diff options
Diffstat (limited to 'fs/afs')
50 files changed, 13510 insertions, 8693 deletions
diff --git a/fs/afs/Kconfig b/fs/afs/Kconfig index 701aaa9b1899..682bd8ec2c10 100644 --- a/fs/afs/Kconfig +++ b/fs/afs/Kconfig @@ -1,13 +1,16 @@ +# SPDX-License-Identifier: GPL-2.0-only config AFS_FS tristate "Andrew File System support (AFS)" depends on INET select AF_RXRPC select DNS_RESOLVER + select NETFS_SUPPORT + select CRYPTO_KRB5 help If you say Y here, you will get an experimental Andrew File System driver. It currently only supports unsecured read-only AFS access. - See <file:Documentation/filesystems/afs.txt> for more information. + See <file:Documentation/filesystems/afs.rst> for more information. If unsure, say N. @@ -17,7 +20,7 @@ config AFS_DEBUG help Say Y here to make runtime controllable debugging messages appear. - See <file:Documentation/filesystems/afs.txt> for more information. + See <file:Documentation/filesystems/afs.rst> for more information. If unsure, say N. @@ -36,6 +39,6 @@ config AFS_DEBUG_CURSOR the dmesg log if the server rotation algorithm fails to successfully contact a server. - See <file:Documentation/filesystems/afs.txt> for more information. + See <file:Documentation/filesystems/afs.rst> for more information. If unsure, say N. diff --git a/fs/afs/Makefile b/fs/afs/Makefile index 0738e2bf5193..b49b8fe682f3 100644 --- a/fs/afs/Makefile +++ b/fs/afs/Makefile @@ -3,20 +3,22 @@ # Makefile for Red Hat Linux AFS client. # -afs-cache-$(CONFIG_AFS_FSCACHE) := cache.o - kafs-y := \ - $(afs-cache-y) \ addr_list.o \ + addr_prefs.o \ callback.o \ cell.o \ + cm_security.o \ cmservice.o \ dir.o \ dir_edit.o \ + dir_search.o \ + dir_silly.o \ dynroot.o \ file.o \ flock.o \ fsclient.o \ + fs_operation.o \ fs_probe.o \ inode.o \ main.o \ @@ -28,8 +30,9 @@ kafs-y := \ server.o \ server_list.o \ super.o \ - netdevices.o \ + validation.o \ vlclient.o \ + vl_alias.o \ vl_list.o \ vl_probe.o \ vl_rotate.o \ diff --git a/fs/afs/addr_list.c b/fs/afs/addr_list.c index 967db336d11a..e941da5b6dd9 100644 --- a/fs/afs/addr_list.c +++ b/fs/afs/addr_list.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Server address list management * * Copyright (C) 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/slab.h> @@ -17,26 +13,55 @@ #include "internal.h" #include "afs_fs.h" +static void afs_free_addrlist(struct rcu_head *rcu) +{ + struct afs_addr_list *alist = container_of(rcu, struct afs_addr_list, rcu); + unsigned int i; + + for (i = 0; i < alist->nr_addrs; i++) + rxrpc_kernel_put_peer(alist->addrs[i].peer); + trace_afs_alist(alist->debug_id, refcount_read(&alist->usage), afs_alist_trace_free); + kfree(alist); +} + /* * Release an address list. */ -void afs_put_addrlist(struct afs_addr_list *alist) +void afs_put_addrlist(struct afs_addr_list *alist, enum afs_alist_trace reason) { - if (alist && refcount_dec_and_test(&alist->usage)) - call_rcu(&alist->rcu, (rcu_callback_t)kfree); + unsigned int debug_id; + bool dead; + int r; + + if (!alist) + return; + debug_id = alist->debug_id; + dead = __refcount_dec_and_test(&alist->usage, &r); + trace_afs_alist(debug_id, r - 1, reason); + if (dead) + call_rcu(&alist->rcu, afs_free_addrlist); +} + +struct afs_addr_list *afs_get_addrlist(struct afs_addr_list *alist, enum afs_alist_trace reason) +{ + int r; + + if (alist) { + __refcount_inc(&alist->usage, &r); + trace_afs_alist(alist->debug_id, r + 1, reason); + } + return alist; } /* * Allocate an address list. */ -struct afs_addr_list *afs_alloc_addrlist(unsigned int nr, - unsigned short service, - unsigned short port) +struct afs_addr_list *afs_alloc_addrlist(unsigned int nr) { struct afs_addr_list *alist; - unsigned int i; + static atomic_t debug_id; - _enter("%u,%u,%u", nr, service, port); + _enter("%u", nr); if (nr > AFS_MAX_ADDRESSES) nr = AFS_MAX_ADDRESSES; @@ -47,17 +72,8 @@ struct afs_addr_list *afs_alloc_addrlist(unsigned int nr, refcount_set(&alist->usage, 1); alist->max_addrs = nr; - - for (i = 0; i < nr; i++) { - struct sockaddr_rxrpc *srx = &alist->addrs[i]; - srx->srx_family = AF_RXRPC; - srx->srx_service = service; - srx->transport_type = SOCK_DGRAM; - srx->transport_len = sizeof(srx->transport.sin6); - srx->transport.sin6.sin6_family = AF_INET6; - srx->transport.sin6.sin6_port = htons(port); - } - + alist->debug_id = atomic_inc_return(&debug_id); + trace_afs_alist(alist->debug_id, 1, afs_alist_trace_alloc); return alist; } @@ -130,7 +146,7 @@ struct afs_vlserver_list *afs_parse_text_addrs(struct afs_net *net, if (!vllist->servers[0].server) goto error_vl; - alist = afs_alloc_addrlist(nr, service, AFS_VL_PORT); + alist = afs_alloc_addrlist(nr); if (!alist) goto error; @@ -201,9 +217,11 @@ struct afs_vlserver_list *afs_parse_text_addrs(struct afs_net *net, } if (family == AF_INET) - afs_merge_fs_addr4(alist, x[0], xport); + ret = afs_merge_fs_addr4(net, alist, x[0], xport); else - afs_merge_fs_addr6(alist, x, xport); + ret = afs_merge_fs_addr6(net, alist, x, xport); + if (ret < 0) + goto error; } while (p < end); @@ -220,26 +238,13 @@ bad_address: problem, p - text, (int)len, (int)len, text); ret = -EINVAL; error: - afs_put_addrlist(alist); + afs_put_addrlist(alist, afs_alist_trace_put_parse_error); error_vl: afs_put_vlserverlist(net, vllist); return ERR_PTR(ret); } /* - * Compare old and new address lists to see if there's been any change. - * - How to do this in better than O(Nlog(N)) time? - * - We don't really want to sort the address list, but would rather take the - * list as we got it so as not to undo record rotation by the DNS server. - */ -#if 0 -static int afs_cmp_addr_list(const struct afs_addr_list *a1, - const struct afs_addr_list *a2) -{ -} -#endif - -/* * Perform a DNS query for VL servers and build a up an address list. */ struct afs_vlserver_list *afs_dns_query(struct afs_cell *cell, time64_t *_expiry) @@ -250,8 +255,8 @@ struct afs_vlserver_list *afs_dns_query(struct afs_cell *cell, time64_t *_expiry _enter("%s", cell->name); - ret = dns_query("afsdb", cell->name, cell->name_len, "srv=1", - &result, _expiry); + ret = dns_query(cell->net->net, "afsdb", cell->name, cell->name_len, + "srv=1", &result, _expiry, true); if (ret < 0) { _leave(" = %d [dns]", ret); return ERR_PTR(ret); @@ -275,25 +280,33 @@ struct afs_vlserver_list *afs_dns_query(struct afs_cell *cell, time64_t *_expiry /* * Merge an IPv4 entry into a fileserver address list. */ -void afs_merge_fs_addr4(struct afs_addr_list *alist, __be32 xdr, u16 port) +int afs_merge_fs_addr4(struct afs_net *net, struct afs_addr_list *alist, + __be32 xdr, u16 port) { - struct sockaddr_rxrpc *srx; - u32 addr = ntohl(xdr); + struct sockaddr_rxrpc srx; + struct rxrpc_peer *peer; int i; if (alist->nr_addrs >= alist->max_addrs) - return; + return 0; - for (i = 0; i < alist->nr_ipv4; i++) { - struct sockaddr_in *a = &alist->addrs[i].transport.sin; - u32 a_addr = ntohl(a->sin_addr.s_addr); - u16 a_port = ntohs(a->sin_port); + srx.srx_family = AF_RXRPC; + srx.transport_type = SOCK_DGRAM; + srx.transport_len = sizeof(srx.transport.sin); + srx.transport.sin.sin_family = AF_INET; + srx.transport.sin.sin_port = htons(port); + srx.transport.sin.sin_addr.s_addr = xdr; - if (addr == a_addr && port == a_port) - return; - if (addr == a_addr && port < a_port) - break; - if (addr < a_addr) + peer = rxrpc_kernel_lookup_peer(net->socket, &srx, GFP_KERNEL); + if (!peer) + return -ENOMEM; + + for (i = 0; i < alist->nr_ipv4; i++) { + if (peer == alist->addrs[i].peer) { + rxrpc_kernel_put_peer(peer); + return 0; + } + if (peer <= alist->addrs[i].peer) break; } @@ -302,38 +315,42 @@ void afs_merge_fs_addr4(struct afs_addr_list *alist, __be32 xdr, u16 port) alist->addrs + i, sizeof(alist->addrs[0]) * (alist->nr_addrs - i)); - srx = &alist->addrs[i]; - srx->srx_family = AF_RXRPC; - srx->transport_type = SOCK_DGRAM; - srx->transport_len = sizeof(srx->transport.sin); - srx->transport.sin.sin_family = AF_INET; - srx->transport.sin.sin_port = htons(port); - srx->transport.sin.sin_addr.s_addr = xdr; + alist->addrs[i].peer = peer; alist->nr_ipv4++; alist->nr_addrs++; + return 0; } /* * Merge an IPv6 entry into a fileserver address list. */ -void afs_merge_fs_addr6(struct afs_addr_list *alist, __be32 *xdr, u16 port) +int afs_merge_fs_addr6(struct afs_net *net, struct afs_addr_list *alist, + __be32 *xdr, u16 port) { - struct sockaddr_rxrpc *srx; - int i, diff; + struct sockaddr_rxrpc srx; + struct rxrpc_peer *peer; + int i; if (alist->nr_addrs >= alist->max_addrs) - return; + return 0; - for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) { - struct sockaddr_in6 *a = &alist->addrs[i].transport.sin6; - u16 a_port = ntohs(a->sin6_port); + srx.srx_family = AF_RXRPC; + srx.transport_type = SOCK_DGRAM; + srx.transport_len = sizeof(srx.transport.sin6); + srx.transport.sin6.sin6_family = AF_INET6; + srx.transport.sin6.sin6_port = htons(port); + memcpy(&srx.transport.sin6.sin6_addr, xdr, 16); - diff = memcmp(xdr, &a->sin6_addr, 16); - if (diff == 0 && port == a_port) - return; - if (diff == 0 && port < a_port) - break; - if (diff < 0) + peer = rxrpc_kernel_lookup_peer(net->socket, &srx, GFP_KERNEL); + if (!peer) + return -ENOMEM; + + for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) { + if (peer == alist->addrs[i].peer) { + rxrpc_kernel_put_peer(peer); + return 0; + } + if (peer <= alist->addrs[i].peer) break; } @@ -341,68 +358,57 @@ void afs_merge_fs_addr6(struct afs_addr_list *alist, __be32 *xdr, u16 port) memmove(alist->addrs + i + 1, alist->addrs + i, sizeof(alist->addrs[0]) * (alist->nr_addrs - i)); - - srx = &alist->addrs[i]; - srx->srx_family = AF_RXRPC; - srx->transport_type = SOCK_DGRAM; - srx->transport_len = sizeof(srx->transport.sin6); - srx->transport.sin6.sin6_family = AF_INET6; - srx->transport.sin6.sin6_port = htons(port); - memcpy(&srx->transport.sin6.sin6_addr, xdr, 16); + alist->addrs[i].peer = peer; alist->nr_addrs++; + return 0; } /* - * Get an address to try. + * Set the app data on the rxrpc peers an address list points to */ -bool afs_iterate_addresses(struct afs_addr_cursor *ac) +void afs_set_peer_appdata(struct afs_server *server, + struct afs_addr_list *old_alist, + struct afs_addr_list *new_alist) { - unsigned long set, failed; - int index; - - if (!ac->alist) - return false; - - set = ac->alist->responded; - failed = ac->alist->failed; - _enter("%lx-%lx-%lx,%d", set, failed, ac->tried, ac->index); - - ac->nr_iterations++; - - set &= ~(failed | ac->tried); - - if (!set) - return false; - - index = READ_ONCE(ac->alist->preferred); - if (test_bit(index, &set)) - goto selected; + unsigned long data = (unsigned long)server; + int n = 0, o = 0; - index = __ffs(set); - -selected: - ac->index = index; - set_bit(index, &ac->tried); - ac->responded = false; - return true; -} + if (!old_alist) { + /* New server. Just set all. */ + for (; n < new_alist->nr_addrs; n++) + rxrpc_kernel_set_peer_data(new_alist->addrs[n].peer, data); + return; + } + if (!new_alist) { + /* Dead server. Just remove all. */ + for (; o < old_alist->nr_addrs; o++) + rxrpc_kernel_set_peer_data(old_alist->addrs[o].peer, 0); + return; + } -/* - * Release an address list cursor. - */ -int afs_end_cursor(struct afs_addr_cursor *ac) -{ - struct afs_addr_list *alist; + /* Walk through the two lists simultaneously, setting new peers and + * clearing old ones. The two lists are ordered by pointer to peer + * record. + */ + while (n < new_alist->nr_addrs && o < old_alist->nr_addrs) { + struct rxrpc_peer *pn = new_alist->addrs[n].peer; + struct rxrpc_peer *po = old_alist->addrs[o].peer; - alist = ac->alist; - if (alist) { - if (ac->responded && - ac->index != alist->preferred && - test_bit(ac->alist->preferred, &ac->tried)) - WRITE_ONCE(alist->preferred, ac->index); - afs_put_addrlist(alist); - ac->alist = NULL; + if (pn == po) + continue; + if (pn < po) { + rxrpc_kernel_set_peer_data(pn, data); + n++; + } else { + rxrpc_kernel_set_peer_data(po, 0); + o++; + } } - return ac->error; + if (n < new_alist->nr_addrs) + for (; n < new_alist->nr_addrs; n++) + rxrpc_kernel_set_peer_data(new_alist->addrs[n].peer, data); + if (o < old_alist->nr_addrs) + for (; o < old_alist->nr_addrs; o++) + rxrpc_kernel_set_peer_data(old_alist->addrs[o].peer, 0); } diff --git a/fs/afs/addr_prefs.c b/fs/afs/addr_prefs.c new file mode 100644 index 000000000000..133736412c3d --- /dev/null +++ b/fs/afs/addr_prefs.c @@ -0,0 +1,533 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Address preferences management + * + * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": addr_prefs: " fmt +#include <linux/slab.h> +#include <linux/ctype.h> +#include <linux/inet.h> +#include <linux/seq_file.h> +#include <keys/rxrpc-type.h> +#include "internal.h" + +static inline struct afs_net *afs_seq2net_single(struct seq_file *m) +{ + return afs_net(seq_file_single_net(m)); +} + +/* + * Split a NUL-terminated string up to the first newline around spaces. The + * source string will be modified to have NUL-terminations inserted. + */ +static int afs_split_string(char **pbuf, char *strv[], unsigned int maxstrv) +{ + unsigned int count = 0; + char *p = *pbuf; + + maxstrv--; /* Allow for terminal NULL */ + for (;;) { + /* Skip over spaces */ + while (isspace(*p)) { + if (*p == '\n') { + p++; + break; + } + p++; + } + if (!*p) + break; + + /* Mark start of word */ + if (count >= maxstrv) { + pr_warn("Too many elements in string\n"); + return -EINVAL; + } + strv[count++] = p; + + /* Skip over word */ + while (!isspace(*p) && *p) + p++; + if (!*p) + break; + + /* Mark end of word */ + if (*p == '\n') { + *p++ = 0; + break; + } + *p++ = 0; + } + + *pbuf = p; + strv[count] = NULL; + return count; +} + +/* + * Parse an address with an optional subnet mask. + */ +static int afs_parse_address(char *p, struct afs_addr_preference *pref) +{ + const char *stop; + unsigned long mask, tmp; + char *end = p + strlen(p); + bool bracket = false; + + if (*p == '[') { + p++; + bracket = true; + } + +#if 0 + if (*p == '[') { + p++; + q = memchr(p, ']', end - p); + if (!q) { + pr_warn("Can't find closing ']'\n"); + return -EINVAL; + } + } else { + for (q = p; q < end; q++) + if (*q == '/') + break; + } +#endif + + if (in4_pton(p, end - p, (u8 *)&pref->ipv4_addr, -1, &stop)) { + pref->family = AF_INET; + mask = 32; + } else if (in6_pton(p, end - p, (u8 *)&pref->ipv6_addr, -1, &stop)) { + pref->family = AF_INET6; + mask = 128; + } else { + pr_warn("Can't determine address family\n"); + return -EINVAL; + } + + p = (char *)stop; + if (bracket) { + if (*p != ']') { + pr_warn("Can't find closing ']'\n"); + return -EINVAL; + } + p++; + } + + if (*p == '/') { + p++; + tmp = simple_strtoul(p, &p, 10); + if (tmp > mask) { + pr_warn("Subnet mask too large\n"); + return -EINVAL; + } + if (tmp == 0) { + pr_warn("Subnet mask too small\n"); + return -EINVAL; + } + mask = tmp; + } + + if (*p) { + pr_warn("Invalid address\n"); + return -EINVAL; + } + + pref->subnet_mask = mask; + return 0; +} + +enum cmp_ret { + CONTINUE_SEARCH, + INSERT_HERE, + EXACT_MATCH, + SUBNET_MATCH, +}; + +/* + * See if a candidate address matches a listed address. + */ +static enum cmp_ret afs_cmp_address_pref(const struct afs_addr_preference *a, + const struct afs_addr_preference *b) +{ + int subnet = min(a->subnet_mask, b->subnet_mask); + const __be32 *pa, *pb; + u32 mask, na, nb; + int diff; + + if (a->family != b->family) + return INSERT_HERE; + + switch (a->family) { + case AF_INET6: + pa = a->ipv6_addr.s6_addr32; + pb = b->ipv6_addr.s6_addr32; + break; + case AF_INET: + pa = &a->ipv4_addr.s_addr; + pb = &b->ipv4_addr.s_addr; + break; + } + + while (subnet > 32) { + diff = ntohl(*pa++) - ntohl(*pb++); + if (diff < 0) + return INSERT_HERE; /* a<b */ + if (diff > 0) + return CONTINUE_SEARCH; /* a>b */ + subnet -= 32; + } + + if (subnet == 0) + return EXACT_MATCH; + + mask = 0xffffffffU << (32 - subnet); + na = ntohl(*pa); + nb = ntohl(*pb); + diff = (na & mask) - (nb & mask); + //kdebug("diff %08x %08x %08x %d", na, nb, mask, diff); + if (diff < 0) + return INSERT_HERE; /* a<b */ + if (diff > 0) + return CONTINUE_SEARCH; /* a>b */ + if (a->subnet_mask == b->subnet_mask) + return EXACT_MATCH; + if (a->subnet_mask > b->subnet_mask) + return SUBNET_MATCH; /* a binds tighter than b */ + return CONTINUE_SEARCH; /* b binds tighter than a */ +} + +/* + * Insert an address preference. + */ +static int afs_insert_address_pref(struct afs_addr_preference_list **_preflist, + struct afs_addr_preference *pref, + int index) +{ + struct afs_addr_preference_list *preflist = *_preflist, *old = preflist; + size_t size, max_prefs; + + _enter("{%u/%u/%u},%u", preflist->ipv6_off, preflist->nr, preflist->max_prefs, index); + + if (preflist->nr == 255) + return -ENOSPC; + if (preflist->nr >= preflist->max_prefs) { + max_prefs = preflist->max_prefs + 1; + size = struct_size(preflist, prefs, max_prefs); + size = roundup_pow_of_two(size); + max_prefs = min_t(size_t, (size - sizeof(*preflist)) / sizeof(*pref), 255); + preflist = kmalloc(size, GFP_KERNEL); + if (!preflist) + return -ENOMEM; + *preflist = **_preflist; + preflist->max_prefs = max_prefs; + *_preflist = preflist; + + if (index < preflist->nr) + memcpy(preflist->prefs + index + 1, old->prefs + index, + sizeof(*pref) * (preflist->nr - index)); + if (index > 0) + memcpy(preflist->prefs, old->prefs, sizeof(*pref) * index); + } else { + if (index < preflist->nr) + memmove(preflist->prefs + index + 1, preflist->prefs + index, + sizeof(*pref) * (preflist->nr - index)); + } + + preflist->prefs[index] = *pref; + preflist->nr++; + if (pref->family == AF_INET) + preflist->ipv6_off++; + return 0; +} + +/* + * Add an address preference. + * echo "add <proto> <IP>[/<mask>] <prior>" >/proc/fs/afs/addr_prefs + */ +static int afs_add_address_pref(struct afs_net *net, struct afs_addr_preference_list **_preflist, + int argc, char **argv) +{ + struct afs_addr_preference_list *preflist = *_preflist; + struct afs_addr_preference pref; + enum cmp_ret cmp; + int ret, i, stop; + + if (argc != 3) { + pr_warn("Wrong number of params\n"); + return -EINVAL; + } + + if (strcmp(argv[0], "udp") != 0) { + pr_warn("Unsupported protocol\n"); + return -EINVAL; + } + + ret = afs_parse_address(argv[1], &pref); + if (ret < 0) + return ret; + + ret = kstrtou16(argv[2], 10, &pref.prio); + if (ret < 0) { + pr_warn("Invalid priority\n"); + return ret; + } + + if (pref.family == AF_INET) { + i = 0; + stop = preflist->ipv6_off; + } else { + i = preflist->ipv6_off; + stop = preflist->nr; + } + + for (; i < stop; i++) { + cmp = afs_cmp_address_pref(&pref, &preflist->prefs[i]); + switch (cmp) { + case CONTINUE_SEARCH: + continue; + case INSERT_HERE: + case SUBNET_MATCH: + return afs_insert_address_pref(_preflist, &pref, i); + case EXACT_MATCH: + preflist->prefs[i].prio = pref.prio; + return 0; + } + } + + return afs_insert_address_pref(_preflist, &pref, i); +} + +/* + * Delete an address preference. + */ +static int afs_delete_address_pref(struct afs_addr_preference_list **_preflist, + int index) +{ + struct afs_addr_preference_list *preflist = *_preflist; + + _enter("{%u/%u/%u},%u", preflist->ipv6_off, preflist->nr, preflist->max_prefs, index); + + if (preflist->nr == 0) + return -ENOENT; + + if (index < preflist->nr - 1) + memmove(preflist->prefs + index, preflist->prefs + index + 1, + sizeof(preflist->prefs[0]) * (preflist->nr - index - 1)); + + if (index < preflist->ipv6_off) + preflist->ipv6_off--; + preflist->nr--; + return 0; +} + +/* + * Delete an address preference. + * echo "del <proto> <IP>[/<mask>]" >/proc/fs/afs/addr_prefs + */ +static int afs_del_address_pref(struct afs_net *net, struct afs_addr_preference_list **_preflist, + int argc, char **argv) +{ + struct afs_addr_preference_list *preflist = *_preflist; + struct afs_addr_preference pref; + enum cmp_ret cmp; + int ret, i, stop; + + if (argc != 2) { + pr_warn("Wrong number of params\n"); + return -EINVAL; + } + + if (strcmp(argv[0], "udp") != 0) { + pr_warn("Unsupported protocol\n"); + return -EINVAL; + } + + ret = afs_parse_address(argv[1], &pref); + if (ret < 0) + return ret; + + if (pref.family == AF_INET) { + i = 0; + stop = preflist->ipv6_off; + } else { + i = preflist->ipv6_off; + stop = preflist->nr; + } + + for (; i < stop; i++) { + cmp = afs_cmp_address_pref(&pref, &preflist->prefs[i]); + switch (cmp) { + case CONTINUE_SEARCH: + continue; + case INSERT_HERE: + case SUBNET_MATCH: + return 0; + case EXACT_MATCH: + return afs_delete_address_pref(_preflist, i); + } + } + + return -ENOANO; +} + +/* + * Handle writes to /proc/fs/afs/addr_prefs + */ +int afs_proc_addr_prefs_write(struct file *file, char *buf, size_t size) +{ + struct afs_addr_preference_list *preflist, *old; + struct seq_file *m = file->private_data; + struct afs_net *net = afs_seq2net_single(m); + size_t psize; + char *argv[5]; + int ret, argc, max_prefs; + + inode_lock(file_inode(file)); + + /* Allocate a candidate new list and initialise it from the old. */ + old = rcu_dereference_protected(net->address_prefs, + lockdep_is_held(&file_inode(file)->i_rwsem)); + + if (old) + max_prefs = old->nr + 1; + else + max_prefs = 1; + + psize = struct_size(old, prefs, max_prefs); + psize = roundup_pow_of_two(psize); + max_prefs = min_t(size_t, (psize - sizeof(*old)) / sizeof(old->prefs[0]), 255); + + ret = -ENOMEM; + preflist = kmalloc(struct_size(preflist, prefs, max_prefs), GFP_KERNEL); + if (!preflist) + goto done; + + if (old) + memcpy(preflist, old, struct_size(preflist, prefs, old->nr)); + else + memset(preflist, 0, sizeof(*preflist)); + preflist->max_prefs = max_prefs; + + do { + argc = afs_split_string(&buf, argv, ARRAY_SIZE(argv)); + if (argc < 0) { + ret = argc; + goto done; + } + if (argc < 2) + goto inval; + + if (strcmp(argv[0], "add") == 0) + ret = afs_add_address_pref(net, &preflist, argc - 1, argv + 1); + else if (strcmp(argv[0], "del") == 0) + ret = afs_del_address_pref(net, &preflist, argc - 1, argv + 1); + else + goto inval; + if (ret < 0) + goto done; + } while (*buf); + + preflist->version++; + rcu_assign_pointer(net->address_prefs, preflist); + /* Store prefs before version */ + smp_store_release(&net->address_pref_version, preflist->version); + kfree_rcu(old, rcu); + preflist = NULL; + ret = 0; + +done: + kfree(preflist); + inode_unlock(file_inode(file)); + _leave(" = %d", ret); + return ret; + +inval: + pr_warn("Invalid Command\n"); + ret = -EINVAL; + goto done; +} + +/* + * Mark the priorities on an address list if the address preferences table has + * changed. The caller must hold the RCU read lock. + */ +void afs_get_address_preferences_rcu(struct afs_net *net, struct afs_addr_list *alist) +{ + const struct afs_addr_preference_list *preflist = + rcu_dereference(net->address_prefs); + const struct sockaddr_in6 *sin6; + const struct sockaddr_in *sin; + const struct sockaddr *sa; + struct afs_addr_preference test; + enum cmp_ret cmp; + int i, j; + + if (!preflist || !preflist->nr || !alist->nr_addrs || + smp_load_acquire(&alist->addr_pref_version) == preflist->version) + return; + + test.family = AF_INET; + test.subnet_mask = 32; + test.prio = 0; + for (i = 0; i < alist->nr_ipv4; i++) { + sa = rxrpc_kernel_remote_addr(alist->addrs[i].peer); + sin = (const struct sockaddr_in *)sa; + test.ipv4_addr = sin->sin_addr; + for (j = 0; j < preflist->ipv6_off; j++) { + cmp = afs_cmp_address_pref(&test, &preflist->prefs[j]); + switch (cmp) { + case CONTINUE_SEARCH: + continue; + case INSERT_HERE: + break; + case EXACT_MATCH: + case SUBNET_MATCH: + WRITE_ONCE(alist->addrs[i].prio, preflist->prefs[j].prio); + break; + } + } + } + + test.family = AF_INET6; + test.subnet_mask = 128; + test.prio = 0; + for (; i < alist->nr_addrs; i++) { + sa = rxrpc_kernel_remote_addr(alist->addrs[i].peer); + sin6 = (const struct sockaddr_in6 *)sa; + test.ipv6_addr = sin6->sin6_addr; + for (j = preflist->ipv6_off; j < preflist->nr; j++) { + cmp = afs_cmp_address_pref(&test, &preflist->prefs[j]); + switch (cmp) { + case CONTINUE_SEARCH: + continue; + case INSERT_HERE: + break; + case EXACT_MATCH: + case SUBNET_MATCH: + WRITE_ONCE(alist->addrs[i].prio, preflist->prefs[j].prio); + break; + } + } + } + + smp_store_release(&alist->addr_pref_version, preflist->version); +} + +/* + * Mark the priorities on an address list if the address preferences table has + * changed. Avoid taking the RCU read lock if we can. + */ +void afs_get_address_preferences(struct afs_net *net, struct afs_addr_list *alist) +{ + if (!net->address_prefs || + /* Load version before prefs */ + smp_load_acquire(&net->address_pref_version) == alist->addr_pref_version) + return; + + rcu_read_lock(); + afs_get_address_preferences_rcu(net, alist); + rcu_read_unlock(); +} diff --git a/fs/afs/afs.h b/fs/afs/afs.h index d12ffb457e47..ec3db00bd081 100644 --- a/fs/afs/afs.h +++ b/fs/afs/afs.h @@ -1,12 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* AFS common types * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #ifndef AFS_H @@ -14,7 +10,7 @@ #include <linux/in.h> -#define AFS_MAXCELLNAME 64 /* Maximum length of a cell name */ +#define AFS_MAXCELLNAME 253 /* Maximum length of a cell name (DNS limited) */ #define AFS_MAXVOLNAME 64 /* Maximum length of a volume name */ #define AFS_MAXNSERVERS 8 /* Maximum servers in a basic volume record */ #define AFS_NMAXNSERVERS 13 /* Maximum servers in a N/U-class volume record */ @@ -23,6 +19,9 @@ #define AFSPATHMAX 1024 /* Maximum length of a pathname plus NUL */ #define AFSOPAQUEMAX 1024 /* Maximum length of an opaque field */ +#define AFS_VL_MAX_LIFESPAN 120 +#define AFS_PROBE_MAX_LIFESPAN 30 + typedef u64 afs_volid_t; typedef u64 afs_vnodeid_t; typedef u64 afs_dataversion_t; @@ -69,8 +68,8 @@ typedef enum { struct afs_callback { time64_t expires_at; /* Time at which expires */ - unsigned version; /* Callback version */ - afs_callback_type_t type; /* Type of callback */ + //unsigned version; /* Callback version */ + //afs_callback_type_t type; /* Type of callback */ }; struct afs_callback_break { @@ -144,6 +143,14 @@ struct afs_file_status { u32 abort_code; /* Abort if bulk-fetching this failed */ }; +struct afs_status_cb { + struct afs_file_status status; + struct afs_callback callback; + bool have_status; /* True if status record was retrieved */ + bool have_cb; /* True if cb record was retrieved */ + bool have_error; /* True if status.abort_code indicates an error */ +}; + /* * AFS file status change request */ @@ -158,7 +165,8 @@ struct afs_file_status { * AFS volume synchronisation information */ struct afs_volsync { - time64_t creation; /* volume creation time */ + time64_t creation; /* Volume creation time (or TIME64_MIN) */ + time64_t update; /* Volume update time (or TIME64_MIN) */ }; /* diff --git a/fs/afs/afs_cm.h b/fs/afs/afs_cm.h index 255f5dd6040c..565cbe0a8af6 100644 --- a/fs/afs/afs_cm.h +++ b/fs/afs/afs_cm.h @@ -1,12 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* AFS Cache Manager definitions * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #ifndef AFS_CM_H diff --git a/fs/afs/afs_fs.h b/fs/afs/afs_fs.h index ddfa88a7a9c0..20ab344baf9d 100644 --- a/fs/afs/afs_fs.h +++ b/fs/afs/afs_fs.h @@ -1,12 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* AFS File Service definitions * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #ifndef AFS_FS_H @@ -17,8 +13,10 @@ enum AFS_FS_Operations { FSFETCHDATA = 130, /* AFS Fetch file data */ + FSFETCHACL = 131, /* AFS Fetch file ACL */ FSFETCHSTATUS = 132, /* AFS Fetch file status */ FSSTOREDATA = 133, /* AFS Store file data */ + FSSTOREACL = 134, /* AFS Store file ACL */ FSSTORESTATUS = 135, /* AFS Store file status */ FSREMOVEFILE = 136, /* AFS Remove a file */ FSCREATEFILE = 137, /* AFS Create a file */ diff --git a/fs/afs/afs_vl.h b/fs/afs/afs_vl.h index e3c4688f573b..b835e25a2c02 100644 --- a/fs/afs/afs_vl.h +++ b/fs/afs/afs_vl.h @@ -1,12 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* AFS Volume Location Service client interface * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #ifndef AFS_VL_H @@ -17,6 +13,7 @@ #define AFS_VL_PORT 7003 /* volume location service port */ #define VL_SERVICE 52 /* RxRPC service ID for the Volume Location service */ #define YFS_VL_SERVICE 2503 /* Service ID for AuriStor upgraded VL service */ +#define YFS_VL_MAXCELLNAME 256 /* Maximum length of a cell name in YFS protocol */ enum AFSVL_Operations { VLGETENTRYBYID = 503, /* AFS Get VLDB entry by ID */ @@ -26,6 +23,7 @@ enum AFSVL_Operations { VLGETENTRYBYNAMEU = 527, /* AFS Get VLDB entry by name (UUID-variant) */ VLGETADDRSU = 533, /* AFS Get addrs for fileserver */ YVLGETENDPOINTS = 64002, /* YFS Get endpoints for file/volume server */ + YVLGETCELLNAME = 64014, /* YFS Get actual cell name */ VLGETCAPABILITIES = 65537, /* AFS Get server capabilities */ }; @@ -137,13 +135,4 @@ struct afs_uvldbentry__xdr { __be32 spares9; }; -struct afs_address_list { - refcount_t usage; - unsigned int version; - unsigned int nr_addrs; - struct sockaddr_rxrpc addrs[]; -}; - -extern void afs_put_address_list(struct afs_address_list *alist); - #endif /* AFS_VL_H */ diff --git a/fs/afs/cache.c b/fs/afs/cache.c deleted file mode 100644 index f6d0a21e8052..000000000000 --- a/fs/afs/cache.c +++ /dev/null @@ -1,72 +0,0 @@ -/* AFS caching stuff - * - * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/sched.h> -#include "internal.h" - -static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, - const void *buffer, - uint16_t buflen, - loff_t object_size); - -struct fscache_netfs afs_cache_netfs = { - .name = "afs", - .version = 2, -}; - -struct fscache_cookie_def afs_cell_cache_index_def = { - .name = "AFS.cell", - .type = FSCACHE_COOKIE_TYPE_INDEX, -}; - -struct fscache_cookie_def afs_volume_cache_index_def = { - .name = "AFS.volume", - .type = FSCACHE_COOKIE_TYPE_INDEX, -}; - -struct fscache_cookie_def afs_vnode_cache_index_def = { - .name = "AFS.vnode", - .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .check_aux = afs_vnode_cache_check_aux, -}; - -/* - * check that the auxiliary data indicates that the entry is still valid - */ -static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, - const void *buffer, - uint16_t buflen, - loff_t object_size) -{ - struct afs_vnode *vnode = cookie_netfs_data; - struct afs_vnode_cache_aux aux; - - _enter("{%llx,%x,%llx},%p,%u", - vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, - buffer, buflen); - - memcpy(&aux, buffer, sizeof(aux)); - - /* check the size of the data is what we're expecting */ - if (buflen != sizeof(aux)) { - _leave(" = OBSOLETE [len %hx != %zx]", buflen, sizeof(aux)); - return FSCACHE_CHECKAUX_OBSOLETE; - } - - if (vnode->status.data_version != aux.data_version) { - _leave(" = OBSOLETE [vers %llx != %llx]", - aux.data_version, vnode->status.data_version); - return FSCACHE_CHECKAUX_OBSOLETE; - } - - _leave(" = SUCCESS"); - return FSCACHE_CHECKAUX_OKAY; -} diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 1c7955f5cdaf..894d2bad6b6c 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -21,220 +21,162 @@ #include "internal.h" /* - * Create volume and callback interests on a server. + * Handle invalidation of an mmap'd file. We invalidate all the PTEs referring + * to the pages in this file's pagecache, forcing the kernel to go through + * ->fault() or ->page_mkwrite() - at which point we can handle invalidation + * more fully. */ -static struct afs_cb_interest *afs_create_interest(struct afs_server *server, - struct afs_vnode *vnode) +void afs_invalidate_mmap_work(struct work_struct *work) { - struct afs_vol_interest *new_vi, *vi; - struct afs_cb_interest *new; - struct hlist_node **pp; - - new_vi = kzalloc(sizeof(struct afs_vol_interest), GFP_KERNEL); - if (!new_vi) - return NULL; - - new = kzalloc(sizeof(struct afs_cb_interest), GFP_KERNEL); - if (!new) { - kfree(new_vi); - return NULL; - } + struct afs_vnode *vnode = container_of(work, struct afs_vnode, cb_work); - new_vi->usage = 1; - new_vi->vid = vnode->volume->vid; - INIT_HLIST_NODE(&new_vi->srv_link); - INIT_HLIST_HEAD(&new_vi->cb_interests); + unmap_mapping_pages(vnode->netfs.inode.i_mapping, 0, 0, false); +} - refcount_set(&new->usage, 1); - new->sb = vnode->vfs_inode.i_sb; - new->vid = vnode->volume->vid; - new->server = afs_get_server(server); - INIT_HLIST_NODE(&new->cb_vlink); +static void afs_volume_init_callback(struct afs_volume *volume) +{ + struct afs_vnode *vnode; - write_lock(&server->cb_break_lock); + down_read(&volume->open_mmaps_lock); - for (pp = &server->cb_volumes.first; *pp; pp = &(*pp)->next) { - vi = hlist_entry(*pp, struct afs_vol_interest, srv_link); - if (vi->vid < new_vi->vid) - continue; - if (vi->vid > new_vi->vid) - break; - vi->usage++; - goto found_vi; + list_for_each_entry(vnode, &volume->open_mmaps, cb_mmap_link) { + if (vnode->cb_v_check != atomic_read(&volume->cb_v_break)) { + afs_clear_cb_promise(vnode, afs_cb_promise_clear_vol_init_cb); + queue_work(system_dfl_wq, &vnode->cb_work); + } } - new_vi->srv_link.pprev = pp; - new_vi->srv_link.next = *pp; - if (*pp) - (*pp)->pprev = &new_vi->srv_link.next; - *pp = &new_vi->srv_link; - vi = new_vi; - new_vi = NULL; -found_vi: - - new->vol_interest = vi; - hlist_add_head(&new->cb_vlink, &vi->cb_interests); - - write_unlock(&server->cb_break_lock); - kfree(new_vi); - return new; + up_read(&volume->open_mmaps_lock); } /* - * Set up an interest-in-callbacks record for a volume on a server and - * register it with the server. - * - Called with vnode->io_lock held. + * Allow the fileserver to request callback state (re-)initialisation. + * Unfortunately, UUIDs are not guaranteed unique. */ -int afs_register_server_cb_interest(struct afs_vnode *vnode, - struct afs_server_list *slist, - unsigned int index) +void afs_init_callback_state(struct afs_server *server) { - struct afs_server_entry *entry = &slist->servers[index]; - struct afs_cb_interest *cbi, *vcbi, *new, *old; - struct afs_server *server = entry->server; - -again: - if (vnode->cb_interest && - likely(vnode->cb_interest == entry->cb_interest)) - return 0; - - read_lock(&slist->lock); - cbi = afs_get_cb_interest(entry->cb_interest); - read_unlock(&slist->lock); - - vcbi = vnode->cb_interest; - if (vcbi) { - if (vcbi == cbi) { - afs_put_cb_interest(afs_v2net(vnode), cbi); - return 0; - } + struct afs_server_entry *se; - /* Use a new interest in the server list for the same server - * rather than an old one that's still attached to a vnode. - */ - if (cbi && vcbi->server == cbi->server) { - write_seqlock(&vnode->cb_lock); - old = vnode->cb_interest; - vnode->cb_interest = cbi; - write_sequnlock(&vnode->cb_lock); - afs_put_cb_interest(afs_v2net(vnode), old); - return 0; - } + down_read(&server->cell->vs_lock); - /* Re-use the one attached to the vnode. */ - if (!cbi && vcbi->server == server) { - write_lock(&slist->lock); - if (entry->cb_interest) { - write_unlock(&slist->lock); - afs_put_cb_interest(afs_v2net(vnode), cbi); - goto again; - } - - entry->cb_interest = cbi; - write_unlock(&slist->lock); - return 0; - } + list_for_each_entry(se, &server->volumes, slink) { + se->cb_expires_at = AFS_NO_CB_PROMISE; + se->volume->cb_expires_at = AFS_NO_CB_PROMISE; + trace_afs_cb_v_break(se->volume->vid, atomic_read(&se->volume->cb_v_break), + afs_cb_break_for_s_reinit); + if (!list_empty(&se->volume->open_mmaps)) + afs_volume_init_callback(se->volume); } - if (!cbi) { - new = afs_create_interest(server, vnode); - if (!new) - return -ENOMEM; - - write_lock(&slist->lock); - if (!entry->cb_interest) { - entry->cb_interest = afs_get_cb_interest(new); - cbi = new; - new = NULL; - } else { - cbi = afs_get_cb_interest(entry->cb_interest); - } - write_unlock(&slist->lock); - afs_put_cb_interest(afs_v2net(vnode), new); - } + up_read(&server->cell->vs_lock); +} - ASSERT(cbi); +/* + * actually break a callback + */ +void __afs_break_callback(struct afs_vnode *vnode, enum afs_cb_break_reason reason) +{ + _enter(""); - /* Change the server the vnode is using. This entails scrubbing any - * interest the vnode had in the previous server it was using. - */ - write_seqlock(&vnode->cb_lock); + clear_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); + if (afs_clear_cb_promise(vnode, afs_cb_promise_clear_cb_break)) { + vnode->cb_break++; + vnode->cb_v_check = atomic_read(&vnode->volume->cb_v_break); + afs_clear_permits(vnode); + + if (vnode->lock_state == AFS_VNODE_LOCK_WAITING_FOR_CB) + afs_lock_may_be_available(vnode); - old = vnode->cb_interest; - vnode->cb_interest = cbi; - vnode->cb_s_break = cbi->server->cb_s_break; - vnode->cb_v_break = vnode->volume->cb_v_break; - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); + if (reason != afs_cb_break_for_deleted && + vnode->status.type == AFS_FTYPE_FILE && + atomic_read(&vnode->cb_nr_mmap)) + queue_work(system_dfl_wq, &vnode->cb_work); + trace_afs_cb_break(&vnode->fid, vnode->cb_break, reason, true); + } else { + trace_afs_cb_break(&vnode->fid, vnode->cb_break, reason, false); + } +} + +void afs_break_callback(struct afs_vnode *vnode, enum afs_cb_break_reason reason) +{ + write_seqlock(&vnode->cb_lock); + __afs_break_callback(vnode, reason); write_sequnlock(&vnode->cb_lock); - afs_put_cb_interest(afs_v2net(vnode), old); - return 0; } /* - * Remove an interest on a server. + * Look up a volume by volume ID under RCU conditions. */ -void afs_put_cb_interest(struct afs_net *net, struct afs_cb_interest *cbi) +static struct afs_volume *afs_lookup_volume_rcu(struct afs_cell *cell, + afs_volid_t vid) { - struct afs_vol_interest *vi; + struct afs_volume *volume = NULL; + struct rb_node *p; + int seq = 1; + + for (;;) { + /* Unfortunately, rbtree walking doesn't give reliable results + * under just the RCU read lock, so we have to check for + * changes. + */ + seq++; /* 2 on the 1st/lockless path, otherwise odd */ + read_seqbegin_or_lock(&cell->volume_lock, &seq); - if (cbi && refcount_dec_and_test(&cbi->usage)) { - if (!hlist_unhashed(&cbi->cb_vlink)) { - write_lock(&cbi->server->cb_break_lock); + p = rcu_dereference_raw(cell->volumes.rb_node); + while (p) { + volume = rb_entry(p, struct afs_volume, cell_node); - hlist_del_init(&cbi->cb_vlink); - vi = cbi->vol_interest; - cbi->vol_interest = NULL; - if (--vi->usage == 0) - hlist_del(&vi->srv_link); + if (volume->vid < vid) + p = rcu_dereference_raw(p->rb_left); + else if (volume->vid > vid) + p = rcu_dereference_raw(p->rb_right); else - vi = NULL; - - write_unlock(&cbi->server->cb_break_lock); - kfree(vi); - afs_put_server(net, cbi->server); + break; + volume = NULL; } - kfree(cbi); + + if (volume && afs_try_get_volume(volume, afs_volume_trace_get_callback)) + break; + if (!need_seqretry(&cell->volume_lock, seq)) + break; + seq |= 1; /* Want a lock next time */ } -} -/* - * allow the fileserver to request callback state (re-)initialisation - */ -void afs_init_callback_state(struct afs_server *server) -{ - if (!test_and_clear_bit(AFS_SERVER_FL_NEW, &server->flags)) - server->cb_s_break++; + done_seqretry(&cell->volume_lock, seq); + return volume; } /* - * actually break a callback + * Allow the fileserver to break callbacks at the volume-level. This is + * typically done when, for example, a R/W volume is snapshotted to a R/O + * volume (the only way to change an R/O volume). It may also, however, happen + * when a volserver takes control of a volume (offlining it, moving it, etc.). + * + * Every file in that volume will need to be reevaluated. */ -void __afs_break_callback(struct afs_vnode *vnode) +static void afs_break_volume_callback(struct afs_server *server, + struct afs_volume *volume) + __releases(RCU) { - _enter(""); + struct afs_server_list *slist = rcu_dereference(volume->servers); + unsigned int i, cb_v_break; - clear_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); - if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { - vnode->cb_break++; - afs_clear_permits(vnode); + write_lock(&volume->cb_v_break_lock); - spin_lock(&vnode->lock); + for (i = 0; i < slist->nr_servers; i++) + if (slist->servers[i].server == server) + slist->servers[i].cb_expires_at = AFS_NO_CB_PROMISE; + volume->cb_expires_at = AFS_NO_CB_PROMISE; - _debug("break callback"); + cb_v_break = atomic_inc_return_release(&volume->cb_v_break); + trace_afs_cb_v_break(volume->vid, cb_v_break, afs_cb_break_for_volume_callback); - if (list_empty(&vnode->granted_locks) && - !list_empty(&vnode->pending_locks)) - afs_lock_may_be_available(vnode); - spin_unlock(&vnode->lock); - } -} + write_unlock(&volume->cb_v_break_lock); + rcu_read_unlock(); -void afs_break_callback(struct afs_vnode *vnode) -{ - write_seqlock(&vnode->cb_lock); - __afs_break_callback(vnode); - write_sequnlock(&vnode->cb_lock); + if (!list_empty(&volume->open_mmaps)) + afs_volume_init_callback(volume); } /* @@ -244,59 +186,68 @@ void afs_break_callback(struct afs_vnode *vnode) * - a lock is released */ static void afs_break_one_callback(struct afs_server *server, + struct afs_volume *volume, struct afs_fid *fid) { - struct afs_vol_interest *vi; - struct afs_cb_interest *cbi; - struct afs_iget_data data; + struct super_block *sb; struct afs_vnode *vnode; struct inode *inode; - read_lock(&server->cb_break_lock); - hlist_for_each_entry(vi, &server->cb_volumes, srv_link) { - if (vi->vid < fid->vid) - continue; - if (vi->vid > fid->vid) { - vi = NULL; - break; - } - //atomic_inc(&vi->usage); - break; + /* See if we can find a matching inode - even an I_NEW inode needs to + * be marked as it can have its callback broken before we finish + * setting up the local inode. + */ + sb = rcu_dereference(volume->sb); + if (!sb) + return; + + inode = find_inode_rcu(sb, fid->vnode, afs_ilookup5_test_by_fid, fid); + if (inode) { + vnode = AFS_FS_I(inode); + afs_break_callback(vnode, afs_cb_break_for_callback); + } else { + trace_afs_cb_miss(fid, afs_cb_break_for_callback); } +} - /* TODO: Find all matching volumes if we couldn't match the server and - * break them anyway. - */ - if (!vi) - goto out; +static void afs_break_some_callbacks(struct afs_server *server, + struct afs_callback_break *cbb, + size_t *_count) +{ + struct afs_callback_break *residue = cbb; + struct afs_volume *volume; + afs_volid_t vid = cbb->fid.vid; + size_t i; + + rcu_read_lock(); + volume = afs_lookup_volume_rcu(server->cell, vid); + if (cbb->fid.vnode == 0 && cbb->fid.unique == 0) { + afs_break_volume_callback(server, volume); + *_count -= 1; + if (*_count) + memmove(cbb, cbb + 1, sizeof(*cbb) * *_count); + } else { + /* TODO: Find all matching volumes if we couldn't match the server and + * break them anyway. + */ - /* Step through all interested superblocks. There may be more than one - * because of cell aliasing. - */ - hlist_for_each_entry(cbi, &vi->cb_interests, cb_vlink) { - if (fid->vnode == 0 && fid->unique == 0) { - /* The callback break applies to an entire volume. */ - struct afs_super_info *as = AFS_FS_S(cbi->sb); - struct afs_volume *volume = as->volume; - - write_lock(&volume->cb_break_lock); - volume->cb_v_break++; - write_unlock(&volume->cb_break_lock); - } else { - data.volume = NULL; - data.fid = *fid; - inode = ilookup5_nowait(cbi->sb, fid->vnode, - afs_iget5_test, &data); - if (inode) { - vnode = AFS_FS_I(inode); - afs_break_callback(vnode); - iput(inode); + for (i = *_count; i > 0; cbb++, i--) { + if (cbb->fid.vid == vid) { + _debug("- Fid { vl=%08llx n=%llu u=%u }", + cbb->fid.vid, + cbb->fid.vnode, + cbb->fid.unique); + --*_count; + if (volume) + afs_break_one_callback(server, volume, &cbb->fid); + } else { + *residue++ = *cbb; } } + rcu_read_unlock(); } -out: - read_unlock(&server->cb_break_lock); + afs_put_volume(volume, afs_volume_trace_put_callback); } /* @@ -308,31 +259,7 @@ void afs_break_callbacks(struct afs_server *server, size_t count, _enter("%p,%zu,", server, count); ASSERT(server != NULL); - ASSERTCMP(count, <=, AFSCBMAX); - - /* TODO: Sort the callback break list by volume ID */ - for (; count > 0; callbacks++, count--) { - _debug("- Fid { vl=%08llx n=%llu u=%u }", - callbacks->fid.vid, - callbacks->fid.vnode, - callbacks->fid.unique); - afs_break_one_callback(server, &callbacks->fid); - } - - _leave(""); - return; -} - -/* - * Clear the callback interests in a server list. - */ -void afs_clear_callback_interests(struct afs_net *net, struct afs_server_list *slist) -{ - int i; - - for (i = 0; i < slist->nr_servers; i++) { - afs_put_cb_interest(net, slist->servers[i].cb_interest); - slist->servers[i].cb_interest = NULL; - } + while (count > 0) + afs_break_some_callbacks(server, callbacks, &count); } diff --git a/fs/afs/cell.c b/fs/afs/cell.c index cf445dbd5f2e..71c10a05cebe 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS cell and server record management * * Copyright (C) 2002, 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/slab.h> @@ -22,8 +18,11 @@ static unsigned __read_mostly afs_cell_gc_delay = 10; static unsigned __read_mostly afs_cell_min_ttl = 10 * 60; static unsigned __read_mostly afs_cell_max_ttl = 24 * 60 * 60; +static atomic_t cell_debug_id; -static void afs_manage_cell(struct work_struct *); +static void afs_cell_timer(struct timer_list *timer); +static void afs_destroy_cell_work(struct work_struct *work); +static void afs_manage_cell_work(struct work_struct *work); static void afs_dec_cells_outstanding(struct afs_net *net) { @@ -31,29 +30,24 @@ static void afs_dec_cells_outstanding(struct afs_net *net) wake_up_var(&net->cells_outstanding); } -/* - * Set the cell timer to fire after a given delay, assuming it's not already - * set for an earlier time. - */ -static void afs_set_cell_timer(struct afs_net *net, time64_t delay) +static void afs_set_cell_state(struct afs_cell *cell, enum afs_cell_state state) { - if (net->live) { - atomic_inc(&net->cells_outstanding); - if (timer_reduce(&net->cells_timer, jiffies + delay * HZ)) - afs_dec_cells_outstanding(net); - } + smp_store_release(&cell->state, state); /* Commit cell changes before state */ + smp_wmb(); /* Set cell state before task state */ + wake_up_var(&cell->state); } /* - * Look up and get an activation reference on a cell record under RCU - * conditions. The caller must hold the RCU read lock. + * Look up and get an activation reference on a cell record. The caller must + * hold net->cells_lock at least read-locked. */ -struct afs_cell *afs_lookup_cell_rcu(struct afs_net *net, - const char *name, unsigned int namesz) +static struct afs_cell *afs_find_cell_locked(struct afs_net *net, + const char *name, unsigned int namesz, + enum afs_cell_trace reason) { struct afs_cell *cell = NULL; struct rb_node *p; - int n, seq = 0, ret = 0; + int n; _enter("%*.*s", namesz, namesz, name); @@ -62,57 +56,49 @@ struct afs_cell *afs_lookup_cell_rcu(struct afs_net *net, if (namesz > AFS_MAXCELLNAME) return ERR_PTR(-ENAMETOOLONG); - do { - /* Unfortunately, rbtree walking doesn't give reliable results - * under just the RCU read lock, so we have to check for - * changes. - */ - if (cell) - afs_put_cell(net, cell); - cell = NULL; - ret = -ENOENT; + if (!name) { + cell = rcu_dereference_protected(net->ws_cell, + lockdep_is_held(&net->cells_lock)); + if (!cell) + return ERR_PTR(-EDESTADDRREQ); + goto found; + } - read_seqbegin_or_lock(&net->cells_lock, &seq); + p = net->cells.rb_node; + while (p) { + cell = rb_entry(p, struct afs_cell, net_node); - if (!name) { - cell = rcu_dereference_raw(net->ws_cell); - if (cell) { - afs_get_cell(cell); - break; - } - ret = -EDESTADDRREQ; - continue; - } + n = strncasecmp(cell->name, name, + min_t(size_t, cell->name_len, namesz)); + if (n == 0) + n = cell->name_len - namesz; + if (n < 0) + p = p->rb_left; + else if (n > 0) + p = p->rb_right; + else + goto found; + } - p = rcu_dereference_raw(net->cells.rb_node); - while (p) { - cell = rb_entry(p, struct afs_cell, net_node); - - n = strncasecmp(cell->name, name, - min_t(size_t, cell->name_len, namesz)); - if (n == 0) - n = cell->name_len - namesz; - if (n < 0) { - p = rcu_dereference_raw(p->rb_left); - } else if (n > 0) { - p = rcu_dereference_raw(p->rb_right); - } else { - if (atomic_inc_not_zero(&cell->usage)) { - ret = 0; - break; - } - /* We want to repeat the search, this time with - * the lock properly locked. - */ - } - cell = NULL; - } + return ERR_PTR(-ENOENT); - } while (need_seqretry(&net->cells_lock, seq)); +found: + return afs_use_cell(cell, reason); +} - done_seqretry(&net->cells_lock, seq); +/* + * Look up and get an activation reference on a cell record. + */ +struct afs_cell *afs_find_cell(struct afs_net *net, + const char *name, unsigned int namesz, + enum afs_cell_trace reason) +{ + struct afs_cell *cell; - return ret == 0 ? cell : ERR_PTR(ret); + down_read(&net->cells_lock); + cell = afs_find_cell_locked(net, name, namesz, reason); + up_read(&net->cells_lock); + return cell; } /* @@ -123,6 +109,7 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net, const char *name, unsigned int namelen, const char *addresses) { + struct afs_vlserver_list *vllist = NULL; struct afs_cell *cell; int i, ret; @@ -133,8 +120,17 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net, _leave(" = -ENAMETOOLONG"); return ERR_PTR(-ENAMETOOLONG); } - if (namelen == 5 && memcmp(name, "@cell", 5) == 0) + + /* Prohibit cell names that contain unprintable chars, '/' and '@' or + * that begin with a dot. This also precludes "@cell". + */ + if (name[0] == '.') return ERR_PTR(-EINVAL); + for (i = 0; i < namelen; i++) { + char ch = name[i]; + if (!isprint(ch) || ch == '/' || ch == '@') + return ERR_PTR(-EINVAL); + } _enter("%*.*s,%s", namelen, namelen, name, addresses); @@ -144,45 +140,90 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net, return ERR_PTR(-ENOMEM); } - cell->net = net; + /* Allocate the cell name and the key name in one go. */ + cell->name = kmalloc(1 + namelen + 1 + + 4 + namelen + 1, GFP_KERNEL); + if (!cell->name) { + kfree(cell); + return ERR_PTR(-ENOMEM); + } + + cell->name[0] = '.'; + cell->name++; cell->name_len = namelen; for (i = 0; i < namelen; i++) cell->name[i] = tolower(name[i]); + cell->name[i++] = 0; + + cell->key_desc = cell->name + i; + memcpy(cell->key_desc, "afs@", 4); + memcpy(cell->key_desc + 4, cell->name, cell->name_len + 1); - atomic_set(&cell->usage, 2); - INIT_WORK(&cell->manager, afs_manage_cell); - cell->flags = ((1 << AFS_CELL_FL_NOT_READY) | - (1 << AFS_CELL_FL_NO_LOOKUP_YET)); - INIT_LIST_HEAD(&cell->proc_volumes); - rwlock_init(&cell->proc_lock); + cell->net = net; + refcount_set(&cell->ref, 1); + atomic_set(&cell->active, 0); + INIT_WORK(&cell->destroyer, afs_destroy_cell_work); + INIT_WORK(&cell->manager, afs_manage_cell_work); + timer_setup(&cell->management_timer, afs_cell_timer, 0); + init_rwsem(&cell->vs_lock); + cell->volumes = RB_ROOT; + INIT_HLIST_HEAD(&cell->proc_volumes); + seqlock_init(&cell->volume_lock); + cell->fs_servers = RB_ROOT; + init_rwsem(&cell->fs_lock); rwlock_init(&cell->vl_servers_lock); + cell->flags = (1 << AFS_CELL_FL_CHECK_ALIAS); - /* Fill in the VL server list if we were given a list of addresses to - * use. + /* Provide a VL server list, filling it in if we were given a list of + * addresses to use. */ if (addresses) { - struct afs_vlserver_list *vllist; - vllist = afs_parse_text_addrs(net, addresses, strlen(addresses), ':', VL_SERVICE, AFS_VL_PORT); if (IS_ERR(vllist)) { ret = PTR_ERR(vllist); + vllist = NULL; goto parse_failed; } - rcu_assign_pointer(cell->vl_servers, vllist); + vllist->source = DNS_RECORD_FROM_CONFIG; + vllist->status = DNS_LOOKUP_NOT_DONE; cell->dns_expiry = TIME64_MAX; } else { + ret = -ENOMEM; + vllist = afs_alloc_vlserver_list(0); + if (!vllist) + goto error; + vllist->source = DNS_RECORD_UNAVAILABLE; + vllist->status = DNS_LOOKUP_NOT_DONE; cell->dns_expiry = ktime_get_real_seconds(); } + rcu_assign_pointer(cell->vl_servers, vllist); + + cell->dns_source = vllist->source; + cell->dns_status = vllist->status; + smp_store_release(&cell->dns_lookup_count, 1); /* vs source/status */ + atomic_inc(&net->cells_outstanding); + ret = idr_alloc_cyclic(&net->cells_dyn_ino, cell, + 2, INT_MAX / 2, GFP_KERNEL); + if (ret < 0) + goto error; + cell->dynroot_ino = ret; + cell->debug_id = atomic_inc_return(&cell_debug_id); + + trace_afs_cell(cell->debug_id, 1, 0, afs_cell_trace_alloc); + _leave(" = %p", cell); return cell; parse_failed: if (ret == -EINVAL) printk(KERN_ERR "kAFS: bad VL server IP address\n"); +error: + afs_put_vlserverlist(cell->net, vllist); + kfree(cell->name - 1); kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); @@ -194,7 +235,8 @@ parse_failed: * @name: The name of the cell. * @namesz: The strlen of the cell name. * @vllist: A colon/comma separated list of numeric IP addresses or NULL. - * @excl: T if an error should be given if the cell name already exists. + * @reason: The reason we're doing the lookup + * @trace: The reason to be logged if the lookup is successful. * * Look up a cell record by name and query the DNS for VL server addresses if * needed. Note that that actual DNS query is punted off to the manager thread @@ -203,20 +245,27 @@ parse_failed: */ struct afs_cell *afs_lookup_cell(struct afs_net *net, const char *name, unsigned int namesz, - const char *vllist, bool excl) + const char *vllist, + enum afs_lookup_cell_for reason, + enum afs_cell_trace trace) { struct afs_cell *cell, *candidate, *cursor; struct rb_node *parent, **pp; + enum afs_cell_state state; int ret, n; - _enter("%s,%s", name, vllist); + _enter("%s,%s,%u", name, vllist, reason); - if (!excl) { - rcu_read_lock(); - cell = afs_lookup_cell_rcu(net, name, namesz); - rcu_read_unlock(); - if (!IS_ERR(cell)) + if (reason != AFS_LOOKUP_CELL_PRELOAD) { + cell = afs_find_cell(net, name, namesz, trace); + if (!IS_ERR(cell)) { + if (reason == AFS_LOOKUP_CELL_DYNROOT) + goto no_wait; + if (cell->state == AFS_CELL_SETTING_UP || + cell->state == AFS_CELL_UNLOOKED) + goto lookup_cell; goto wait_for_cell; + } } /* Assume we're probably going to create a cell and preallocate and @@ -235,7 +284,7 @@ struct afs_cell *afs_lookup_cell(struct afs_net *net, /* Find the insertion point and check to see if someone else added a * cell whilst we were allocating. */ - write_seqlock(&net->cells_lock); + down_write(&net->cells_lock); pp = &net->cells.rb_node; parent = NULL; @@ -257,27 +306,73 @@ struct afs_cell *afs_lookup_cell(struct afs_net *net, cell = candidate; candidate = NULL; + afs_use_cell(cell, trace); rb_link_node_rcu(&cell->net_node, parent, pp); rb_insert_color(&cell->net_node, &net->cells); - atomic_inc(&net->cells_outstanding); - write_sequnlock(&net->cells_lock); + up_write(&net->cells_lock); - queue_work(afs_wq, &cell->manager); +lookup_cell: + if (reason != AFS_LOOKUP_CELL_PRELOAD && + reason != AFS_LOOKUP_CELL_ROOTCELL) { + set_bit(AFS_CELL_FL_DO_LOOKUP, &cell->flags); + afs_queue_cell(cell, afs_cell_trace_queue_new); + } wait_for_cell: - _debug("wait_for_cell"); - ret = wait_on_bit(&cell->flags, AFS_CELL_FL_NOT_READY, TASK_INTERRUPTIBLE); - smp_rmb(); + state = smp_load_acquire(&cell->state); /* vs error */ + switch (state) { + case AFS_CELL_ACTIVE: + case AFS_CELL_DEAD: + break; + case AFS_CELL_UNLOOKED: + default: + if (reason == AFS_LOOKUP_CELL_PRELOAD || + reason == AFS_LOOKUP_CELL_ROOTCELL) + break; + _debug("wait_for_cell"); + afs_see_cell(cell, afs_cell_trace_wait); + wait_var_event(&cell->state, + ({ + state = smp_load_acquire(&cell->state); /* vs error */ + state == AFS_CELL_ACTIVE || state == AFS_CELL_DEAD; + })); + _debug("waited_for_cell %d %d", cell->state, cell->error); + } - switch (READ_ONCE(cell->state)) { - case AFS_CELL_FAILED: +no_wait: + /* Check the state obtained from the wait check. */ + state = smp_load_acquire(&cell->state); /* vs error */ + if (state == AFS_CELL_DEAD) { ret = cell->error; goto error; - default: - _debug("weird %u %d", cell->state, cell->error); - goto error; - case AFS_CELL_ACTIVE: - break; + } + if (state == AFS_CELL_ACTIVE) { + switch (cell->dns_status) { + case DNS_LOOKUP_NOT_DONE: + if (cell->dns_source == DNS_RECORD_FROM_CONFIG) { + ret = 0; + break; + } + fallthrough; + default: + ret = -EIO; + goto error; + case DNS_LOOKUP_GOOD: + case DNS_LOOKUP_GOOD_WITH_BAD: + ret = 0; + break; + case DNS_LOOKUP_GOT_NOT_FOUND: + ret = -ENOENT; + goto error; + case DNS_LOOKUP_BAD: + ret = -EREMOTEIO; + goto error; + case DNS_LOOKUP_GOT_LOCAL_FAILURE: + case DNS_LOOKUP_GOT_TEMP_FAILURE: + case DNS_LOOKUP_GOT_NS_FAILURE: + ret = -EDESTADDRREQ; + goto error; + } } _leave(" = %p [cell]", cell); @@ -286,19 +381,20 @@ wait_for_cell: cell_already_exists: _debug("cell exists"); cell = cursor; - if (excl) { + if (reason == AFS_LOOKUP_CELL_PRELOAD) { ret = -EEXIST; } else { - afs_get_cell(cursor); + afs_use_cell(cursor, trace); ret = 0; } - write_sequnlock(&net->cells_lock); - kfree(candidate); + up_write(&net->cells_lock); + if (candidate) + afs_put_cell(candidate, afs_cell_trace_put_candidate); if (ret == 0) goto wait_for_cell; goto error_noput; error: - afs_put_cell(net, cell); + afs_unuse_cell(cell, afs_cell_trace_unuse_lookup_error); error_noput: _leave(" = %d [error]", ret); return ERR_PTR(ret); @@ -335,23 +431,33 @@ int afs_cell_init(struct afs_net *net, const char *rootcell) len = cp - rootcell; } - /* allocate a cell record for the root cell */ - new_root = afs_lookup_cell(net, rootcell, len, vllist, false); + if (len == 0 || !rootcell[0] || rootcell[0] == '.' || rootcell[len - 1] == '.') + return -EINVAL; + if (memchr(rootcell, '/', len)) + return -EINVAL; + cp = strstr(rootcell, ".."); + if (cp && cp < rootcell + len) + return -EINVAL; + + /* allocate a cell record for the root/workstation cell */ + new_root = afs_lookup_cell(net, rootcell, len, vllist, + AFS_LOOKUP_CELL_ROOTCELL, + afs_cell_trace_use_lookup_ws); if (IS_ERR(new_root)) { _leave(" = %ld", PTR_ERR(new_root)); return PTR_ERR(new_root); } if (!test_and_set_bit(AFS_CELL_FL_NO_GC, &new_root->flags)) - afs_get_cell(new_root); + afs_use_cell(new_root, afs_cell_trace_use_pin); /* install the new cell */ - write_seqlock(&net->cells_lock); - old_root = rcu_access_pointer(net->ws_cell); - rcu_assign_pointer(net->ws_cell, new_root); - write_sequnlock(&net->cells_lock); + down_write(&net->cells_lock); + old_root = rcu_replace_pointer(net->ws_cell, new_root, + lockdep_is_held(&net->cells_lock)); + up_write(&net->cells_lock); - afs_put_cell(net, old_root); + afs_unuse_cell(old_root, afs_cell_trace_unuse_ws); _leave(" = 0"); return 0; } @@ -359,16 +465,48 @@ int afs_cell_init(struct afs_net *net, const char *rootcell) /* * Update a cell's VL server address list from the DNS. */ -static void afs_update_cell(struct afs_cell *cell) +static int afs_update_cell(struct afs_cell *cell) { - struct afs_vlserver_list *vllist, *old; + struct afs_vlserver_list *vllist, *old = NULL, *p; unsigned int min_ttl = READ_ONCE(afs_cell_min_ttl); unsigned int max_ttl = READ_ONCE(afs_cell_max_ttl); time64_t now, expiry = 0; + int ret = 0; _enter("%s", cell->name); vllist = afs_dns_query(cell, &expiry); + if (IS_ERR(vllist)) { + ret = PTR_ERR(vllist); + + _debug("%s: fail %d", cell->name, ret); + if (ret == -ENOMEM) + goto out_wake; + + vllist = afs_alloc_vlserver_list(0); + if (!vllist) { + if (ret >= 0) + ret = -ENOMEM; + goto out_wake; + } + + switch (ret) { + case -ENODATA: + case -EDESTADDRREQ: + vllist->status = DNS_LOOKUP_GOT_NOT_FOUND; + break; + case -EAGAIN: + case -ECONNREFUSED: + vllist->status = DNS_LOOKUP_GOT_TEMP_FAILURE; + break; + default: + vllist->status = DNS_LOOKUP_GOT_LOCAL_FAILURE; + break; + } + } + + _debug("%s: got list %d %d", cell->name, vllist->source, vllist->status); + cell->dns_status = vllist->status; now = ktime_get_real_seconds(); if (min_ttl > max_ttl) @@ -378,48 +516,47 @@ static void afs_update_cell(struct afs_cell *cell) else if (expiry > now + max_ttl) expiry = now + max_ttl; - if (IS_ERR(vllist)) { - switch (PTR_ERR(vllist)) { - case -ENODATA: - case -EDESTADDRREQ: + _debug("%s: status %d", cell->name, vllist->status); + if (vllist->source == DNS_RECORD_UNAVAILABLE) { + switch (vllist->status) { + case DNS_LOOKUP_GOT_NOT_FOUND: /* The DNS said that the cell does not exist or there * weren't any addresses to be had. */ - set_bit(AFS_CELL_FL_NOT_FOUND, &cell->flags); - clear_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags); cell->dns_expiry = expiry; break; - case -EAGAIN: - case -ECONNREFUSED: + case DNS_LOOKUP_BAD: + case DNS_LOOKUP_GOT_LOCAL_FAILURE: + case DNS_LOOKUP_GOT_TEMP_FAILURE: + case DNS_LOOKUP_GOT_NS_FAILURE: default: - set_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags); cell->dns_expiry = now + 10; break; } - - cell->error = -EDESTADDRREQ; } else { - clear_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags); - clear_bit(AFS_CELL_FL_NOT_FOUND, &cell->flags); - - /* Exclusion on changing vl_addrs is achieved by a - * non-reentrant work item. - */ - old = rcu_dereference_protected(cell->vl_servers, true); - rcu_assign_pointer(cell->vl_servers, vllist); cell->dns_expiry = expiry; - - if (old) - afs_put_vlserverlist(cell->net, old); } - if (test_and_clear_bit(AFS_CELL_FL_NO_LOOKUP_YET, &cell->flags)) - wake_up_bit(&cell->flags, AFS_CELL_FL_NO_LOOKUP_YET); + /* Replace the VL server list if the new record has servers or the old + * record doesn't. + */ + write_lock(&cell->vl_servers_lock); + p = rcu_dereference_protected(cell->vl_servers, true); + if (vllist->nr_servers > 0 || p->nr_servers == 0) { + rcu_assign_pointer(cell->vl_servers, vllist); + cell->dns_source = vllist->source; + old = p; + } + write_unlock(&cell->vl_servers_lock); + afs_put_vlserverlist(cell->net, old); - now = ktime_get_real_seconds(); - afs_set_cell_timer(cell->net, cell->dns_expiry - now); - _leave(""); +out_wake: + smp_store_release(&cell->dns_lookup_count, + cell->dns_lookup_count + 1); /* vs source/status */ + wake_up_var(&cell->dns_lookup_count); + _leave(" = %d", ret); + return ret; } /* @@ -428,59 +565,93 @@ static void afs_update_cell(struct afs_cell *cell) static void afs_cell_destroy(struct rcu_head *rcu) { struct afs_cell *cell = container_of(rcu, struct afs_cell, rcu); + struct afs_net *net = cell->net; + int r; _enter("%p{%s}", cell, cell->name); - ASSERTCMP(atomic_read(&cell->usage), ==, 0); + r = refcount_read(&cell->ref); + ASSERTCMP(r, ==, 0); + trace_afs_cell(cell->debug_id, r, atomic_read(&cell->active), afs_cell_trace_free); - afs_put_vlserverlist(cell->net, rcu_access_pointer(cell->vl_servers)); + afs_put_vlserverlist(net, rcu_access_pointer(cell->vl_servers)); + afs_unuse_cell(cell->alias_of, afs_cell_trace_unuse_alias); key_put(cell->anonymous_key); + idr_remove(&net->cells_dyn_ino, cell->dynroot_ino); + kfree(cell->name - 1); kfree(cell); + afs_dec_cells_outstanding(net); _leave(" [destroyed]"); } +static void afs_destroy_cell_work(struct work_struct *work) +{ + struct afs_cell *cell = container_of(work, struct afs_cell, destroyer); + + afs_see_cell(cell, afs_cell_trace_destroy); + timer_delete_sync(&cell->management_timer); + cancel_work_sync(&cell->manager); + call_rcu(&cell->rcu, afs_cell_destroy); +} + /* - * Queue the cell manager. + * Get a reference on a cell record. */ -static void afs_queue_cell_manager(struct afs_net *net) +struct afs_cell *afs_get_cell(struct afs_cell *cell, enum afs_cell_trace reason) { - int outstanding = atomic_inc_return(&net->cells_outstanding); + int r; - _enter("%d", outstanding); - - if (!queue_work(afs_wq, &net->cells_manager)) - afs_dec_cells_outstanding(net); + __refcount_inc(&cell->ref, &r); + trace_afs_cell(cell->debug_id, r + 1, atomic_read(&cell->active), reason); + return cell; } /* - * Cell management timer. We have an increment on cells_outstanding that we - * need to pass along to the work item. + * Drop a reference on a cell record. */ -void afs_cells_timer(struct timer_list *timer) +void afs_put_cell(struct afs_cell *cell, enum afs_cell_trace reason) { - struct afs_net *net = container_of(timer, struct afs_net, cells_timer); - - _enter(""); - if (!queue_work(afs_wq, &net->cells_manager)) - afs_dec_cells_outstanding(net); + if (cell) { + unsigned int debug_id = cell->debug_id; + unsigned int a; + bool zero; + int r; + + a = atomic_read(&cell->active); + zero = __refcount_dec_and_test(&cell->ref, &r); + trace_afs_cell(debug_id, r - 1, a, reason); + if (zero) { + a = atomic_read(&cell->active); + WARN(a != 0, "Cell active count %u > 0\n", a); + WARN_ON(!queue_work(afs_wq, &cell->destroyer)); + } + } } /* - * Get a reference on a cell record. + * Note a cell becoming more active. */ -struct afs_cell *afs_get_cell(struct afs_cell *cell) +struct afs_cell *afs_use_cell(struct afs_cell *cell, enum afs_cell_trace reason) { - atomic_inc(&cell->usage); + int r, a; + + __refcount_inc(&cell->ref, &r); + a = atomic_inc_return(&cell->active); + trace_afs_cell(cell->debug_id, r + 1, a, reason); return cell; } /* - * Drop a reference on a cell record. + * Record a cell becoming less active. When the active counter reaches 1, it + * is scheduled for destruction, but may get reactivated. */ -void afs_put_cell(struct afs_net *net, struct afs_cell *cell) +void afs_unuse_cell(struct afs_cell *cell, enum afs_cell_trace reason) { + unsigned int debug_id; time64_t now, expire_delay; + bool zero; + int r, a; if (!cell) return; @@ -490,42 +661,59 @@ void afs_put_cell(struct afs_net *net, struct afs_cell *cell) now = ktime_get_real_seconds(); cell->last_inactive = now; expire_delay = 0; - if (!test_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags) && - !test_bit(AFS_CELL_FL_NOT_FOUND, &cell->flags)) + if (cell->vl_servers->nr_servers) expire_delay = afs_cell_gc_delay; - if (atomic_dec_return(&cell->usage) > 1) - return; + debug_id = cell->debug_id; + a = atomic_dec_return(&cell->active); + if (!a) + /* 'cell' may now be garbage collected. */ + afs_set_cell_timer(cell, expire_delay); - /* 'cell' may now be garbage collected. */ - afs_set_cell_timer(net, expire_delay); + zero = __refcount_dec_and_test(&cell->ref, &r); + trace_afs_cell(debug_id, r - 1, a, reason); + if (zero) + WARN_ON(!queue_work(afs_wq, &cell->destroyer)); } /* - * Allocate a key to use as a placeholder for anonymous user security. + * Note that a cell has been seen. */ -static int afs_alloc_anon_key(struct afs_cell *cell) +void afs_see_cell(struct afs_cell *cell, enum afs_cell_trace reason) { - struct key *key; - char keyname[4 + AFS_MAXCELLNAME + 1], *cp, *dp; + int r, a; - /* Create a key to represent an anonymous user. */ - memcpy(keyname, "afs@", 4); - dp = keyname + 4; - cp = cell->name; - do { - *dp++ = tolower(*cp); - } while (*cp++); + r = refcount_read(&cell->ref); + a = atomic_read(&cell->active); + trace_afs_cell(cell->debug_id, r, a, reason); +} + +/* + * Queue a cell for management, giving the workqueue a ref to hold. + */ +void afs_queue_cell(struct afs_cell *cell, enum afs_cell_trace reason) +{ + queue_work(afs_wq, &cell->manager); +} - key = rxrpc_get_null_key(keyname); - if (IS_ERR(key)) - return PTR_ERR(key); +/* + * Cell-specific management timer. + */ +static void afs_cell_timer(struct timer_list *timer) +{ + struct afs_cell *cell = container_of(timer, struct afs_cell, management_timer); - cell->anonymous_key = key; + afs_see_cell(cell, afs_cell_trace_see_mgmt_timer); + if (refcount_read(&cell->ref) > 0 && cell->net->live) + queue_work(afs_wq, &cell->manager); +} - _debug("anon key %p{%x}", - cell->anonymous_key, key_serial(cell->anonymous_key)); - return 0; +/* + * Set/reduce the cell timer. + */ +void afs_set_cell_timer(struct afs_cell *cell, unsigned int delay_secs) +{ + timer_reduce(&cell->management_timer, jiffies + delay_secs * HZ); } /* @@ -537,19 +725,6 @@ static int afs_activate_cell(struct afs_net *net, struct afs_cell *cell) struct afs_cell *pcell; int ret; - if (!cell->anonymous_key) { - ret = afs_alloc_anon_key(cell); - if (ret < 0) - return ret; - } - -#ifdef CONFIG_AFS_FSCACHE - cell->cache = fscache_acquire_cookie(afs_cache_netfs.primary_index, - &afs_cell_cache_index_def, - cell->name, strlen(cell->name), - NULL, 0, - cell, 0, true); -#endif ret = afs_proc_cell_setup(cell); if (ret < 0) return ret; @@ -567,7 +742,6 @@ static int afs_activate_cell(struct afs_net *net, struct afs_cell *cell) if (cell->proc_link.next) cell->proc_link.next->pprev = &cell->proc_link.next; - afs_dynroot_mkdir(net, cell); mutex_unlock(&net->proc_cells_lock); return 0; } @@ -582,203 +756,135 @@ static void afs_deactivate_cell(struct afs_net *net, struct afs_cell *cell) afs_proc_cell_remove(cell); mutex_lock(&net->proc_cells_lock); - hlist_del_rcu(&cell->proc_link); - afs_dynroot_rmdir(net, cell); + if (!hlist_unhashed(&cell->proc_link)) + hlist_del_rcu(&cell->proc_link); mutex_unlock(&net->proc_cells_lock); -#ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(cell->cache, NULL, false); - cell->cache = NULL; -#endif - _leave(""); } +static bool afs_has_cell_expired(struct afs_cell *cell, time64_t *_next_manage) +{ + const struct afs_vlserver_list *vllist; + time64_t expire_at = cell->last_inactive; + time64_t now = ktime_get_real_seconds(); + + if (atomic_read(&cell->active)) + return false; + if (!cell->net->live) + return true; + + vllist = rcu_dereference_protected(cell->vl_servers, true); + if (vllist && vllist->nr_servers > 0) + expire_at += afs_cell_gc_delay; + + if (expire_at <= now) + return true; + if (expire_at < *_next_manage) + *_next_manage = expire_at; + return false; +} + /* * Manage a cell record, initialising and destroying it, maintaining its DNS * records. */ -static void afs_manage_cell(struct work_struct *work) +static bool afs_manage_cell(struct afs_cell *cell) { - struct afs_cell *cell = container_of(work, struct afs_cell, manager); struct afs_net *net = cell->net; - bool deleted; - int ret, usage; + time64_t next_manage = TIME64_MAX; + int ret; _enter("%s", cell->name); -again: _debug("state %u", cell->state); switch (cell->state) { - case AFS_CELL_INACTIVE: - case AFS_CELL_FAILED: - write_seqlock(&net->cells_lock); - usage = 1; - deleted = atomic_try_cmpxchg_relaxed(&cell->usage, &usage, 0); - if (deleted) - rb_erase(&cell->net_node, &net->cells); - write_sequnlock(&net->cells_lock); - if (deleted) - goto final_destruction; - if (cell->state == AFS_CELL_FAILED) - goto done; - cell->state = AFS_CELL_UNSET; - goto again; - - case AFS_CELL_UNSET: - cell->state = AFS_CELL_ACTIVATING; - goto again; - - case AFS_CELL_ACTIVATING: - ret = afs_activate_cell(net, cell); - if (ret < 0) - goto activation_failed; - - cell->state = AFS_CELL_ACTIVE; - smp_wmb(); - clear_bit(AFS_CELL_FL_NOT_READY, &cell->flags); - wake_up_bit(&cell->flags, AFS_CELL_FL_NOT_READY); - goto again; - + case AFS_CELL_SETTING_UP: + goto set_up_cell; + case AFS_CELL_UNLOOKED: case AFS_CELL_ACTIVE: - if (atomic_read(&cell->usage) > 1) { - time64_t now = ktime_get_real_seconds(); - if (cell->dns_expiry <= now && net->live) - afs_update_cell(cell); - goto done; - } - cell->state = AFS_CELL_DEACTIVATING; - goto again; - - case AFS_CELL_DEACTIVATING: - set_bit(AFS_CELL_FL_NOT_READY, &cell->flags); - if (atomic_read(&cell->usage) > 1) - goto reverse_deactivation; - afs_deactivate_cell(net, cell); - cell->state = AFS_CELL_INACTIVE; - goto again; - + goto cell_is_active; + case AFS_CELL_REMOVING: + WARN_ON_ONCE(1); + return false; + case AFS_CELL_DEAD: + return false; default: - break; + _debug("bad state %u", cell->state); + WARN_ON_ONCE(1); /* Unhandled state */ + return false; } - _debug("bad state %u", cell->state); - BUG(); /* Unhandled state */ - -activation_failed: - cell->error = ret; - afs_deactivate_cell(net, cell); - - cell->state = AFS_CELL_FAILED; - smp_wmb(); - if (test_and_clear_bit(AFS_CELL_FL_NOT_READY, &cell->flags)) - wake_up_bit(&cell->flags, AFS_CELL_FL_NOT_READY); - goto again; - -reverse_deactivation: - cell->state = AFS_CELL_ACTIVE; - smp_wmb(); - clear_bit(AFS_CELL_FL_NOT_READY, &cell->flags); - wake_up_bit(&cell->flags, AFS_CELL_FL_NOT_READY); - _leave(" [deact->act]"); - return; - -done: - _leave(" [done %u]", cell->state); - return; -final_destruction: - call_rcu(&cell->rcu, afs_cell_destroy); - afs_dec_cells_outstanding(net); - _leave(" [destruct %d]", atomic_read(&net->cells_outstanding)); -} +set_up_cell: + ret = afs_activate_cell(net, cell); + if (ret < 0) { + cell->error = ret; + goto remove_cell; + } -/* - * Manage the records of cells known to a network namespace. This includes - * updating the DNS records and garbage collecting unused cells that were - * automatically added. - * - * Note that constructed cell records may only be removed from net->cells by - * this work item, so it is safe for this work item to stash a cursor pointing - * into the tree and then return to caller (provided it skips cells that are - * still under construction). - * - * Note also that we were given an increment on net->cells_outstanding by - * whoever queued us that we need to deal with before returning. - */ -void afs_manage_cells(struct work_struct *work) -{ - struct afs_net *net = container_of(work, struct afs_net, cells_manager); - struct rb_node *cursor; - time64_t now = ktime_get_real_seconds(), next_manage = TIME64_MAX; - bool purging = !net->live; + afs_set_cell_state(cell, AFS_CELL_UNLOOKED); - _enter(""); +cell_is_active: + if (afs_has_cell_expired(cell, &next_manage)) + goto remove_cell; - /* Trawl the cell database looking for cells that have expired from - * lack of use and cells whose DNS results have expired and dispatch - * their managers. - */ - read_seqlock_excl(&net->cells_lock); + if (test_and_clear_bit(AFS_CELL_FL_DO_LOOKUP, &cell->flags)) { + ret = afs_update_cell(cell); + if (ret < 0) + cell->error = ret; + if (cell->state == AFS_CELL_UNLOOKED) + afs_set_cell_state(cell, AFS_CELL_ACTIVE); + } - for (cursor = rb_first(&net->cells); cursor; cursor = rb_next(cursor)) { - struct afs_cell *cell = - rb_entry(cursor, struct afs_cell, net_node); - unsigned usage; - bool sched_cell = false; + if (next_manage < TIME64_MAX && cell->net->live) { + time64_t now = ktime_get_real_seconds(); - usage = atomic_read(&cell->usage); - _debug("manage %s %u", cell->name, usage); + if (next_manage - now <= 0) + afs_queue_cell(cell, afs_cell_trace_queue_again); + else + afs_set_cell_timer(cell, next_manage - now); + } + _leave(" [done %u]", cell->state); + return false; - ASSERTCMP(usage, >=, 1); +remove_cell: + down_write(&net->cells_lock); - if (purging) { - if (test_and_clear_bit(AFS_CELL_FL_NO_GC, &cell->flags)) - usage = atomic_dec_return(&cell->usage); - ASSERTCMP(usage, ==, 1); - } + if (atomic_read(&cell->active)) { + up_write(&net->cells_lock); + goto cell_is_active; + } - if (usage == 1) { - time64_t expire_at = cell->last_inactive; + /* Make sure that the expiring server records are going to see the fact + * that the cell is caput. + */ + afs_set_cell_state(cell, AFS_CELL_REMOVING); - if (!test_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags) && - !test_bit(AFS_CELL_FL_NOT_FOUND, &cell->flags)) - expire_at += afs_cell_gc_delay; - if (purging || expire_at <= now) - sched_cell = true; - else if (expire_at < next_manage) - next_manage = expire_at; - } + afs_deactivate_cell(net, cell); + afs_purge_servers(cell); - if (!purging) { - if (cell->dns_expiry <= now) - sched_cell = true; - else if (cell->dns_expiry <= next_manage) - next_manage = cell->dns_expiry; - } + rb_erase(&cell->net_node, &net->cells); + afs_see_cell(cell, afs_cell_trace_unuse_delete); + up_write(&net->cells_lock); - if (sched_cell) - queue_work(afs_wq, &cell->manager); - } + /* The root volume is pinning the cell */ + afs_put_volume(cell->root_volume, afs_volume_trace_put_cell_root); + cell->root_volume = NULL; - read_sequnlock_excl(&net->cells_lock); + afs_set_cell_state(cell, AFS_CELL_DEAD); + return true; +} - /* Update the timer on the way out. We have to pass an increment on - * cells_outstanding in the namespace that we are in to the timer or - * the work scheduler. - */ - if (!purging && next_manage < TIME64_MAX) { - now = ktime_get_real_seconds(); - - if (next_manage - now <= 0) { - if (queue_work(afs_wq, &net->cells_manager)) - atomic_inc(&net->cells_outstanding); - } else { - afs_set_cell_timer(net, next_manage - now); - } - } +static void afs_manage_cell_work(struct work_struct *work) +{ + struct afs_cell *cell = container_of(work, struct afs_cell, manager); + bool final_put; - afs_dec_cells_outstanding(net); - _leave(" [%d]", atomic_read(&net->cells_outstanding)); + afs_see_cell(cell, afs_cell_trace_manage); + final_put = afs_manage_cell(cell); + afs_see_cell(cell, afs_cell_trace_managed); + if (final_put) + afs_put_cell(cell, afs_cell_trace_put_final); } /* @@ -787,21 +893,29 @@ void afs_manage_cells(struct work_struct *work) void afs_cell_purge(struct afs_net *net) { struct afs_cell *ws; + struct rb_node *cursor; _enter(""); - write_seqlock(&net->cells_lock); - ws = rcu_access_pointer(net->ws_cell); - RCU_INIT_POINTER(net->ws_cell, NULL); - write_sequnlock(&net->cells_lock); - afs_put_cell(net, ws); + down_write(&net->cells_lock); + ws = rcu_replace_pointer(net->ws_cell, NULL, + lockdep_is_held(&net->cells_lock)); + up_write(&net->cells_lock); + afs_unuse_cell(ws, afs_cell_trace_unuse_ws); + + _debug("kick cells"); + down_read(&net->cells_lock); + for (cursor = rb_first(&net->cells); cursor; cursor = rb_next(cursor)) { + struct afs_cell *cell = rb_entry(cursor, struct afs_cell, net_node); - _debug("del timer"); - if (del_timer_sync(&net->cells_timer)) - atomic_dec(&net->cells_outstanding); + afs_see_cell(cell, afs_cell_trace_purge); - _debug("kick mgr"); - afs_queue_cell_manager(net); + if (test_and_clear_bit(AFS_CELL_FL_NO_GC, &cell->flags)) + afs_unuse_cell(cell, afs_cell_trace_unuse_pin); + + afs_queue_cell(cell, afs_cell_trace_queue_purge); + } + up_read(&net->cells_lock); _debug("wait"); wait_var_event(&net->cells_outstanding, diff --git a/fs/afs/cm_security.c b/fs/afs/cm_security.c new file mode 100644 index 000000000000..edcbd249d202 --- /dev/null +++ b/fs/afs/cm_security.c @@ -0,0 +1,340 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Cache manager security. + * + * Copyright (C) 2025 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/slab.h> +#include <crypto/krb5.h> +#include "internal.h" +#include "afs_cm.h" +#include "afs_fs.h" +#include "protocol_yfs.h" +#define RXRPC_TRACE_ONLY_DEFINE_ENUMS +#include <trace/events/rxrpc.h> + +#define RXGK_SERVER_ENC_TOKEN 1036U // 0x40c +#define xdr_round_up(x) (round_up((x), sizeof(__be32))) +#define xdr_len_object(x) (4 + round_up((x), sizeof(__be32))) + +#ifdef CONFIG_RXGK +static int afs_create_yfs_cm_token(struct sk_buff *challenge, + struct afs_server *server); +#endif + +/* + * Respond to an RxGK challenge, adding appdata. + */ +static int afs_respond_to_challenge(struct sk_buff *challenge) +{ +#ifdef CONFIG_RXGK + struct krb5_buffer appdata = {}; + struct afs_server *server; +#endif + struct rxrpc_peer *peer; + unsigned long peer_data; + u16 service_id; + u8 security_index; + + rxrpc_kernel_query_challenge(challenge, &peer, &peer_data, + &service_id, &security_index); + + _enter("%u,%u", service_id, security_index); + + switch (service_id) { + /* We don't send CM_SERVICE RPCs, so don't expect a challenge + * therefrom. + */ + case FS_SERVICE: + case VL_SERVICE: + case YFS_FS_SERVICE: + case YFS_VL_SERVICE: + break; + default: + pr_warn("Can't respond to unknown challenge %u:%u", + service_id, security_index); + return rxrpc_kernel_reject_challenge(challenge, RX_USER_ABORT, -EPROTO, + afs_abort_unsupported_sec_class); + } + + switch (security_index) { +#ifdef CONFIG_RXKAD + case RXRPC_SECURITY_RXKAD: + return rxkad_kernel_respond_to_challenge(challenge); +#endif + +#ifdef CONFIG_RXGK + case RXRPC_SECURITY_RXGK: + return rxgk_kernel_respond_to_challenge(challenge, &appdata); + + case RXRPC_SECURITY_YFS_RXGK: + switch (service_id) { + case FS_SERVICE: + case YFS_FS_SERVICE: + server = (struct afs_server *)peer_data; + if (!server->cm_rxgk_appdata.data) { + mutex_lock(&server->cm_token_lock); + if (!server->cm_rxgk_appdata.data) + afs_create_yfs_cm_token(challenge, server); + mutex_unlock(&server->cm_token_lock); + } + if (server->cm_rxgk_appdata.data) + appdata = server->cm_rxgk_appdata; + break; + } + return rxgk_kernel_respond_to_challenge(challenge, &appdata); +#endif + + default: + return rxrpc_kernel_reject_challenge(challenge, RX_USER_ABORT, -EPROTO, + afs_abort_unsupported_sec_class); + } +} + +/* + * Process the OOB message queue, processing challenge packets. + */ +void afs_process_oob_queue(struct work_struct *work) +{ + struct afs_net *net = container_of(work, struct afs_net, rx_oob_work); + struct sk_buff *oob; + enum rxrpc_oob_type type; + + while ((oob = rxrpc_kernel_dequeue_oob(net->socket, &type))) { + switch (type) { + case RXRPC_OOB_CHALLENGE: + afs_respond_to_challenge(oob); + break; + } + rxrpc_kernel_free_oob(oob); + } +} + +#ifdef CONFIG_RXGK +/* + * Create a securities keyring for the cache manager and attach a key to it for + * the RxGK tokens we want to use to secure the callback connection back from + * the fileserver. + */ +int afs_create_token_key(struct afs_net *net, struct socket *socket) +{ + const struct krb5_enctype *krb5; + struct key *ring; + key_ref_t key; + char K0[32], *desc; + int ret; + + ring = keyring_alloc("kafs", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, current_cred(), + KEY_POS_SEARCH | KEY_POS_WRITE | + KEY_USR_VIEW | KEY_USR_READ | KEY_USR_SEARCH, + KEY_ALLOC_NOT_IN_QUOTA, + NULL, NULL); + if (IS_ERR(ring)) + return PTR_ERR(ring); + + ret = rxrpc_sock_set_security_keyring(socket->sk, ring); + if (ret < 0) + goto out; + + ret = -ENOPKG; + krb5 = crypto_krb5_find_enctype(KRB5_ENCTYPE_AES128_CTS_HMAC_SHA1_96); + if (!krb5) + goto out; + + if (WARN_ON_ONCE(krb5->key_len > sizeof(K0))) + goto out; + + ret = -ENOMEM; + desc = kasprintf(GFP_KERNEL, "%u:%u:%u:%u", + YFS_CM_SERVICE, RXRPC_SECURITY_YFS_RXGK, 1, krb5->etype); + if (!desc) + goto out; + + wait_for_random_bytes(); + get_random_bytes(K0, krb5->key_len); + + key = key_create(make_key_ref(ring, true), + "rxrpc_s", desc, + K0, krb5->key_len, + KEY_POS_VIEW | KEY_POS_READ | KEY_POS_SEARCH | KEY_USR_VIEW, + KEY_ALLOC_NOT_IN_QUOTA); + kfree(desc); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + goto out; + } + + net->fs_cm_token_key = key_ref_to_ptr(key); + ret = 0; +out: + key_put(ring); + return ret; +} + +/* + * Create an YFS RxGK GSS token to use as a ticket to the specified fileserver. + */ +static int afs_create_yfs_cm_token(struct sk_buff *challenge, + struct afs_server *server) +{ + const struct krb5_enctype *conn_krb5, *token_krb5; + const struct krb5_buffer *token_key; + struct crypto_aead *aead; + struct scatterlist sg; + struct afs_net *net = server->cell->net; + const struct key *key = net->fs_cm_token_key; + size_t keysize, uuidsize, authsize, toksize, encsize, contsize, adatasize, offset; + __be32 caps[1] = { + [0] = htonl(AFS_CAP_ERROR_TRANSLATION), + }; + __be32 *xdr; + void *appdata, *K0, *encbase; + u32 enctype; + int ret; + + if (!key) + return -ENOKEY; + + /* Assume that the fileserver is happy to use the same encoding type as + * we were told to use by the token obtained by the user. + */ + enctype = rxgk_kernel_query_challenge(challenge); + + conn_krb5 = crypto_krb5_find_enctype(enctype); + if (!conn_krb5) + return -ENOPKG; + token_krb5 = key->payload.data[0]; + token_key = (const struct krb5_buffer *)&key->payload.data[2]; + + /* struct rxgk_key { + * afs_uint32 enctype; + * opaque key<>; + * }; + */ + keysize = 4 + xdr_len_object(conn_krb5->key_len); + + /* struct RXGK_AuthName { + * afs_int32 kind; + * opaque data<AUTHDATAMAX>; + * opaque display<AUTHPRINTABLEMAX>; + * }; + */ + uuidsize = sizeof(server->uuid); + authsize = 4 + xdr_len_object(uuidsize) + xdr_len_object(0); + + /* struct RXGK_Token { + * rxgk_key K0; + * RXGK_Level level; + * rxgkTime starttime; + * afs_int32 lifetime; + * afs_int32 bytelife; + * rxgkTime expirationtime; + * struct RXGK_AuthName identities<>; + * }; + */ + toksize = keysize + 8 + 4 + 4 + 8 + xdr_len_object(authsize); + + offset = 0; + encsize = crypto_krb5_how_much_buffer(token_krb5, KRB5_ENCRYPT_MODE, toksize, &offset); + + /* struct RXGK_TokenContainer { + * afs_int32 kvno; + * afs_int32 enctype; + * opaque encrypted_token<>; + * }; + */ + contsize = 4 + 4 + xdr_len_object(encsize); + + /* struct YFSAppData { + * opr_uuid initiatorUuid; + * opr_uuid acceptorUuid; + * Capabilities caps; + * afs_int32 enctype; + * opaque callbackKey<>; + * opaque callbackToken<>; + * }; + */ + adatasize = 16 + 16 + + xdr_len_object(sizeof(caps)) + + 4 + + xdr_len_object(conn_krb5->key_len) + + xdr_len_object(contsize); + + ret = -ENOMEM; + appdata = kzalloc(adatasize, GFP_KERNEL); + if (!appdata) + goto out; + xdr = appdata; + + memcpy(xdr, &net->uuid, 16); /* appdata.initiatorUuid */ + xdr += 16 / 4; + memcpy(xdr, &server->uuid, 16); /* appdata.acceptorUuid */ + xdr += 16 / 4; + *xdr++ = htonl(ARRAY_SIZE(caps)); /* appdata.caps.len */ + memcpy(xdr, &caps, sizeof(caps)); /* appdata.caps */ + xdr += ARRAY_SIZE(caps); + *xdr++ = htonl(conn_krb5->etype); /* appdata.enctype */ + + *xdr++ = htonl(conn_krb5->key_len); /* appdata.callbackKey.len */ + K0 = xdr; + get_random_bytes(K0, conn_krb5->key_len); /* appdata.callbackKey.data */ + xdr += xdr_round_up(conn_krb5->key_len) / 4; + + *xdr++ = htonl(contsize); /* appdata.callbackToken.len */ + *xdr++ = htonl(1); /* cont.kvno */ + *xdr++ = htonl(token_krb5->etype); /* cont.enctype */ + *xdr++ = htonl(encsize); /* cont.encrypted_token.len */ + + encbase = xdr; + xdr += offset / 4; + *xdr++ = htonl(conn_krb5->etype); /* token.K0.enctype */ + *xdr++ = htonl(conn_krb5->key_len); /* token.K0.key.len */ + memcpy(xdr, K0, conn_krb5->key_len); /* token.K0.key.data */ + xdr += xdr_round_up(conn_krb5->key_len) / 4; + + *xdr++ = htonl(RXRPC_SECURITY_ENCRYPT); /* token.level */ + *xdr++ = htonl(0); /* token.starttime */ + *xdr++ = htonl(0); /* " */ + *xdr++ = htonl(0); /* token.lifetime */ + *xdr++ = htonl(0); /* token.bytelife */ + *xdr++ = htonl(0); /* token.expirationtime */ + *xdr++ = htonl(0); /* " */ + *xdr++ = htonl(1); /* token.identities.count */ + *xdr++ = htonl(0); /* token.identities[0].kind */ + *xdr++ = htonl(uuidsize); /* token.identities[0].data.len */ + memcpy(xdr, &server->uuid, uuidsize); + xdr += xdr_round_up(uuidsize) / 4; + *xdr++ = htonl(0); /* token.identities[0].display.len */ + + xdr = encbase + xdr_round_up(encsize); + + if ((unsigned long)xdr - (unsigned long)appdata != adatasize) + pr_err("Appdata size incorrect %lx != %zx\n", + (unsigned long)xdr - (unsigned long)appdata, adatasize); + + aead = crypto_krb5_prepare_encryption(token_krb5, token_key, RXGK_SERVER_ENC_TOKEN, + GFP_KERNEL); + if (IS_ERR(aead)) { + ret = PTR_ERR(aead); + goto out_token; + } + + sg_init_one(&sg, encbase, encsize); + ret = crypto_krb5_encrypt(token_krb5, aead, &sg, 1, encsize, offset, toksize, false); + if (ret < 0) + goto out_aead; + + server->cm_rxgk_appdata.len = adatasize; + server->cm_rxgk_appdata.data = appdata; + appdata = NULL; + +out_aead: + crypto_free_aead(aead); +out_token: + kfree(appdata); +out: + return ret; +} +#endif /* CONFIG_RXGK */ diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index 8ee5972893ed..1a906805a9e3 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS Cache Manager Service * * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/module.h> @@ -17,6 +13,8 @@ #include "internal.h" #include "afs_cm.h" #include "protocol_yfs.h" +#define RXRPC_TRACE_ONLY_DEFINE_ENUMS +#include <trace/events/rxrpc.h> static int afs_deliver_cb_init_call_back_state(struct afs_call *); static int afs_deliver_cb_init_call_back_state3(struct afs_call *); @@ -33,16 +31,11 @@ static void SRXAFSCB_TellMeAboutYourself(struct work_struct *); static int afs_deliver_yfs_cb_callback(struct afs_call *); -#define CM_NAME(name) \ - const char afs_SRXCB##name##_name[] __tracepoint_string = \ - "CB." #name - /* * CB.CallBack operation type */ -static CM_NAME(CallBack); static const struct afs_call_type afs_SRXCBCallBack = { - .name = afs_SRXCBCallBack_name, + .name = "CB.CallBack", .deliver = afs_deliver_cb_callback, .destructor = afs_cm_destructor, .work = SRXAFSCB_CallBack, @@ -51,9 +44,8 @@ static const struct afs_call_type afs_SRXCBCallBack = { /* * CB.InitCallBackState operation type */ -static CM_NAME(InitCallBackState); static const struct afs_call_type afs_SRXCBInitCallBackState = { - .name = afs_SRXCBInitCallBackState_name, + .name = "CB.InitCallBackState", .deliver = afs_deliver_cb_init_call_back_state, .destructor = afs_cm_destructor, .work = SRXAFSCB_InitCallBackState, @@ -62,9 +54,8 @@ static const struct afs_call_type afs_SRXCBInitCallBackState = { /* * CB.InitCallBackState3 operation type */ -static CM_NAME(InitCallBackState3); static const struct afs_call_type afs_SRXCBInitCallBackState3 = { - .name = afs_SRXCBInitCallBackState3_name, + .name = "CB.InitCallBackState3", .deliver = afs_deliver_cb_init_call_back_state3, .destructor = afs_cm_destructor, .work = SRXAFSCB_InitCallBackState, @@ -73,9 +64,8 @@ static const struct afs_call_type afs_SRXCBInitCallBackState3 = { /* * CB.Probe operation type */ -static CM_NAME(Probe); static const struct afs_call_type afs_SRXCBProbe = { - .name = afs_SRXCBProbe_name, + .name = "CB.Probe", .deliver = afs_deliver_cb_probe, .destructor = afs_cm_destructor, .work = SRXAFSCB_Probe, @@ -84,9 +74,8 @@ static const struct afs_call_type afs_SRXCBProbe = { /* * CB.ProbeUuid operation type */ -static CM_NAME(ProbeUuid); static const struct afs_call_type afs_SRXCBProbeUuid = { - .name = afs_SRXCBProbeUuid_name, + .name = "CB.ProbeUuid", .deliver = afs_deliver_cb_probe_uuid, .destructor = afs_cm_destructor, .work = SRXAFSCB_ProbeUuid, @@ -95,9 +84,8 @@ static const struct afs_call_type afs_SRXCBProbeUuid = { /* * CB.TellMeAboutYourself operation type */ -static CM_NAME(TellMeAboutYourself); static const struct afs_call_type afs_SRXCBTellMeAboutYourself = { - .name = afs_SRXCBTellMeAboutYourself_name, + .name = "CB.TellMeAboutYourself", .deliver = afs_deliver_cb_tell_me_about_yourself, .destructor = afs_cm_destructor, .work = SRXAFSCB_TellMeAboutYourself, @@ -106,9 +94,8 @@ static const struct afs_call_type afs_SRXCBTellMeAboutYourself = { /* * YFS CB.CallBack operation type */ -static CM_NAME(YFS_CallBack); static const struct afs_call_type afs_SRXYFSCB_CallBack = { - .name = afs_SRXCBYFS_CallBack_name, + .name = "YFSCB.CallBack", .deliver = afs_deliver_yfs_cb_callback, .destructor = afs_cm_destructor, .work = SRXAFSCB_CallBack, @@ -122,8 +109,6 @@ bool afs_cm_incoming_call(struct afs_call *call) { _enter("{%u, CB.OP %u}", call->service_id, call->operation_ID); - call->epoch = rxrpc_kernel_get_epoch(call->net->socket, call->rxcall); - switch (call->operation_ID) { case CBCallBack: call->type = &afs_SRXCBCallBack; @@ -154,97 +139,23 @@ bool afs_cm_incoming_call(struct afs_call *call) } /* - * Record a probe to the cache manager from a server. - */ -static int afs_record_cm_probe(struct afs_call *call, struct afs_server *server) -{ - _enter(""); - - if (test_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags) && - !test_bit(AFS_SERVER_FL_PROBING, &server->flags)) { - if (server->cm_epoch == call->epoch) - return 0; - - if (!server->probe.said_rebooted) { - pr_notice("kAFS: FS rebooted %pU\n", &server->uuid); - server->probe.said_rebooted = true; - } - } - - spin_lock(&server->probe_lock); - - if (!test_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags)) { - server->cm_epoch = call->epoch; - server->probe.cm_epoch = call->epoch; - goto out; - } - - if (server->probe.cm_probed && - call->epoch != server->probe.cm_epoch && - !server->probe.said_inconsistent) { - pr_notice("kAFS: FS endpoints inconsistent %pU\n", - &server->uuid); - server->probe.said_inconsistent = true; - } - - if (!server->probe.cm_probed || call->epoch == server->cm_epoch) - server->probe.cm_epoch = server->cm_epoch; - -out: - server->probe.cm_probed = true; - spin_unlock(&server->probe_lock); - return 0; -} - -/* - * Find the server record by peer address and record a probe to the cache - * manager from a server. - */ -static int afs_find_cm_server_by_peer(struct afs_call *call) -{ - struct sockaddr_rxrpc srx; - struct afs_server *server; - - rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx); - - server = afs_find_server(call->net, &srx); - if (!server) { - trace_afs_cm_no_server(call, &srx); - return 0; - } - - call->cm_server = server; - return afs_record_cm_probe(call, server); -} - -/* - * Find the server record by server UUID and record a probe to the cache - * manager from a server. + * Clean up a cache manager call. */ -static int afs_find_cm_server_by_uuid(struct afs_call *call, - struct afs_uuid *uuid) +static void afs_cm_destructor(struct afs_call *call) { - struct afs_server *server; - - rcu_read_lock(); - server = afs_find_server_by_uuid(call->net, call->request); - rcu_read_unlock(); - if (!server) { - trace_afs_cm_no_server_u(call, call->request); - return 0; - } - - call->cm_server = server; - return afs_record_cm_probe(call, server); + kfree(call->buffer); + call->buffer = NULL; } /* - * Clean up a cache manager call. + * Abort a service call from within an action function. */ -static void afs_cm_destructor(struct afs_call *call) +static void afs_abort_service_call(struct afs_call *call, u32 abort_code, int error, + enum rxrpc_abort_reason why) { - kfree(call->buffer); - call->buffer = NULL; + rxrpc_kernel_abort_call(call->net->socket, call->rxcall, + abort_code, error, why); + afs_set_call_complete(call, error, 0); } /* @@ -260,8 +171,13 @@ static void SRXAFSCB_CallBack(struct work_struct *work) * server holds up change visibility till it receives our reply so as * to maintain cache coherency. */ - if (call->cm_server) - afs_break_callbacks(call->cm_server, call->count, call->request); + if (call->server) { + trace_afs_server(call->server->debug_id, + refcount_read(&call->server->ref), + atomic_read(&call->server->active), + afs_server_trace_callback); + afs_break_callbacks(call->server, call->count, call->request); + } afs_send_empty_reply(call); afs_put_call(call); @@ -285,6 +201,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->unmarshall++; /* extract the FID array and its count in two steps */ + fallthrough; case 1: _debug("extract FID count"); ret = afs_extract_data(call, true); @@ -294,8 +211,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->count = ntohl(call->tmp); _debug("FID count: %u", call->count); if (call->count > AFSCBMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_cb_fid_count); + return afs_protocol_error(call, afs_eproto_cb_fid_count); call->buffer = kmalloc(array3_size(call->count, 3, 4), GFP_KERNEL); @@ -304,6 +220,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) afs_extract_to_buf(call, call->count * 3 * 4); call->unmarshall++; + fallthrough; case 2: _debug("extract FID array"); ret = afs_extract_data(call, true); @@ -329,6 +246,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->unmarshall++; /* extract the callback array and its count in two steps */ + fallthrough; case 3: _debug("extract CB count"); ret = afs_extract_data(call, true); @@ -338,31 +256,30 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->count2 = ntohl(call->tmp); _debug("CB count: %u", call->count2); if (call->count2 != call->count && call->count2 != 0) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_cb_count); - call->_iter = &call->iter; - iov_iter_discard(&call->iter, READ, call->count2 * 3 * 4); + return afs_protocol_error(call, afs_eproto_cb_count); + call->iter = &call->def_iter; + iov_iter_discard(&call->def_iter, ITER_DEST, call->count2 * 3 * 4); call->unmarshall++; + fallthrough; case 4: _debug("extract discard %zu/%u", - iov_iter_count(&call->iter), call->count2 * 3 * 4); + iov_iter_count(call->iter), call->count2 * 3 * 4); ret = afs_extract_data(call, false); if (ret < 0) return ret; call->unmarshall++; + fallthrough; + case 5: break; } if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - - /* we'll need the file server record as that tells us which set of - * vnodes to operate upon */ - return afs_find_cm_server_by_peer(call); + return 0; } /* @@ -372,10 +289,10 @@ static void SRXAFSCB_InitCallBackState(struct work_struct *work) { struct afs_call *call = container_of(work, struct afs_call, work); - _enter("{%p}", call->cm_server); + _enter("{%p}", call->server); - if (call->cm_server) - afs_init_callback_state(call->cm_server); + if (call->server) + afs_init_callback_state(call->server); afs_send_empty_reply(call); afs_put_call(call); _leave(""); @@ -386,18 +303,10 @@ static void SRXAFSCB_InitCallBackState(struct work_struct *work) */ static int afs_deliver_cb_init_call_back_state(struct afs_call *call) { - int ret; - _enter(""); afs_extract_discard(call, 0); - ret = afs_extract_data(call, false); - if (ret < 0) - return ret; - - /* we'll need the file server record as that tells us which set of - * vnodes to operate upon */ - return afs_find_cm_server_by_peer(call); + return afs_extract_data(call, false); } /* @@ -410,8 +319,6 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) __be32 *b; int ret; - _enter(""); - _enter("{%u}", call->unmarshall); switch (call->unmarshall) { @@ -422,6 +329,7 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) afs_extract_to_buf(call, 11 * sizeof(__be32)); call->unmarshall++; + fallthrough; case 1: _debug("extract UUID"); ret = afs_extract_data(call, false); @@ -448,6 +356,7 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) r->node[loop] = ntohl(b[loop + 5]); call->unmarshall++; + fallthrough; case 2: break; @@ -456,9 +365,13 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - /* we'll need the file server record as that tells us which set of - * vnodes to operate upon */ - return afs_find_cm_server_by_uuid(call, call->request); + if (memcmp(call->request, &call->server->_uuid, sizeof(call->server->_uuid)) != 0) { + pr_notice("Callback UUID does not match fileserver UUID\n"); + trace_afs_cm_no_server_u(call, call->request); + return 0; + } + + return 0; } /* @@ -490,29 +403,25 @@ static int afs_deliver_cb_probe(struct afs_call *call) if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - return afs_find_cm_server_by_peer(call); + return 0; } /* - * allow the fileserver to quickly find out if the fileserver has been rebooted + * Allow the fileserver to quickly find out if the cache manager has been + * rebooted. */ static void SRXAFSCB_ProbeUuid(struct work_struct *work) { struct afs_call *call = container_of(work, struct afs_call, work); struct afs_uuid *r = call->request; - struct { - __be32 match; - } reply; - _enter(""); if (memcmp(r, &call->net->uuid, sizeof(call->net->uuid)) == 0) - reply.match = htonl(0); + afs_send_empty_reply(call); else - reply.match = htonl(1); + afs_abort_service_call(call, 1, 1, afs_abort_probeuuid_negative); - afs_send_simple_reply(call, &reply, sizeof(reply)); afs_put_call(call); _leave(""); } @@ -537,6 +446,7 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call) afs_extract_to_buf(call, 11 * sizeof(__be32)); call->unmarshall++; + fallthrough; case 1: _debug("extract UUID"); ret = afs_extract_data(call, false); @@ -563,6 +473,7 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call) r->node[loop] = ntohl(b[loop + 5]); call->unmarshall++; + fallthrough; case 2: break; @@ -570,7 +481,7 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call) if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - return afs_find_cm_server_by_uuid(call, call->request); + return 0; } /* @@ -578,9 +489,8 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call) */ static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work) { - struct afs_interface *ifs; struct afs_call *call = container_of(work, struct afs_call, work); - int loop, nifs; + int loop; struct { struct /* InterfaceAddr */ { @@ -598,19 +508,7 @@ static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work) _enter(""); - nifs = 0; - ifs = kcalloc(32, sizeof(*ifs), GFP_KERNEL); - if (ifs) { - nifs = afs_get_ipv4_interfaces(call->net, ifs, 32, false); - if (nifs < 0) { - kfree(ifs); - ifs = NULL; - nifs = 0; - } - } - memset(&reply, 0, sizeof(reply)); - reply.ia.nifs = htonl(nifs); reply.ia.uuid[0] = call->net->uuid.time_low; reply.ia.uuid[1] = htonl(ntohs(call->net->uuid.time_mid)); @@ -620,15 +518,6 @@ static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work) for (loop = 0; loop < 6; loop++) reply.ia.uuid[loop + 5] = htonl((s8) call->net->uuid.node[loop]); - if (ifs) { - for (loop = 0; loop < nifs; loop++) { - reply.ia.ifaddr[loop] = ifs[loop].address.s_addr; - reply.ia.netmask[loop] = ifs[loop].netmask.s_addr; - reply.ia.mtu[loop] = htonl(ifs[loop].mtu); - } - kfree(ifs); - } - reply.cap.capcount = htonl(1); reply.cap.caps[0] = htonl(AFS_CAP_ERROR_TRANSLATION); afs_send_simple_reply(call, &reply, sizeof(reply)); @@ -652,7 +541,7 @@ static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *call) if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - return afs_find_cm_server_by_peer(call); + return 0; } /* @@ -673,6 +562,7 @@ static int afs_deliver_yfs_cb_callback(struct afs_call *call) call->unmarshall++; /* extract the FID array and its count in two steps */ + fallthrough; case 1: _debug("extract FID count"); ret = afs_extract_data(call, true); @@ -682,8 +572,7 @@ static int afs_deliver_yfs_cb_callback(struct afs_call *call) call->count = ntohl(call->tmp); _debug("FID count: %u", call->count); if (call->count > YFSCBMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_cb_fid_count); + return afs_protocol_error(call, afs_eproto_cb_fid_count); size = array_size(call->count, sizeof(struct yfs_xdr_YFSFid)); call->buffer = kmalloc(size, GFP_KERNEL); @@ -692,6 +581,7 @@ static int afs_deliver_yfs_cb_callback(struct afs_call *call) afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; case 2: _debug("extract FID array"); ret = afs_extract_data(call, false); @@ -717,6 +607,7 @@ static int afs_deliver_yfs_cb_callback(struct afs_call *call) afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; case 3: break; @@ -724,9 +615,5 @@ static int afs_deliver_yfs_cb_callback(struct afs_call *call) if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING)) return afs_io_error(call, afs_io_error_cm_reply); - - /* We'll need the file server record as that tells us which set of - * vnodes to operate upon. - */ - return afs_find_cm_server_by_peer(call); + return 0; } diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 8a2562e3a316..f4e9e12373ac 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* dir.c: AFS filesystem directory handling * * Copyright (C) 2002, 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -16,40 +12,38 @@ #include <linux/swap.h> #include <linux/ctype.h> #include <linux/sched.h> +#include <linux/iversion.h> +#include <linux/iov_iter.h> #include <linux/task_io_accounting_ops.h> #include "internal.h" +#include "afs_fs.h" #include "xdr_fs.h" static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); static int afs_readdir(struct file *file, struct dir_context *ctx); -static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); +static int afs_d_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags); static int afs_d_delete(const struct dentry *dentry); -static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, +static void afs_d_iput(struct dentry *dentry, struct inode *inode); +static bool afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, +static bool afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl); -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); +static int afs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl); +static struct dentry *afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode); static int afs_rmdir(struct inode *dir, struct dentry *dentry); static int afs_unlink(struct inode *dir, struct dentry *dentry); static int afs_link(struct dentry *from, struct inode *dir, struct dentry *dentry); -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content); -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags); -static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags); -static void afs_dir_invalidatepage(struct page *page, unsigned int offset, - unsigned int length); - -static int afs_dir_set_page_dirty(struct page *page) -{ - BUG(); /* This should never happen. */ -} +static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *content); +static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags); const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, @@ -71,13 +65,10 @@ const struct inode_operations afs_dir_inode_operations = { .permission = afs_permission, .getattr = afs_getattr, .setattr = afs_setattr, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_dir_aops = { - .set_page_dirty = afs_dir_set_page_dirty, - .releasepage = afs_dir_releasepage, - .invalidatepage = afs_dir_invalidatepage, + .writepages = afs_single_writepages, }; const struct dentry_operations afs_fs_dentry_operations = { @@ -85,6 +76,7 @@ const struct dentry_operations afs_fs_dentry_operations = { .d_delete = afs_d_delete, .d_release = afs_d_release, .d_automount = afs_d_automount, + .d_iput = afs_d_iput, }; struct afs_lookup_one_cookie { @@ -97,66 +89,124 @@ struct afs_lookup_one_cookie { struct afs_lookup_cookie { struct dir_context ctx; struct qstr name; - bool found; - bool one_only; unsigned short nr_fids; - struct afs_file_status *statuses; - struct afs_callback *callbacks; struct afs_fid fids[50]; }; +static void afs_dir_unuse_cookie(struct afs_vnode *dvnode, int ret) +{ + if (ret == 0) { + struct afs_vnode_cache_aux aux; + loff_t i_size = i_size_read(&dvnode->netfs.inode); + + afs_set_cache_aux(dvnode, &aux); + fscache_unuse_cookie(afs_vnode_cache(dvnode), &aux, &i_size); + } else { + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + } +} + /* - * check that a directory page is valid + * Iterate through a kmapped directory segment, dumping a summary of + * the contents. */ -static bool afs_dir_check_page(struct afs_vnode *dvnode, struct page *page, - loff_t i_size) +static size_t afs_dir_dump_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct afs_xdr_dir_page *dbuf; - loff_t latter, off; - int tmp, qty; + do { + union afs_xdr_dir_block *block = iter_base; - /* Determine how many magic numbers there should be in this page, but - * we must take care because the directory may change size under us. - */ - off = page_offset(page); - if (i_size <= off) - goto checked; + pr_warn("[%05zx] %32phN\n", progress, block); + iter_base += AFS_DIR_BLOCK_SIZE; + progress += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - latter = i_size - off; - if (latter >= PAGE_SIZE) - qty = PAGE_SIZE; - else - qty = latter; - qty /= sizeof(union afs_xdr_dir_block); - - /* check them */ - dbuf = kmap(page); - for (tmp = 0; tmp < qty; tmp++) { - if (dbuf->blocks[tmp].hdr.magic != AFS_DIR_MAGIC) { - printk("kAFS: %s(%lx): bad magic %d/%d is %04hx\n", - __func__, dvnode->vfs_inode.i_ino, tmp, qty, - ntohs(dbuf->blocks[tmp].hdr.magic)); - trace_afs_dir_check_failed(dvnode, off, i_size); - kunmap(page); - trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); - goto error; - } + return len; +} - /* Make sure each block is NUL terminated so we can reasonably - * use string functions on it. The filenames in the page - * *should* be NUL-terminated anyway. - */ - ((u8 *)&dbuf->blocks[tmp])[AFS_DIR_BLOCK_SIZE - 1] = 0; - } +/* + * Dump the contents of a directory. + */ +static void afs_dir_dump(struct afs_vnode *dvnode) +{ + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); + + pr_warn("DIR %llx:%llx is=%llx\n", + dvnode->fid.vid, dvnode->fid.vnode, i_size); - kunmap(page); + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iterate_folioq(&iter, iov_iter_count(&iter), NULL, NULL, + afs_dir_dump_step); +} + +/* + * check that a directory folio is valid + */ +static bool afs_dir_check_block(struct afs_vnode *dvnode, size_t progress, + union afs_xdr_dir_block *block) +{ + if (block->hdr.magic != AFS_DIR_MAGIC) { + pr_warn("%s(%lx): [%zx] bad magic %04x\n", + __func__, dvnode->netfs.inode.i_ino, + progress, ntohs(block->hdr.magic)); + trace_afs_dir_check_failed(dvnode, progress); + trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); + return false; + } -checked: + /* Make sure each block is NUL terminated so we can reasonably + * use string functions on it. The filenames in the folio + * *should* be NUL-terminated anyway. + */ + ((u8 *)block)[AFS_DIR_BLOCK_SIZE - 1] = 0; afs_stat_v(dvnode, n_read_dir); return true; +} -error: - return false; +/* + * Iterate through a kmapped directory segment, checking the content. + */ +static size_t afs_dir_check_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) +{ + struct afs_vnode *dvnode = priv; + + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) + return len; + + do { + if (!afs_dir_check_block(dvnode, progress, iter_base)) + break; + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); + + return len; +} + +/* + * Check all the blocks in a directory. + */ +static int afs_dir_check(struct afs_vnode *dvnode) +{ + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); + size_t checked = 0; + + if (unlikely(!i_size)) + return 0; + + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + checked = iterate_folioq(&iter, iov_iter_count(&iter), dvnode, NULL, + afs_dir_check_step); + if (checked != i_size) { + afs_dir_dump(dvnode); + return -EIO; + } + return 0; } /* @@ -176,143 +226,140 @@ static int afs_dir_open(struct inode *inode, struct file *file) } /* - * Read the directory into the pagecache in one go, scrubbing the previous - * contents. The list of pages is returned, pinning them so that they don't - * get reclaimed during the iteration. + * Read a file in a single download. */ -static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) - __acquires(&dvnode->validate_lock) +static ssize_t afs_do_read_single(struct afs_vnode *dvnode, struct file *file) { - struct afs_read *req; + struct iov_iter iter; + ssize_t ret; loff_t i_size; - int nr_pages, nr_inline, i, n; - int ret = -ENOMEM; - -retry: - i_size = i_size_read(&dvnode->vfs_inode); - if (i_size < 2048) - return ERR_PTR(afs_bad(dvnode, afs_file_error_dir_small)); - if (i_size > 2048 * 1024) { - trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); - return ERR_PTR(-EFBIG); + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); + + i_size = i_size_read(&dvnode->netfs.inode); + if (is_dir) { + if (i_size < AFS_DIR_BLOCK_SIZE) + return afs_bad(dvnode, afs_file_error_dir_small); + if (i_size > AFS_DIR_BLOCK_SIZE * 1024) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } + } else { + if (i_size > AFSPATHMAX) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } } - _enter("%llu", i_size); + /* Expand the storage. TODO: Shrink the storage too. */ + if (dvnode->directory_size < i_size) { + size_t cur_size = dvnode->directory_size; - /* Get a request record to hold the page list. We want to hold it - * inline if we can, but we don't want to make an order 1 allocation. - */ - nr_pages = (i_size + PAGE_SIZE - 1) / PAGE_SIZE; - nr_inline = nr_pages; - if (nr_inline > (PAGE_SIZE - sizeof(*req)) / sizeof(struct page *)) - nr_inline = 0; - - req = kzalloc(sizeof(*req) + sizeof(struct page *) * nr_inline, - GFP_KERNEL); - if (!req) - return ERR_PTR(-ENOMEM); - - refcount_set(&req->usage, 1); - req->nr_pages = nr_pages; - req->actual_len = i_size; /* May change */ - req->len = nr_pages * PAGE_SIZE; /* We can ask for more than there is */ - req->data_version = dvnode->status.data_version; /* May change */ - if (nr_inline > 0) { - req->pages = req->array; - } else { - req->pages = kcalloc(nr_pages, sizeof(struct page *), - GFP_KERNEL); - if (!req->pages) - goto error; + ret = netfs_alloc_folioq_buffer(NULL, + &dvnode->directory, &cur_size, i_size, + mapping_gfp_mask(dvnode->netfs.inode.i_mapping)); + dvnode->directory_size = cur_size; + if (ret < 0) + return ret; } - /* Get a list of all the pages that hold or will hold the directory - * content. We need to fill in any gaps that we might find where the - * memory reclaimer has been at work. If there are any gaps, we will - * need to reread the entire directory contents. + iov_iter_folio_queue(&iter, ITER_DEST, dvnode->directory, 0, 0, dvnode->directory_size); + + /* AFS requires us to perform the read of a directory synchronously as + * a single unit to avoid issues with the directory contents being + * changed between reads. */ - i = 0; - do { - n = find_get_pages_contig(dvnode->vfs_inode.i_mapping, i, - req->nr_pages - i, - req->pages + i); - _debug("find %u at %u/%u", n, i, req->nr_pages); - if (n == 0) { - gfp_t gfp = dvnode->vfs_inode.i_mapping->gfp_mask; - - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); - - ret = -ENOMEM; - req->pages[i] = __page_cache_alloc(gfp); - if (!req->pages[i]) - goto error; - ret = add_to_page_cache_lru(req->pages[i], - dvnode->vfs_inode.i_mapping, - i, gfp); - if (ret < 0) - goto error; - - set_page_private(req->pages[i], 1); - SetPagePrivate(req->pages[i]); - unlock_page(req->pages[i]); - i++; - } else { - i += n; + ret = netfs_read_single(&dvnode->netfs.inode, file, &iter); + if (ret >= 0) { + i_size = i_size_read(&dvnode->netfs.inode); + if (i_size > ret) { + /* The content has grown, so we need to expand the + * buffer. + */ + ret = -ESTALE; + } else if (is_dir) { + int ret2 = afs_dir_check(dvnode); + + if (ret2 < 0) + ret = ret2; + } else if (i_size < folioq_folio_size(dvnode->directory, 0)) { + /* NUL-terminate a symlink. */ + char *symlink = kmap_local_folio(folioq_folio(dvnode->directory, 0), 0); + + symlink[i_size] = 0; + kunmap_local(symlink); } - } while (i < req->nr_pages); + } + + return ret; +} + +ssize_t afs_read_single(struct afs_vnode *dvnode, struct file *file) +{ + ssize_t ret; + + fscache_use_cookie(afs_vnode_cache(dvnode), false); + ret = afs_do_read_single(dvnode, file); + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + return ret; +} + +/* + * Read the directory into a folio_queue buffer in one go, scrubbing the + * previous contents. We return -ESTALE if the caller needs to call us again. + */ +ssize_t afs_read_dir(struct afs_vnode *dvnode, struct file *file) + __acquires(&dvnode->validate_lock) +{ + ssize_t ret; + loff_t i_size; + + i_size = i_size_read(&dvnode->netfs.inode); - /* If we're going to reload, we need to lock all the pages to prevent - * races. - */ ret = -ERESTARTSYS; if (down_read_killable(&dvnode->validate_lock) < 0) goto error; - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - goto success; + /* We only need to reread the data if it became invalid - or if we + * haven't read it yet. + */ + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + ret = i_size; + goto valid; + } up_read(&dvnode->validate_lock); if (down_write_killable(&dvnode->validate_lock) < 0) goto error; - if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { - ret = afs_fetch_data(dvnode, key, req); + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_invalidate_cache(dvnode, 0); + + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) || + !test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + trace_afs_reload_dir(dvnode); + ret = afs_read_single(dvnode, file); if (ret < 0) goto error_unlock; - task_io_account_read(PAGE_SIZE * req->nr_pages); - - if (req->len < req->file_size) - goto content_has_grown; - - /* Validate the data we just read. */ - ret = -EIO; - for (i = 0; i < req->nr_pages; i++) - if (!afs_dir_check_page(dvnode, req->pages[i], - req->actual_len)) - goto error_unlock; - // TODO: Trim excess pages set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + set_bit(AFS_VNODE_DIR_READ, &dvnode->flags); + } else { + ret = i_size; } downgrade_write(&dvnode->validate_lock); -success: - return req; +valid: + return ret; error_unlock: up_write(&dvnode->validate_lock); error: - afs_put_read(req); - _leave(" = %d", ret); - return ERR_PTR(ret); - -content_has_grown: - up_write(&dvnode->validate_lock); - afs_put_read(req); - goto retry; + _leave(" = %zd", ret); + return ret; } /* @@ -320,75 +367,71 @@ content_has_grown: */ static int afs_dir_iterate_block(struct afs_vnode *dvnode, struct dir_context *ctx, - union afs_xdr_dir_block *block, - unsigned blkoff) + union afs_xdr_dir_block *block) { union afs_xdr_dirent *dire; - unsigned offset, next, curr; + unsigned int blknum, base, hdr, pos, next, nr_slots; size_t nlen; int tmp; - _enter("%u,%x,%p,,",(unsigned)ctx->pos,blkoff,block); + blknum = ctx->pos / AFS_DIR_BLOCK_SIZE; + base = blknum * AFS_DIR_SLOTS_PER_BLOCK; + hdr = (blknum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + pos = DIV_ROUND_UP(ctx->pos, AFS_DIR_DIRENT_SIZE) - base; - curr = (ctx->pos - blkoff) / sizeof(union afs_xdr_dirent); + _enter("%llx,%x", ctx->pos, blknum); /* walk through the block, an entry at a time */ - for (offset = (blkoff == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); - offset < AFS_DIR_SLOTS_PER_BLOCK; - offset = next - ) { - next = offset + 1; - + for (unsigned int slot = hdr; slot < AFS_DIR_SLOTS_PER_BLOCK; slot = next) { /* skip entries marked unused in the bitmap */ - if (!(block->hdr.bitmap[offset / 8] & - (1 << (offset % 8)))) { - _debug("ENT[%zu.%u]: unused", - blkoff / sizeof(union afs_xdr_dir_block), offset); - if (offset >= curr) - ctx->pos = blkoff + - next * sizeof(union afs_xdr_dirent); + if (!(block->hdr.bitmap[slot / 8] & + (1 << (slot % 8)))) { + _debug("ENT[%x]: Unused", base + slot); + next = slot + 1; + if (next >= pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; } /* got a valid entry */ - dire = &block->dirents[offset]; + dire = &block->dirents[slot]; nlen = strnlen(dire->u.name, - sizeof(*block) - - offset * sizeof(union afs_xdr_dirent)); + (unsigned long)(block + 1) - (unsigned long)dire->u.name - 1); + if (nlen > AFSNAMEMAX - 1) { + _debug("ENT[%x]: Name too long (len %zx)", + base + slot, nlen); + return afs_bad(dvnode, afs_file_error_dir_name_too_long); + } - _debug("ENT[%zu.%u]: %s %zu \"%s\"", - blkoff / sizeof(union afs_xdr_dir_block), offset, - (offset < curr ? "skip" : "fill"), + _debug("ENT[%x]: %s %zx \"%s\"", + base + slot, (slot < pos ? "skip" : "fill"), nlen, dire->u.name); - /* work out where the next possible entry is */ - for (tmp = nlen; tmp > 15; tmp -= sizeof(union afs_xdr_dirent)) { - if (next >= AFS_DIR_SLOTS_PER_BLOCK) { - _debug("ENT[%zu.%u]:" - " %u travelled beyond end dir block" - " (len %u/%zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, next, tmp, nlen); - return afs_bad(dvnode, afs_file_error_dir_over_end); - } - if (!(block->hdr.bitmap[next / 8] & - (1 << (next % 8)))) { - _debug("ENT[%zu.%u]:" - " %u unmarked extension (len %u/%zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, next, tmp, nlen); + nr_slots = afs_dir_calc_slots(nlen); + next = slot + nr_slots; + if (next > AFS_DIR_SLOTS_PER_BLOCK) { + _debug("ENT[%x]: extends beyond end dir block (len %zx)", + base + slot, nlen); + return afs_bad(dvnode, afs_file_error_dir_over_end); + } + + /* Check that the name-extension dirents are all allocated */ + for (tmp = 1; tmp < nr_slots; tmp++) { + unsigned int xslot = slot + tmp; + + if (!(block->hdr.bitmap[xslot / 8] & (1 << (xslot % 8)))) { + _debug("ENT[%x]: Unmarked extension (%x/%x)", + base + slot, tmp, nr_slots); return afs_bad(dvnode, afs_file_error_dir_unmarked_ext); } - - _debug("ENT[%zu.%u]: ext %u/%zu", - blkoff / sizeof(union afs_xdr_dir_block), - next, tmp, nlen); - next++; } /* skip if starts before the current position */ - if (offset < curr) + if (slot < pos) { + if (next > pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; + } /* found the next entry */ if (!dir_emit(ctx, dire->u.name, nlen, @@ -400,82 +443,110 @@ static int afs_dir_iterate_block(struct afs_vnode *dvnode, return 0; } - ctx->pos = blkoff + next * sizeof(union afs_xdr_dirent); + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); } _leave(" = 1 [more]"); return 1; } +struct afs_dir_iteration_ctx { + struct dir_context *dir_ctx; + int error; +}; + /* - * iterate through the data blob that lists the contents of an AFS directory + * Iterate through a kmapped directory segment. */ -static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, - struct key *key) +static size_t afs_dir_iterate_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_xdr_dir_page *dbuf; - union afs_xdr_dir_block *dblock; - struct afs_read *req; - struct page *page; - unsigned blkoff, limit; + struct afs_dir_iteration_ctx *ctx = priv2; + struct afs_vnode *dvnode = priv; int ret; - _enter("{%lu},%u,,", dir->i_ino, (unsigned)ctx->pos); - - if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { - _leave(" = -ESTALE"); - return -ESTALE; + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) { + pr_err("Mis-iteration prog=%zx len=%zx\n", + progress % AFS_DIR_BLOCK_SIZE, + len % AFS_DIR_BLOCK_SIZE); + return len; } - req = afs_read_dir(dvnode, key); - if (IS_ERR(req)) - return PTR_ERR(req); + do { + ret = afs_dir_iterate_block(dvnode, ctx->dir_ctx, iter_base); + if (ret != 1) + break; - /* round the file position up to the next entry boundary */ - ctx->pos += sizeof(union afs_xdr_dirent) - 1; - ctx->pos &= ~(sizeof(union afs_xdr_dirent) - 1); + ctx->dir_ctx->pos = round_up(ctx->dir_ctx->pos, AFS_DIR_BLOCK_SIZE); + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - /* walk through the blocks in sequence */ - ret = 0; - while (ctx->pos < req->actual_len) { - blkoff = ctx->pos & ~(sizeof(union afs_xdr_dir_block) - 1); + return len; +} - /* Fetch the appropriate page from the directory and re-add it - * to the LRU. - */ - page = req->pages[blkoff / PAGE_SIZE]; - if (!page) { - ret = afs_bad(dvnode, afs_file_error_dir_missing_page); - break; - } - mark_page_accessed(page); +/* + * Iterate through the directory folios. + */ +static int afs_dir_iterate_contents(struct inode *dir, struct dir_context *dir_ctx) +{ + struct afs_dir_iteration_ctx ctx = { .dir_ctx = dir_ctx }; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct iov_iter iter; + unsigned long long i_size = i_size_read(dir); - limit = blkoff & ~(PAGE_SIZE - 1); + /* Round the file position up to the next entry boundary */ + dir_ctx->pos = round_up(dir_ctx->pos, sizeof(union afs_xdr_dirent)); - dbuf = kmap(page); + if (i_size <= 0 || dir_ctx->pos >= i_size) + return 0; - /* deal with the individual blocks stashed on this page */ - do { - dblock = &dbuf->blocks[(blkoff % PAGE_SIZE) / - sizeof(union afs_xdr_dir_block)]; - ret = afs_dir_iterate_block(dvnode, ctx, dblock, blkoff); - if (ret != 1) { - kunmap(page); - goto out; - } + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iov_iter_advance(&iter, round_down(dir_ctx->pos, AFS_DIR_BLOCK_SIZE)); - blkoff += sizeof(union afs_xdr_dir_block); + iterate_folioq(&iter, iov_iter_count(&iter), dvnode, &ctx, + afs_dir_iterate_step); - } while (ctx->pos < dir->i_size && blkoff < limit); + if (ctx.error == -ESTALE) + afs_invalidate_dir(dvnode, afs_dir_invalid_iter_stale); + return ctx.error; +} - kunmap(page); - ret = 0; - } +/* + * iterate through the data blob that lists the contents of an AFS directory + */ +static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, + struct file *file, afs_dataversion_t *_dir_version) +{ + struct afs_vnode *dvnode = AFS_FS_I(dir); + int retry_limit = 100; + int ret; + + _enter("{%lu},%llx,,", dir->i_ino, ctx->pos); + + do { + if (--retry_limit < 0) { + pr_warn("afs_read_dir(): Too many retries\n"); + ret = -ESTALE; + break; + } + ret = afs_read_dir(dvnode, file); + if (ret < 0) { + if (ret != -ESTALE) + break; + if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { + ret = -ESTALE; + break; + } + continue; + } + *_dir_version = inode_peek_iversion_raw(dir); + + ret = afs_dir_iterate_contents(dir, ctx); + up_read(&dvnode->validate_lock); + } while (ret == -ESTALE); -out: - up_read(&dvnode->validate_lock); - afs_put_read(req); _leave(" = %d", ret); return ret; } @@ -485,7 +556,9 @@ out: */ static int afs_readdir(struct file *file, struct dir_context *ctx) { - return afs_dir_iterate(file_inode(file), ctx, afs_file_key(file)); + afs_dataversion_t dir_version; + + return afs_dir_iterate(file_inode(file), ctx, file, &dir_version); } /* @@ -493,7 +566,7 @@ static int afs_readdir(struct file *file, struct dir_context *ctx) * - if afs_dir_iterate_block() spots this function, it'll pass the FID * uniquifier through dtype */ -static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, +static bool afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype) { struct afs_lookup_one_cookie *cookie = @@ -509,43 +582,43 @@ static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, if (cookie->name.len != nlen || memcmp(cookie->name.name, name, nlen) != 0) { - _leave(" = 0 [no]"); - return 0; + _leave(" = true [keep looking]"); + return true; } cookie->fid.vnode = ino; cookie->fid.unique = dtype; cookie->found = 1; - _leave(" = -1 [found]"); - return -1; + _leave(" = false [found]"); + return false; } /* * Do a lookup of a single name in a directory * - just returns the FID the dentry name maps to if found */ -static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, - struct afs_fid *fid, struct key *key) +static int afs_do_lookup_one(struct inode *dir, const struct qstr *name, + struct afs_fid *fid, + afs_dataversion_t *_dir_version) { struct afs_super_info *as = dir->i_sb->s_fs_info; struct afs_lookup_one_cookie cookie = { .ctx.actor = afs_lookup_one_filldir, - .name = dentry->d_name, + .name = *name, .fid.vid = as->volume->vid }; int ret; - _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); + _enter("{%lu},{%.*s},", dir->i_ino, name->len, name->name); /* search the directory */ - ret = afs_dir_iterate(dir, &cookie.ctx, key); + ret = afs_dir_iterate(dir, &cookie.ctx, NULL, _dir_version); if (ret < 0) { _leave(" = %d [iter]", ret); return ret; } - ret = -ENOENT; if (!cookie.found) { _leave(" = -ENOENT [not found]"); return -ENOENT; @@ -561,12 +634,11 @@ static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, * - if afs_dir_iterate_block() spots this function, it'll pass the FID * uniquifier through dtype */ -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, +static bool afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype) { struct afs_lookup_cookie *cookie = container_of(ctx, struct afs_lookup_cookie, ctx); - int ret; _enter("{%s,%u},%s,%u,,%llu,%u", cookie->name.name, cookie->name.len, name, nlen, @@ -576,23 +648,121 @@ static int afs_lookup_filldir(struct dir_context *ctx, const char *name, BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); - if (cookie->found) { - if (cookie->nr_fids < 50) { - cookie->fids[cookie->nr_fids].vnode = ino; - cookie->fids[cookie->nr_fids].unique = dtype; - cookie->nr_fids++; + if (cookie->nr_fids < 50) { + cookie->fids[cookie->nr_fids].vnode = ino; + cookie->fids[cookie->nr_fids].unique = dtype; + cookie->nr_fids++; + } + + return cookie->nr_fids < 50; +} + +/* + * Deal with the result of a successful lookup operation. Turn all the files + * into inodes and save the first one - which is the one we actually want. + */ +static void afs_do_lookup_success(struct afs_operation *op) +{ + struct afs_vnode_param *vp; + struct afs_vnode *vnode; + struct inode *inode; + u32 abort_code; + int i; + + _enter(""); + + for (i = 0; i < op->nr_files; i++) { + switch (i) { + case 0: + vp = &op->file[0]; + abort_code = vp->scb.status.abort_code; + if (abort_code != 0) { + op->call_abort_code = abort_code; + afs_op_set_error(op, afs_abort_to_error(abort_code)); + op->cumul_error.abort_code = abort_code; + } + break; + + case 1: + vp = &op->file[1]; + break; + + default: + vp = &op->more_files[i - 2]; + break; + } + + if (vp->scb.status.abort_code) + trace_afs_bulkstat_error(op, &vp->fid, i, vp->scb.status.abort_code); + if (!vp->scb.have_status && !vp->scb.have_error) + continue; + + _debug("do [%u]", i); + if (vp->vnode) { + if (!test_bit(AFS_VNODE_UNSET, &vp->vnode->flags)) + afs_vnode_commit_status(op, vp); + } else if (vp->scb.status.abort_code == 0) { + inode = afs_iget(op, vp); + if (!IS_ERR(inode)) { + vnode = AFS_FS_I(inode); + afs_cache_permit(vnode, op->key, + 0 /* Assume vnode->cb_break is 0 */ + + op->cb_v_break, + &vp->scb); + vp->vnode = vnode; + vp->put_vnode = true; + } + } else { + _debug("- abort %d %llx:%llx.%x", + vp->scb.status.abort_code, + vp->fid.vid, vp->fid.vnode, vp->fid.unique); } - } else if (cookie->name.len == nlen && - memcmp(cookie->name.name, name, nlen) == 0) { - cookie->fids[0].vnode = ino; - cookie->fids[0].unique = dtype; - cookie->found = 1; - if (cookie->one_only) - return -1; } - ret = cookie->nr_fids >= 50 ? -1 : 0; - _leave(" = %d", ret); + _leave(""); +} + +static const struct afs_operation_ops afs_inline_bulk_status_operation = { + .issue_afs_rpc = afs_fs_inline_bulk_status, + .issue_yfs_rpc = yfs_fs_inline_bulk_status, + .success = afs_do_lookup_success, +}; + +static const struct afs_operation_ops afs_lookup_fetch_status_operation = { + .issue_afs_rpc = afs_fs_fetch_status, + .issue_yfs_rpc = yfs_fs_fetch_status, + .success = afs_do_lookup_success, + .aborted = afs_check_for_remote_deletion, +}; + +/* + * See if we know that the server we expect to use doesn't support + * FS.InlineBulkStatus. + */ +static bool afs_server_supports_ibulk(struct afs_vnode *dvnode) +{ + struct afs_server_list *slist; + struct afs_volume *volume = dvnode->volume; + struct afs_server *server; + bool ret = true; + int i; + + if (!test_bit(AFS_VOLUME_MAYBE_NO_IBULK, &volume->flags)) + return true; + + rcu_read_lock(); + slist = rcu_dereference(volume->servers); + + for (i = 0; i < slist->nr_servers; i++) { + server = slist->servers[i].server; + if (server == dvnode->cb_server) { + if (test_bit(AFS_SERVER_FL_NO_IBULK, &server->flags)) + ret = false; + break; + } + } + + rcu_read_unlock(); return ret; } @@ -601,17 +771,17 @@ static int afs_lookup_filldir(struct dir_context *ctx, const char *name, * files in one go and create inodes for them. The inode of the file we were * asked for is returned. */ -static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry) { struct afs_lookup_cookie *cookie; - struct afs_cb_interest *cbi = NULL; - struct afs_super_info *as = dir->i_sb->s_fs_info; - struct afs_iget_data data; - struct afs_fs_cursor fc; - struct afs_vnode *dvnode = AFS_FS_I(dir); - struct inode *inode = NULL; - int ret, i; + struct afs_vnode_param *vp; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode; + struct inode *inode = NULL, *ti; + afs_dataversion_t data_version = READ_ONCE(dvnode->status.data_version); + bool supports_ibulk, isnew; + long ret; + int i; _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); @@ -619,154 +789,127 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, if (!cookie) return ERR_PTR(-ENOMEM); + for (i = 0; i < ARRAY_SIZE(cookie->fids); i++) + cookie->fids[i].vid = dvnode->fid.vid; cookie->ctx.actor = afs_lookup_filldir; cookie->name = dentry->d_name; - cookie->nr_fids = 1; /* slot 0 is saved for the fid we actually want */ - - read_seqlock_excl(&dvnode->cb_lock); - if (dvnode->cb_interest && - dvnode->cb_interest->server && - test_bit(AFS_SERVER_FL_NO_IBULK, &dvnode->cb_interest->server->flags)) - cookie->one_only = true; - read_sequnlock_excl(&dvnode->cb_lock); - - for (i = 0; i < 50; i++) - cookie->fids[i].vid = as->volume->vid; + cookie->nr_fids = 2; /* slot 1 is saved for the fid we actually want + * and slot 0 for the directory */ - /* search the directory */ - ret = afs_dir_iterate(dir, &cookie->ctx, key); - if (ret < 0) { - inode = ERR_PTR(ret); + /* Search the directory for the named entry using the hash table... */ + ret = afs_dir_search(dvnode, &dentry->d_name, &cookie->fids[1], &data_version); + if (ret < 0) goto out; + + supports_ibulk = afs_server_supports_ibulk(dvnode); + if (supports_ibulk) { + /* ...then scan linearly from that point for entries to lookup-ahead. */ + cookie->ctx.pos = (ret + 1) * AFS_DIR_DIRENT_SIZE; + afs_dir_iterate(dir, &cookie->ctx, NULL, &data_version); } - inode = ERR_PTR(-ENOENT); - if (!cookie->found) - goto out; + dentry->d_fsdata = (void *)(unsigned long)data_version; /* Check to see if we already have an inode for the primary fid. */ - data.volume = dvnode->volume; - data.fid = cookie->fids[0]; - inode = ilookup5(dir->i_sb, cookie->fids[0].vnode, afs_iget5_test, &data); + inode = ilookup5(dir->i_sb, cookie->fids[1].vnode, + afs_ilookup5_test_by_fid, &cookie->fids[1]); if (inode) + goto out; /* We do */ + + /* Okay, we didn't find it. We need to query the server - and whilst + * we're doing that, we're going to attempt to look up a bunch of other + * vnodes also. + */ + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto out; + } + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_fid(op, 1, &cookie->fids[1]); + + op->nr_files = cookie->nr_fids; + _debug("nr_files %u", op->nr_files); /* Need space for examining all the selected files */ - inode = ERR_PTR(-ENOMEM); - cookie->statuses = kcalloc(cookie->nr_fids, sizeof(struct afs_file_status), - GFP_KERNEL); - if (!cookie->statuses) - goto out; + if (op->nr_files > 2) { + op->more_files = kvcalloc(op->nr_files - 2, + sizeof(struct afs_vnode_param), + GFP_KERNEL); + if (!op->more_files) { + afs_op_nomem(op); + goto out_op; + } - cookie->callbacks = kcalloc(cookie->nr_fids, sizeof(struct afs_callback), - GFP_KERNEL); - if (!cookie->callbacks) - goto out_s; + for (i = 2; i < op->nr_files; i++) { + vp = &op->more_files[i - 2]; + vp->fid = cookie->fids[i]; + + /* Find any inodes that already exist and get their + * callback counters. + */ + ti = ilookup5_nowait(dir->i_sb, vp->fid.vnode, + afs_ilookup5_test_by_fid, &vp->fid, &isnew); + if (!IS_ERR_OR_NULL(ti)) { + vnode = AFS_FS_I(ti); + vp->dv_before = vnode->status.data_version; + vp->cb_break_before = afs_calc_vnode_cb_break(vnode); + vp->vnode = vnode; + vp->put_vnode = true; + vp->speculative = true; /* vnode not locked */ + } + } + } /* Try FS.InlineBulkStatus first. Abort codes for the individual * lookups contained therein are stored in the reply without aborting * the whole operation. */ - if (cookie->one_only) - goto no_inline_bulk_status; - - inode = ERR_PTR(-ERESTARTSYS); - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - if (test_bit(AFS_SERVER_FL_NO_IBULK, - &fc.cbi->server->flags)) { - fc.ac.abort_code = RX_INVALID_OPERATION; - fc.ac.error = -ECONNABORTED; - break; - } - afs_fs_inline_bulk_status(&fc, - afs_v2net(dvnode), - cookie->fids, - cookie->statuses, - cookie->callbacks, - cookie->nr_fids, NULL); - } - - if (fc.ac.error == 0) - cbi = afs_get_cb_interest(fc.cbi); - if (fc.ac.abort_code == RX_INVALID_OPERATION) - set_bit(AFS_SERVER_FL_NO_IBULK, &fc.cbi->server->flags); - inode = ERR_PTR(afs_end_vnode_operation(&fc)); + afs_op_set_error(op, -ENOTSUPP); + if (supports_ibulk) { + op->ops = &afs_inline_bulk_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } - if (!IS_ERR(inode)) - goto success; - if (fc.ac.abort_code != RX_INVALID_OPERATION) - goto out_c; - -no_inline_bulk_status: - /* We could try FS.BulkStatus next, but this aborts the entire op if - * any of the lookups fails - so, for the moment, revert to - * FS.FetchStatus for just the primary fid. - */ - cookie->nr_fids = 1; - inode = ERR_PTR(-ERESTARTSYS); - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - afs_fs_fetch_status(&fc, - afs_v2net(dvnode), - cookie->fids, - cookie->statuses, - cookie->callbacks, - NULL); - } - - if (fc.ac.error == 0) - cbi = afs_get_cb_interest(fc.cbi); - inode = ERR_PTR(afs_end_vnode_operation(&fc)); + if (afs_op_error(op) == -ENOTSUPP) { + /* We could try FS.BulkStatus next, but this aborts the entire + * op if any of the lookups fails - so, for the moment, revert + * to FS.FetchStatus for op->file[1]. + */ + op->fetch_status.which = 1; + op->ops = &afs_lookup_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } - if (IS_ERR(inode)) - goto out_c; - - for (i = 0; i < cookie->nr_fids; i++) - cookie->statuses[i].abort_code = 0; - -success: - /* Turn all the files into inodes and save the first one - which is the - * one we actually want. - */ - if (cookie->statuses[0].abort_code != 0) - inode = ERR_PTR(afs_abort_to_error(cookie->statuses[0].abort_code)); - - for (i = 0; i < cookie->nr_fids; i++) { - struct inode *ti; - - if (cookie->statuses[i].abort_code != 0) - continue; - - ti = afs_iget(dir->i_sb, key, &cookie->fids[i], - &cookie->statuses[i], - &cookie->callbacks[i], - cbi); - if (i == 0) { - inode = ti; +out_op: + if (!afs_op_error(op)) { + if (op->file[1].scb.status.abort_code) { + afs_op_accumulate_error(op, -ECONNABORTED, + op->file[1].scb.status.abort_code); } else { - if (!IS_ERR(ti)) - iput(ti); + inode = &op->file[1].vnode->netfs.inode; + op->file[1].vnode = NULL; } } -out_c: - afs_put_cb_interest(afs_v2net(dvnode), cbi); - kfree(cookie->callbacks); -out_s: - kfree(cookie->statuses); + if (op->file[0].scb.have_status) + dentry->d_fsdata = (void *)(unsigned long)op->file[0].scb.status.data_version; + else + dentry->d_fsdata = (void *)(unsigned long)op->file[0].dv_before; + ret = afs_put_operation(op); out: kfree(cookie); - return inode; + _leave(""); + return inode ?: ERR_PTR(ret); } /* * Look up an entry in a directory with @sys substitution. */ -static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry) { struct afs_sysnames *subs; struct afs_net *net = afs_i2net(dir); @@ -800,7 +943,7 @@ static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, } strcpy(p, name); - ret = lookup_one_len(buf, dentry->d_parent, len); + ret = lookup_noperm(&QSTR(buf), dentry->d_parent); if (IS_ERR(ret) || d_is_positive(ret)) goto out_s; dput(ret); @@ -814,7 +957,6 @@ out_s: afs_put_sysnames(subs); kfree(buf); out_p: - key_put(key); return ret; } @@ -825,9 +967,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_fid fid = {}; struct inode *inode; struct dentry *d; - struct key *key; int ret; _enter("{%llx:%llu},%p{%pd},", @@ -845,15 +987,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ESTALE); } - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - _leave(" = %ld [key]", PTR_ERR(key)); - return ERR_CAST(key); - } - - ret = afs_validate(dvnode, key); + ret = afs_validate(dvnode, NULL); if (ret < 0) { - key_put(key); + afs_dir_unuse_cookie(dvnode, ret); _leave(" = %d [val]", ret); return ERR_PTR(ret); } @@ -863,40 +999,76 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, dentry->d_name.name[dentry->d_name.len - 3] == 's' && dentry->d_name.name[dentry->d_name.len - 2] == 'y' && dentry->d_name.name[dentry->d_name.len - 1] == 's') - return afs_lookup_atsys(dir, dentry, key); + return afs_lookup_atsys(dir, dentry); afs_stat_v(dvnode, n_lookup); - inode = afs_do_lookup(dir, dentry, key); - key_put(key); - if (inode == ERR_PTR(-ENOENT)) { - inode = afs_try_auto_mntpt(dentry, dir); - } else { - dentry->d_fsdata = - (void *)(unsigned long)dvnode->status.data_version; - } + inode = afs_do_lookup(dir, dentry); + if (inode == ERR_PTR(-ENOENT)) + inode = NULL; + else if (!IS_ERR_OR_NULL(inode)) + fid = AFS_FS_I(inode)->fid; + + _debug("splice %p", dentry->d_inode); d = d_splice_alias(inode, dentry); - if (!IS_ERR_OR_NULL(d)) + if (!IS_ERR_OR_NULL(d)) { d->d_fsdata = dentry->d_fsdata; + trace_afs_lookup(dvnode, &d->d_name, &fid); + } else { + trace_afs_lookup(dvnode, &dentry->d_name, &fid); + } + _leave(""); return d; } /* + * Check the validity of a dentry under RCU conditions. + */ +static int afs_d_revalidate_rcu(struct afs_vnode *dvnode, struct dentry *dentry) +{ + long dir_version, de_version; + + _enter("%p", dentry); + + if (test_bit(AFS_VNODE_DELETED, &dvnode->flags)) + return -ECHILD; + + if (!afs_check_validity(dvnode)) + return -ECHILD; + + /* We only need to invalidate a dentry if the server's copy changed + * behind our back. If we made the change, it's no problem. Note that + * on a 32-bit system, we only have 32 bits in the dentry to store the + * version. + */ + dir_version = (long)READ_ONCE(dvnode->status.data_version); + de_version = (long)READ_ONCE(dentry->d_fsdata); + if (de_version != dir_version) { + dir_version = (long)READ_ONCE(dvnode->invalid_before); + if (de_version - dir_version < 0) + return -ECHILD; + } + + return 1; /* Still valid */ +} + +/* * check that a dentry lookup hit has found a valid entry * - NOTE! the hit can be a negative hit too, so we can't assume we have an * inode */ -static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) +static int afs_d_revalidate(struct inode *parent_dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) { - struct afs_vnode *vnode, *dir; - struct afs_fid uninitialized_var(fid); - struct dentry *parent; + struct afs_vnode *vnode, *dir = AFS_FS_I(parent_dir); + struct afs_fid fid; struct inode *inode; struct key *key; - long dir_version, de_version; + afs_dataversion_t dir_version, invalid_before; + long de_version; int ret; if (flags & LOOKUP_RCU) - return -ECHILD; + return afs_d_revalidate_rcu(dir, dentry); if (d_really_is_positive(dentry)) { vnode = AFS_FS_I(d_inode(dentry)); @@ -911,26 +1083,16 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) if (IS_ERR(key)) key = NULL; - if (d_really_is_positive(dentry)) { - inode = d_inode(dentry); - if (inode) { - vnode = AFS_FS_I(inode); - afs_validate(vnode, key); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - goto out_bad; - } - } - - /* lock down the parent dentry so we can peer at it */ - parent = dget_parent(dentry); - dir = AFS_FS_I(d_inode(parent)); - /* validate the parent directory */ - afs_validate(dir, key); + ret = afs_validate(dir, key); + if (ret == -ERESTARTSYS) { + key_put(key); + return ret; + } if (test_bit(AFS_VNODE_DELETED, &dir->flags)) { _debug("%pd: parent dir deleted", dentry); - goto out_bad_parent; + goto not_found; } /* We only need to invalidate a dentry if the server's copy changed @@ -938,30 +1100,30 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) * on a 32-bit system, we only have 32 bits in the dentry to store the * version. */ - dir_version = (long)dir->status.data_version; + dir_version = dir->status.data_version; de_version = (long)dentry->d_fsdata; - if (de_version == dir_version) - goto out_valid; + if (de_version == (long)dir_version) + goto out_valid_noupdate; - dir_version = (long)dir->invalid_before; - if (de_version - dir_version >= 0) + invalid_before = dir->invalid_before; + if (de_version - (long)invalid_before >= 0) goto out_valid; _debug("dir modified"); afs_stat_v(dir, n_reval); /* search the directory for this vnode */ - ret = afs_do_lookup_one(&dir->vfs_inode, dentry, &fid, key); + ret = afs_do_lookup_one(&dir->netfs.inode, name, &fid, &dir_version); switch (ret) { case 0: /* the filename maps to something */ if (d_really_is_negative(dentry)) - goto out_bad_parent; + goto not_found; inode = d_inode(dentry); if (is_bad_inode(inode)) { printk("kAFS: afs_d_revalidate: %pd2 has bad inode\n", dentry); - goto out_bad_parent; + goto not_found; } vnode = AFS_FS_I(inode); @@ -982,10 +1144,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) _debug("%pd: file deleted (uq %u -> %u I:%u)", dentry, fid.unique, vnode->fid.unique, - vnode->vfs_inode.i_generation); - write_seqlock(&vnode->cb_lock); - set_bit(AFS_VNODE_DELETED, &vnode->flags); - write_sequnlock(&vnode->cb_lock); + vnode->netfs.inode.i_generation); goto not_found; } goto out_valid; @@ -998,28 +1157,19 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) goto out_valid; default: - _debug("failed to iterate dir %pd: %d", - parent, ret); - goto out_bad_parent; + _debug("failed to iterate parent %pd2: %d", dentry, ret); + goto not_found; } out_valid: - dentry->d_fsdata = (void *)dir_version; - dput(parent); + dentry->d_fsdata = (void *)(unsigned long)dir_version; +out_valid_noupdate: key_put(key); _leave(" = 1 [valid]"); return 1; - /* the dirent, if it exists, now points to a different vnode */ not_found: - spin_lock(&dentry->d_lock); - dentry->d_flags |= DCACHE_NFSFS_RENAMED; - spin_unlock(&dentry->d_lock); - -out_bad_parent: _debug("dropping dentry %pd2", dentry); - dput(parent); -out_bad: key_put(key); _leave(" = 0 [bad]"); @@ -1053,6 +1203,16 @@ zap: } /* + * Clean up sillyrename files on dentry removal. + */ +static void afs_d_iput(struct dentry *dentry, struct inode *inode) +{ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + afs_silly_iput(dentry, inode); + iput(inode); +} + +/* * handle dentry release */ void afs_d_release(struct dentry *dentry) @@ -1060,95 +1220,130 @@ void afs_d_release(struct dentry *dentry) _enter("%pd", dentry); } +void afs_check_for_remote_deletion(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->file[0].vnode; + + switch (afs_op_abort_code(op)) { + case VNOVNODE: + set_bit(AFS_VNODE_DELETED, &vnode->flags); + clear_nlink(&vnode->netfs.inode); + afs_break_callback(vnode, afs_cb_break_for_deleted); + } +} + /* * Create a new inode for create/mkdir/symlink */ -static void afs_vnode_new_inode(struct afs_fs_cursor *fc, - struct dentry *new_dentry, - struct afs_fid *newfid, - struct afs_file_status *newstatus, - struct afs_callback *newcb) +static void afs_vnode_new_inode(struct afs_operation *op) { + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_vnode *vnode; struct inode *inode; - if (fc->ac.error < 0) - return; + _enter(""); - inode = afs_iget(fc->vnode->vfs_inode.i_sb, fc->key, - newfid, newstatus, newcb, fc->cbi); + ASSERTCMP(afs_op_error(op), ==, 0); + + inode = afs_iget(op, vp); if (IS_ERR(inode)) { /* ENOMEM or EINTR at a really inconvenient time - just abandon * the new directory on the server. */ - fc->ac.error = PTR_ERR(inode); + afs_op_accumulate_error(op, PTR_ERR(inode), 0); return; } vnode = AFS_FS_I(inode); set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); - afs_vnode_commit_status(fc, vnode, 0); - d_instantiate(new_dentry, inode); + if (S_ISDIR(inode->i_mode)) + afs_mkdir_init_dir(vnode, dvp->vnode); + else if (S_ISLNK(inode->i_mode)) + afs_init_new_symlink(vnode, op); + if (!afs_op_error(op)) + afs_cache_permit(vnode, op->key, vnode->cb_break, &vp->scb); + d_instantiate(op->dentry, inode); +} + +static void afs_create_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[0]); + afs_update_dentry_version(op, &op->file[0], op->dentry); + afs_vnode_new_inode(op); } +static void afs_create_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_add(dvnode, &op->dentry->d_name, &vp->fid, + op->create.reason); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_create_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + + if (afs_op_error(op)) + d_drop(op->dentry); +} + +static const struct afs_operation_ops afs_mkdir_operation = { + .issue_afs_rpc = afs_fs_make_dir, + .issue_yfs_rpc = yfs_fs_make_dir, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a directory on an AFS filesystem */ -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static struct dentry *afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode) { - struct afs_file_status newstatus; - struct afs_fs_cursor fc; - struct afs_callback newcb; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; - mode |= S_IFDIR; - _enter("{%llx:%llu},{%pd},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry, mode); - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + d_drop(dentry); + return ERR_CAST(op); } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_create(&fc, dentry->d_name.name, mode, data_version, - &newfid, &newstatus, &newcb); - } - - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } - - if (ret == 0 && - test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_create); - - key_put(key); - _leave(" = 0"); - return 0; - -error_key: - key_put(key); -error: - d_drop(dentry); - _leave(" = %d", ret); - return ret; + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + op->dentry = dentry; + op->create.mode = S_IFDIR | mode; + op->create.reason = afs_edit_dir_for_mkdir; + op->mtime = current_time(dir); + op->ops = &afs_mkdir_operation; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ERR_PTR(ret); } /* @@ -1159,63 +1354,109 @@ static void afs_dir_remove_subdir(struct dentry *dentry) if (d_really_is_positive(dentry)) { struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - clear_nlink(&vnode->vfs_inode); + clear_nlink(&vnode->netfs.inode); set_bit(AFS_VNODE_DELETED, &vnode->flags); - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_rmdir); + afs_invalidate_dir(vnode, afs_dir_invalid_subdir_removed); } } +static void afs_rmdir_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[0]); + afs_update_dentry_version(op, &op->file[0], op->dentry); +} + +static void afs_rmdir_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + afs_dir_remove_subdir(op->dentry); + + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_remove(dvnode, &op->dentry->d_name, + afs_edit_dir_for_rmdir); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_rmdir_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->file[1].vnode) + up_write(&op->file[1].vnode->rmdir_lock); +} + +static const struct afs_operation_ops afs_rmdir_operation = { + .issue_afs_rpc = afs_fs_remove_dir, + .issue_yfs_rpc = yfs_fs_remove_dir, + .success = afs_rmdir_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_rmdir_edit_dir, + .put = afs_rmdir_put, +}; + /* * remove a directory from an AFS filesystem */ static int afs_rmdir(struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; _enter("{%llx:%llu},{%pd}", dvnode->fid.vid, dvnode->fid.vnode, dentry); - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + + op->dentry = dentry; + op->ops = &afs_rmdir_operation; /* Try to make sure we have a callback promise on the victim. */ if (d_really_is_positive(dentry)) { vnode = AFS_FS_I(d_inode(dentry)); - ret = afs_validate(vnode, key); + ret = afs_validate(vnode, op->key); if (ret < 0) - goto error_key; + goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_remove(&fc, vnode, dentry->d_name.name, true, - data_version); - } - - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - if (ret == 0) { - afs_dir_remove_subdir(dentry); - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_remove(dvnode, &dentry->d_name, - afs_edit_dir_for_rmdir); - } + if (vnode) { + ret = down_write_killable(&vnode->rmdir_lock); + if (ret < 0) + goto error; + op->file[1].vnode = vnode; } -error_key: - key_put(key); -error: + ret = afs_do_sync_operation(op); + + /* Not all systems that can host afs servers have ENOTEMPTY. */ + if (ret == -EEXIST) + ret = -ENOTEMPTY; +out: + afs_dir_unuse_cookie(dvnode, ret); return ret; + +error: + ret = afs_put_operation(op); + goto out; } /* @@ -1228,56 +1469,95 @@ error: * However, if we didn't have a callback promise outstanding, or it was * outstanding on a different server, then it won't break it either... */ -static int afs_dir_remove_link(struct dentry *dentry, struct key *key, - unsigned long d_version_before, - unsigned long d_version_after) +static void afs_dir_remove_link(struct afs_operation *op) { - bool dir_valid; - int ret = 0; - - /* There were no intervening changes on the server if the version - * number we got back was incremented by exactly 1. - */ - dir_valid = (d_version_after == d_version_before + 1); + struct afs_vnode *dvnode = op->file[0].vnode; + struct afs_vnode *vnode = op->file[1].vnode; + struct dentry *dentry = op->dentry; + int ret; - if (d_really_is_positive(dentry)) { - struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); + if (afs_op_error(op) || + (op->file[1].scb.have_status && op->file[1].scb.have_error)) + return; + if (d_really_is_positive(dentry)) + return; - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { - /* Already done */ - } else if (dir_valid) { - drop_nlink(&vnode->vfs_inode); - if (vnode->vfs_inode.i_nlink == 0) { - set_bit(AFS_VNODE_DELETED, &vnode->flags); - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - } - ret = 0; - } else { - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { + /* Already done */ + } else if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { + write_seqlock(&vnode->cb_lock); + drop_nlink(&vnode->netfs.inode); + if (vnode->netfs.inode.i_nlink == 0) { + set_bit(AFS_VNODE_DELETED, &vnode->flags); + __afs_break_callback(vnode, afs_cb_break_for_unlink); + } + write_sequnlock(&vnode->cb_lock); + } else { + afs_break_callback(vnode, afs_cb_break_for_unlink); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - kdebug("AFS_VNODE_DELETED"); + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) + _debug("AFS_VNODE_DELETED"); - ret = afs_validate(vnode, key); - if (ret == -ESTALE) - ret = 0; - } - _debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, ret); + ret = afs_validate(vnode, op->key); + if (ret != -ESTALE) + afs_op_set_error(op, ret); } - return ret; + _debug("nlink %d [val %d]", vnode->netfs.inode.i_nlink, afs_op_error(op)); +} + +static void afs_unlink_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_check_dir_conflict(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[1]); + afs_update_dentry_version(op, &op->file[0], op->dentry); + afs_dir_remove_link(op); } +static void afs_unlink_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_remove(dvnode, &op->dentry->d_name, + afs_edit_dir_for_unlink); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_unlink_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->unlink.need_rehash && afs_op_error(op) < 0 && afs_op_error(op) != -ENOENT) + d_rehash(op->dentry); +} + +static const struct afs_operation_ops afs_unlink_operation = { + .issue_afs_rpc = afs_fs_remove_file, + .issue_yfs_rpc = yfs_fs_remove_file, + .success = afs_unlink_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_unlink_edit_dir, + .put = afs_unlink_put, +}; + /* * Remove a file or symlink from an AFS filesystem. */ static int afs_unlink(struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; - struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL; - struct key *key; - unsigned long d_version = (unsigned long)dentry->d_fsdata; - u64 data_version = dvnode->status.data_version; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); int ret; _enter("{%llx:%llu},{%pd}", @@ -1286,207 +1566,221 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) if (dentry->d_name.len >= AFSNAMEMAX) return -ENAMETOOLONG; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; /* Try to make sure we have a callback promise on the victim. */ - if (d_really_is_positive(dentry)) { - vnode = AFS_FS_I(d_inode(dentry)); - ret = afs_validate(vnode, key); - if (ret < 0) - goto error_key; + ret = afs_validate(vnode, op->key); + if (ret < 0) { + afs_op_set_error(op, ret); + goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc.cbi->server->flags) && - !test_bit(AFS_SERVER_FL_NO_RM2, &fc.cbi->server->flags)) { - yfs_fs_remove_file2(&fc, vnode, dentry->d_name.name, - data_version); - if (fc.ac.error != -ECONNABORTED || - fc.ac.abort_code != RXGEN_OPCODE) - continue; - set_bit(AFS_SERVER_FL_NO_RM2, &fc.cbi->server->flags); - } + spin_lock(&dentry->d_lock); + if (d_count(dentry) > 1) { + spin_unlock(&dentry->d_lock); + /* Start asynchronous writeout of the inode */ + write_inode_now(d_inode(dentry), 0); + afs_op_set_error(op, afs_sillyrename(dvnode, vnode, dentry, op->key)); + goto error; + } + if (!d_unhashed(dentry)) { + /* Prevent a race with RCU lookup. */ + __d_drop(dentry); + op->unlink.need_rehash = true; + } + spin_unlock(&dentry->d_lock); - afs_fs_remove(&fc, vnode, dentry->d_name.name, false, - data_version); - } + op->file[1].vnode = vnode; + op->file[1].update_ctime = true; + op->file[1].op_unlinked = true; + op->dentry = dentry; + op->ops = &afs_unlink_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - if (ret == 0) - ret = afs_dir_remove_link( - dentry, key, d_version, - (unsigned long)dvnode->status.data_version); - if (ret == 0 && - test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_remove(dvnode, &dentry->d_name, - afs_edit_dir_for_unlink); + /* If there was a conflict with a third party, check the status of the + * unlinked vnode. + */ + if (afs_op_error(op) == 0 && (op->flags & AFS_OPERATION_DIR_CONFLICT)) { + op->file[1].update_ctime = false; + op->fetch_status.which = 1; + op->ops = &afs_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } -error_key: - key_put(key); error: - _leave(" = %d", ret); + ret = afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); return ret; } +static const struct afs_operation_ops afs_create_operation = { + .issue_afs_rpc = afs_fs_create_file, + .issue_yfs_rpc = yfs_fs_create_file, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a regular file on an AFS filesystem */ -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int afs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - struct afs_fs_cursor fc; - struct afs_file_status newstatus; - struct afs_callback newcb; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; - int ret; - - mode |= S_IFREG; + int ret = -ENAMETOOLONG; - _enter("{%llx:%llu},{%pd},%ho,", + _enter("{%llx:%llu},{%pd},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry, mode); - ret = -ENAMETOOLONG; if (dentry->d_name.len >= AFSNAMEMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_create(&fc, dentry->d_name.name, mode, data_version, - &newfid, &newstatus, &newcb); - } + fscache_use_cookie(afs_vnode_cache(dvnode), true); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_create); - - key_put(key); - _leave(" = 0"); - return 0; + op->dentry = dentry; + op->create.mode = S_IFREG | mode; + op->create.reason = afs_edit_dir_for_create; + op->mtime = current_time(dir); + op->ops = &afs_create_operation; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static void afs_link_success(struct afs_operation *op) +{ + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; + + _enter("op=%08x", op->debug_id); + op->ctime = dvp->scb.status.mtime_client; + afs_vnode_commit_status(op, dvp); + afs_vnode_commit_status(op, vp); + afs_update_dentry_version(op, dvp, op->dentry); + if (op->dentry_2->d_parent == op->dentry->d_parent) + afs_update_dentry_version(op, dvp, op->dentry_2); + ihold(&vp->vnode->netfs.inode); + d_instantiate(op->dentry, &vp->vnode->netfs.inode); +} + +static void afs_link_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (afs_op_error(op)) + d_drop(op->dentry); +} + +static const struct afs_operation_ops afs_link_operation = { + .issue_afs_rpc = afs_fs_link, + .issue_yfs_rpc = yfs_fs_link, + .success = afs_link_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_link_put, +}; + /* * create a hard link between files in an AFS filesystem */ static int afs_link(struct dentry *from, struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; - struct afs_vnode *dvnode, *vnode; - struct key *key; - u64 data_version; - int ret; - - vnode = AFS_FS_I(d_inode(from)); - dvnode = AFS_FS_I(dir); - data_version = dvnode->status.data_version; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_vnode *vnode = AFS_FS_I(d_inode(from)); + int ret = -ENAMETOOLONG; _enter("{%llx:%llu},{%llx:%llu},{%pd}", vnode->fid.vid, vnode->fid.vnode, dvnode->fid.vid, dvnode->fid.vnode, dentry); - ret = -ENAMETOOLONG; if (dentry->d_name.len >= AFSNAMEMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) { - afs_end_vnode_operation(&fc); - goto error_key; - } - - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - fc.cb_break_2 = afs_calc_vnode_cb_break(vnode); - afs_fs_link(&fc, vnode, dentry->d_name.name, data_version); - } - - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_commit_status(&fc, vnode, fc.cb_break_2); - ihold(&vnode->vfs_inode); - d_instantiate(dentry, &vnode->vfs_inode); - - mutex_unlock(&vnode->io_lock); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } - - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid, - afs_edit_dir_for_link); - - key_put(key); - _leave(" = 0"); - return 0; + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + ret = afs_validate(vnode, op->key); + if (ret < 0) + goto error_op; + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_vnode(op, 1, vnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + op->file[1].update_ctime = true; + + op->dentry = dentry; + op->dentry_2 = from; + op->ops = &afs_link_operation; + op->create.reason = afs_edit_dir_for_link; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); +error_op: + afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static const struct afs_operation_ops afs_symlink_operation = { + .issue_afs_rpc = afs_fs_symlink, + .issue_yfs_rpc = yfs_fs_symlink, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a symlink in an AFS filesystem */ -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content) +static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *content) { - struct afs_fs_cursor fc; - struct afs_file_status newstatus; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; _enter("{%llx:%llu},{%pd},%s", @@ -1501,69 +1795,285 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, if (strlen(content) >= AFSPATHMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_symlink(&fc, dentry->d_name.name, - content, data_version, - &newfid, &newstatus); - } - - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, NULL); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } + fscache_use_cookie(afs_vnode_cache(dvnode), true); - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_symlink); + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; - key_put(key); - _leave(" = 0"); - return 0; + op->dentry = dentry; + op->ops = &afs_symlink_operation; + op->create.reason = afs_edit_dir_for_symlink; + op->create.symlink = content; + op->mtime = current_time(dir); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static void afs_rename_success(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->more_files[0].vnode; + struct afs_vnode *new_vnode = op->more_files[1].vnode; + + _enter("op=%08x", op->debug_id); + + op->ctime = op->file[0].scb.status.mtime_client; + afs_check_dir_conflict(op, &op->file[1]); + afs_vnode_commit_status(op, &op->file[0]); + if (op->file[1].vnode != op->file[0].vnode) { + op->ctime = op->file[1].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[1]); + } + if (op->more_files[0].scb.have_status) + afs_vnode_commit_status(op, &op->more_files[0]); + if (op->more_files[1].scb.have_status) + afs_vnode_commit_status(op, &op->more_files[1]); + + /* If we're moving a subdir between dirs, we need to update + * its DV counter too as the ".." will be altered. + */ + if (op->file[0].vnode != op->file[1].vnode) { + if (S_ISDIR(vnode->netfs.inode.i_mode)) { + u64 new_dv; + + write_seqlock(&vnode->cb_lock); + + new_dv = vnode->status.data_version + 1; + trace_afs_set_dv(vnode, new_dv); + vnode->status.data_version = new_dv; + inode_set_iversion_raw(&vnode->netfs.inode, new_dv); + + write_sequnlock(&vnode->cb_lock); + } + + if ((op->rename.rename_flags & RENAME_EXCHANGE) && + S_ISDIR(new_vnode->netfs.inode.i_mode)) { + u64 new_dv; + + write_seqlock(&new_vnode->cb_lock); + + new_dv = new_vnode->status.data_version + 1; + new_vnode->status.data_version = new_dv; + inode_set_iversion_raw(&new_vnode->netfs.inode, new_dv); + + write_sequnlock(&new_vnode->cb_lock); + } + } +} + +static void afs_rename_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources orig_cres = {}, new_cres = {}; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode *orig_dvnode = orig_dvp->vnode; + struct afs_vnode *new_dvnode = new_dvp->vnode; + struct afs_vnode *vnode = AFS_FS_I(d_inode(op->dentry)); + struct dentry *old_dentry = op->dentry; + struct dentry *new_dentry = op->dentry_2; + struct inode *new_inode; + + _enter("op=%08x", op->debug_id); + + if (op->rename.rehash) { + d_rehash(op->rename.rehash); + op->rename.rehash = NULL; + } + + fscache_begin_write_operation(&orig_cres, afs_vnode_cache(orig_dvnode)); + if (new_dvnode != orig_dvnode) + fscache_begin_write_operation(&new_cres, afs_vnode_cache(new_dvnode)); + + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) + afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name, + afs_edit_dir_for_rename_0); + + if (new_dvnode != orig_dvnode) { + up_write(&orig_dvnode->validate_lock); + down_write(&new_dvnode->validate_lock); + } + + if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) && + new_dvnode->status.data_version == new_dvp->dv_before + new_dvp->dv_delta) { + if (!op->rename.new_negative) + afs_edit_dir_remove(new_dvnode, &new_dentry->d_name, + afs_edit_dir_for_rename_1); + + afs_edit_dir_add(new_dvnode, &new_dentry->d_name, + &vnode->fid, afs_edit_dir_for_rename_2); + } + + if (S_ISDIR(vnode->netfs.inode.i_mode) && + new_dvnode != orig_dvnode && + test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + afs_edit_dir_update(vnode, &dotdot_name, new_dvnode, + afs_edit_dir_for_rename_sub); + + new_inode = d_inode(new_dentry); + if (new_inode) { + spin_lock(&new_inode->i_lock); + if (S_ISDIR(new_inode->i_mode)) + clear_nlink(new_inode); + else if (new_inode->i_nlink > 0) + drop_nlink(new_inode); + spin_unlock(&new_inode->i_lock); + } + + /* Now we can update d_fsdata on the dentries to reflect their + * new parent's data_version. + */ + afs_update_dentry_version(op, new_dvp, op->dentry); + afs_update_dentry_version(op, new_dvp, op->dentry_2); + + d_move(old_dentry, new_dentry); + + up_write(&new_dvnode->validate_lock); + fscache_end_operation(&orig_cres); + if (new_dvnode != orig_dvnode) + fscache_end_operation(&new_cres); +} + +static void afs_rename_exchange_edit_dir(struct afs_operation *op) +{ + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode *orig_dvnode = orig_dvp->vnode; + struct afs_vnode *new_dvnode = new_dvp->vnode; + struct afs_vnode *old_vnode = op->more_files[0].vnode; + struct afs_vnode *new_vnode = op->more_files[1].vnode; + struct dentry *old_dentry = op->dentry; + struct dentry *new_dentry = op->dentry_2; + + _enter("op=%08x", op->debug_id); + + if (new_dvnode == orig_dvnode) { + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) { + afs_edit_dir_update(orig_dvnode, &old_dentry->d_name, + new_vnode, afs_edit_dir_for_rename_0); + afs_edit_dir_update(orig_dvnode, &new_dentry->d_name, + old_vnode, afs_edit_dir_for_rename_1); + } + + d_exchange(old_dentry, new_dentry); + up_write(&orig_dvnode->validate_lock); + } else { + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) + afs_edit_dir_update(orig_dvnode, &old_dentry->d_name, + new_vnode, afs_edit_dir_for_rename_0); + + up_write(&orig_dvnode->validate_lock); + down_write(&new_dvnode->validate_lock); + + if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) && + new_dvnode->status.data_version == new_dvp->dv_before + new_dvp->dv_delta) + afs_edit_dir_update(new_dvnode, &new_dentry->d_name, + old_vnode, afs_edit_dir_for_rename_1); + + if (S_ISDIR(old_vnode->netfs.inode.i_mode) && + test_bit(AFS_VNODE_DIR_VALID, &old_vnode->flags)) + afs_edit_dir_update(old_vnode, &dotdot_name, new_dvnode, + afs_edit_dir_for_rename_sub); + + if (S_ISDIR(new_vnode->netfs.inode.i_mode) && + test_bit(AFS_VNODE_DIR_VALID, &new_vnode->flags)) + afs_edit_dir_update(new_vnode, &dotdot_name, orig_dvnode, + afs_edit_dir_for_rename_sub); + + /* Now we can update d_fsdata on the dentries to reflect their + * new parents' data_version. + */ + afs_update_dentry_version(op, new_dvp, old_dentry); + afs_update_dentry_version(op, orig_dvp, new_dentry); + + d_exchange(old_dentry, new_dentry); + up_write(&new_dvnode->validate_lock); + } +} + +static void afs_rename_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->rename.rehash) + d_rehash(op->rename.rehash); + dput(op->rename.tmp); + if (afs_op_error(op)) + d_rehash(op->dentry); +} + +static const struct afs_operation_ops afs_rename_operation = { + .issue_afs_rpc = afs_fs_rename, + .issue_yfs_rpc = yfs_fs_rename, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; + +#if 0 /* Autoswitched in yfs_fs_rename_replace(). */ +static const struct afs_operation_ops afs_rename_replace_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_replace, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; +#endif + +static const struct afs_operation_ops afs_rename_noreplace_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_noreplace, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; + +static const struct afs_operation_ops afs_rename_exchange_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_exchange, + .success = afs_rename_success, + .edit_dir = afs_rename_exchange_edit_dir, + .put = afs_rename_put, +}; + /* * rename a file in an AFS filesystem and/or move it between directories */ -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { - struct afs_fs_cursor fc; - struct afs_vnode *orig_dvnode, *new_dvnode, *vnode; - struct key *key; - u64 orig_data_version, new_data_version; - bool new_negative = d_is_negative(new_dentry); + struct afs_operation *op; + struct afs_vnode *orig_dvnode, *new_dvnode, *vnode, *new_vnode = NULL; int ret; - if (flags) + if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE)) + return -EINVAL; + + /* Don't allow silly-rename files be moved around. */ + if (old_dentry->d_flags & DCACHE_NFSFS_RENAMED) return -EINVAL; vnode = AFS_FS_I(d_inode(old_dentry)); orig_dvnode = AFS_FS_I(old_dir); new_dvnode = AFS_FS_I(new_dir); - orig_data_version = orig_dvnode->status.data_version; - new_data_version = new_dvnode->status.data_version; + if (d_is_positive(new_dentry)) + new_vnode = AFS_FS_I(d_inode(new_dentry)); _enter("{%llx:%llu},{%llx:%llu},{%llx:%llu},{%pd}", orig_dvnode->fid.vid, orig_dvnode->fid.vnode, @@ -1571,99 +2081,142 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dvnode->fid.vid, new_dvnode->fid.vnode, new_dentry); - key = afs_request_key(orig_dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } - - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, orig_dvnode, key)) { - if (orig_dvnode != new_dvnode) { - if (mutex_lock_interruptible_nested(&new_dvnode->io_lock, 1) < 0) { - afs_end_vnode_operation(&fc); - goto error_key; - } - } - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(orig_dvnode); - fc.cb_break_2 = afs_calc_vnode_cb_break(new_dvnode); - afs_fs_rename(&fc, old_dentry->d_name.name, - new_dvnode, new_dentry->d_name.name, - orig_data_version, new_data_version); - } + op = afs_alloc_operation(NULL, orig_dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break); - afs_vnode_commit_status(&fc, new_dvnode, fc.cb_break_2); - if (orig_dvnode != new_dvnode) - mutex_unlock(&new_dvnode->io_lock); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } + fscache_use_cookie(afs_vnode_cache(orig_dvnode), true); + if (new_dvnode != orig_dvnode) + fscache_use_cookie(afs_vnode_cache(new_dvnode), true); - if (ret == 0) { - if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags)) - afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name, - afs_edit_dir_for_rename); + ret = afs_validate(vnode, op->key); + afs_op_set_error(op, ret); + if (ret < 0) + goto error; - if (!new_negative && - test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) - afs_edit_dir_remove(new_dvnode, &new_dentry->d_name, - afs_edit_dir_for_rename); + ret = -ENOMEM; + op->more_files = kvcalloc(2, sizeof(struct afs_vnode_param), GFP_KERNEL); + if (!op->more_files) + goto error; - if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) - afs_edit_dir_add(new_dvnode, &new_dentry->d_name, - &vnode->fid, afs_edit_dir_for_rename); - } + afs_op_set_vnode(op, 0, orig_dvnode); + afs_op_set_vnode(op, 1, new_dvnode); /* May be same as orig_dvnode */ + op->file[0].dv_delta = 1; + op->file[1].dv_delta = 1; + op->file[0].modification = true; + op->file[1].modification = true; + op->file[0].update_ctime = true; + op->file[1].update_ctime = true; + op->more_files[0].vnode = vnode; + op->more_files[0].speculative = true; + op->more_files[1].vnode = new_vnode; + op->more_files[1].speculative = true; + op->nr_files = 4; + + op->dentry = old_dentry; + op->dentry_2 = new_dentry; + op->rename.rename_flags = flags; + op->rename.new_negative = d_is_negative(new_dentry); + + if (flags & RENAME_NOREPLACE) { + op->ops = &afs_rename_noreplace_operation; + } else if (flags & RENAME_EXCHANGE) { + op->ops = &afs_rename_exchange_operation; + d_drop(new_dentry); + } else { + /* If we might displace the target, we might need to do silly + * rename. + */ + op->ops = &afs_rename_operation; -error_key: - key_put(key); -error: - _leave(" = %d", ret); - return ret; -} + /* For non-directories, check whether the target is busy and if + * so, make a copy of the dentry and then do a silly-rename. + * If the silly-rename succeeds, the copied dentry is hashed + * and becomes the new target. + */ + if (d_is_positive(new_dentry) && !d_is_dir(new_dentry)) { + /* To prevent any new references to the target during + * the rename, we unhash the dentry in advance. + */ + if (!d_unhashed(new_dentry)) { + d_drop(new_dentry); + op->rename.rehash = new_dentry; + } -/* - * Release a directory page and clean up its private state if it's not busy - * - return true if the page can now be released, false if not - */ -static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags) -{ - struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); + if (d_count(new_dentry) > 2) { + /* copy the target dentry's name */ + op->rename.tmp = d_alloc(new_dentry->d_parent, + &new_dentry->d_name); + if (!op->rename.tmp) { + afs_op_nomem(op); + goto error; + } + + ret = afs_sillyrename(new_dvnode, + AFS_FS_I(d_inode(new_dentry)), + new_dentry, op->key); + if (ret) { + afs_op_set_error(op, ret); + goto error; + } + + op->dentry_2 = op->rename.tmp; + op->rename.rehash = NULL; + op->rename.new_negative = true; + } + } + } - _enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, page->index); + /* This bit is potentially nasty as there's a potential race with + * afs_d_revalidate{,_rcu}(). We have to change d_fsdata on the dentry + * to reflect it's new parent's new data_version after the op, but + * d_revalidate may see old_dentry between the op having taken place + * and the version being updated. + * + * So drop the old_dentry for now to make other threads go through + * lookup instead - which we hold a lock against. + */ + d_drop(old_dentry); - set_page_private(page, 0); - ClearPagePrivate(page); + ret = afs_do_sync_operation(op); + if (ret == -ENOTSUPP) + ret = -EINVAL; +out: + afs_dir_unuse_cookie(orig_dvnode, ret); + if (new_dvnode != orig_dvnode) + afs_dir_unuse_cookie(new_dvnode, ret); + return ret; - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_relpg); - return 1; +error: + ret = afs_put_operation(op); + goto out; } /* - * invalidate part or all of a page - * - release a page and clean up its private data if offset is 0 (indicating - * the entire page) + * Write the file contents to the cache as a single blob. */ -static void afs_dir_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) +int afs_single_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); - - _enter("{%lu},%u,%u", page->index, offset, length); - - BUG_ON(!PageLocked(page)); - - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); + struct afs_vnode *dvnode = AFS_FS_I(mapping->host); + struct iov_iter iter; + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); + int ret = 0; - /* we clean up only if the entire page is being invalidated */ - if (offset == 0 && length == PAGE_SIZE) { - set_page_private(page, 0); - ClearPagePrivate(page); + /* Need to lock to prevent the folio queue and folios from being thrown + * away. + */ + down_read(&dvnode->validate_lock); + + if (is_dir ? + test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) : + atomic64_read(&dvnode->cb_expires_at) != AFS_NO_CB_PROMISE) { + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, + i_size_read(&dvnode->netfs.inode)); + ret = netfs_writeback_single(mapping, wbc, &iter); } + + up_read(&dvnode->validate_lock); + return ret; } diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c index 8b400f5aead5..fd3aa9f97ce6 100644 --- a/fs/afs/dir_edit.c +++ b/fs/afs/dir_edit.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS filesystem directory editing * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/kernel.h> @@ -14,6 +10,7 @@ #include <linux/namei.h> #include <linux/pagemap.h> #include <linux/iversion.h> +#include <linux/folio_queue.h> #include "internal.h" #include "xdr_fs.h" @@ -72,13 +69,11 @@ static int afs_find_contig_bits(union afs_xdr_dir_block *block, unsigned int nr_ static void afs_set_contig_bits(union afs_xdr_dir_block *block, int bit, unsigned int nr_slots) { - u64 mask, before, after; + u64 mask; mask = (1 << nr_slots) - 1; mask <<= bit; - before = *(u64 *)block->hdr.bitmap; - block->hdr.bitmap[0] |= (u8)(mask >> 0 * 8); block->hdr.bitmap[1] |= (u8)(mask >> 1 * 8); block->hdr.bitmap[2] |= (u8)(mask >> 2 * 8); @@ -87,8 +82,6 @@ static void afs_set_contig_bits(union afs_xdr_dir_block *block, block->hdr.bitmap[5] |= (u8)(mask >> 5 * 8); block->hdr.bitmap[6] |= (u8)(mask >> 6 * 8); block->hdr.bitmap[7] |= (u8)(mask >> 7 * 8); - - after = *(u64 *)block->hdr.bitmap; } /* @@ -97,13 +90,11 @@ static void afs_set_contig_bits(union afs_xdr_dir_block *block, static void afs_clear_contig_bits(union afs_xdr_dir_block *block, int bit, unsigned int nr_slots) { - u64 mask, before, after; + u64 mask; mask = (1 << nr_slots) - 1; mask <<= bit; - before = *(u64 *)block->hdr.bitmap; - block->hdr.bitmap[0] &= ~(u8)(mask >> 0 * 8); block->hdr.bitmap[1] &= ~(u8)(mask >> 1 * 8); block->hdr.bitmap[2] &= ~(u8)(mask >> 2 * 8); @@ -112,17 +103,69 @@ static void afs_clear_contig_bits(union afs_xdr_dir_block *block, block->hdr.bitmap[5] &= ~(u8)(mask >> 5 * 8); block->hdr.bitmap[6] &= ~(u8)(mask >> 6 * 8); block->hdr.bitmap[7] &= ~(u8)(mask >> 7 * 8); +} - after = *(u64 *)block->hdr.bitmap; +/* + * Get a specific block, extending the directory storage to cover it as needed. + */ +static union afs_xdr_dir_block *afs_dir_get_block(struct afs_dir_iter *iter, size_t block) +{ + struct folio_queue *fq; + struct afs_vnode *dvnode = iter->dvnode; + struct folio *folio; + size_t blpos = block * AFS_DIR_BLOCK_SIZE; + size_t blend = (block + 1) * AFS_DIR_BLOCK_SIZE, fpos = iter->fpos; + int ret; + + if (dvnode->directory_size < blend) { + size_t cur_size = dvnode->directory_size; + + ret = netfs_alloc_folioq_buffer( + NULL, &dvnode->directory, &cur_size, blend, + mapping_gfp_mask(dvnode->netfs.inode.i_mapping)); + dvnode->directory_size = cur_size; + if (ret < 0) + goto fail; + } + + fq = iter->fq; + if (!fq) + fq = dvnode->directory; + + /* Search the folio queue for the folio containing the block... */ + for (; fq; fq = fq->next) { + for (int s = iter->fq_slot; s < folioq_count(fq); s++) { + size_t fsize = folioq_folio_size(fq, s); + + if (blend <= fpos + fsize) { + /* ... and then return the mapped block. */ + folio = folioq_folio(fq, s); + if (WARN_ON_ONCE(folio_pos(folio) != fpos)) + goto fail; + iter->fq = fq; + iter->fq_slot = s; + iter->fpos = fpos; + return kmap_local_folio(folio, blpos - fpos); + } + fpos += fsize; + } + iter->fq_slot = 0; + } + +fail: + iter->fq = NULL; + iter->fq_slot = 0; + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_get_block); + return NULL; } /* * Scan a directory block looking for a dirent of the right name. */ -static int afs_dir_scan_block(union afs_xdr_dir_block *block, struct qstr *name, +static int afs_dir_scan_block(const union afs_xdr_dir_block *block, const struct qstr *name, unsigned int blocknum) { - union afs_xdr_dirent *de; + const union afs_xdr_dirent *de; u64 bitmap; int d, len, n; @@ -196,77 +239,58 @@ static void afs_edit_init_block(union afs_xdr_dir_block *meta, * The caller must hold the inode locked. */ void afs_edit_dir_add(struct afs_vnode *vnode, - struct qstr *name, struct afs_fid *new_fid, + const struct qstr *name, struct afs_fid *new_fid, enum afs_edit_dir_reason why) { union afs_xdr_dir_block *meta, *block; - struct afs_xdr_dir_page *meta_page, *dir_page; union afs_xdr_dirent *de; - struct page *page0, *page; - unsigned int need_slots, nr_blocks, b; - pgoff_t index; + struct afs_dir_iter iter = { .dvnode = vnode }; + unsigned int nr_blocks, b, entry; loff_t i_size; - gfp_t gfp; int slot; _enter(",,{%d,%s},", name->len, name->name); - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_bad_size); return; } - gfp = vnode->vfs_inode.i_mapping->gfp_mask; - page0 = find_or_create_page(vnode->vfs_inode.i_mapping, 0, gfp); - if (!page0) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - _leave(" [fgp]"); + meta = afs_dir_get_block(&iter, 0); + if (!meta) return; - } /* Work out how many slots we're going to need. */ - need_slots = round_up(12 + name->len + 1 + 4, AFS_DIR_DIRENT_SIZE); - need_slots /= AFS_DIR_DIRENT_SIZE; + iter.nr_slots = afs_dir_calc_slots(name->len); - meta_page = kmap(page0); - meta = &meta_page->blocks[0]; if (i_size == 0) goto new_directory; nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; - /* Find a block that has sufficient slots available. Each VM page + /* Find a block that has sufficient slots available. Each folio * contains two or more directory blocks. */ for (b = 0; b < nr_blocks + 1; b++) { - /* If the directory extended into a new page, then we need to - * tack a new page on the end. + /* If the directory extended into a new folio, then we need to + * tack a new folio on the end. */ - index = b / AFS_DIR_BLOCKS_PER_PAGE; - if (index == 0) { - page = page0; - dir_page = meta_page; - } else { - if (nr_blocks >= AFS_DIR_MAX_BLOCKS) - goto error; - gfp = vnode->vfs_inode.i_mapping->gfp_mask; - page = find_or_create_page(vnode->vfs_inode.i_mapping, - index, gfp); - if (!page) - goto error; - if (!PagePrivate(page)) { - set_page_private(page, 1); - SetPagePrivate(page); - } - dir_page = kmap(page); - } + if (nr_blocks >= AFS_DIR_MAX_BLOCKS) + goto error_too_many_blocks; + + /* Lower dir blocks have a counter in the header we can check. */ + if (b < AFS_DIR_BLOCKS_WITH_CTR && + meta->meta.alloc_ctrs[b] < iter.nr_slots) + continue; + + block = afs_dir_get_block(&iter, b); + if (!block) + goto error; /* Abandon the edit if we got a callback break. */ if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - goto invalidated; - - block = &dir_page->blocks[b % AFS_DIR_BLOCKS_PER_PAGE]; + goto already_invalidated; _debug("block %u: %2u %3u %u", b, @@ -278,43 +302,34 @@ void afs_edit_dir_add(struct afs_vnode *vnode, if (b == nr_blocks) { _debug("init %u", b); afs_edit_init_block(meta, block, b); - i_size_write(&vnode->vfs_inode, (b + 1) * AFS_DIR_BLOCK_SIZE); + afs_set_i_size(vnode, (b + 1) * AFS_DIR_BLOCK_SIZE); } - /* Only lower dir pages have a counter in the header. */ - if (b >= AFS_DIR_BLOCKS_WITH_CTR || - meta->meta.alloc_ctrs[b] >= need_slots) { - /* We need to try and find one or more consecutive - * slots to hold the entry. - */ - slot = afs_find_contig_bits(block, need_slots); - if (slot >= 0) { - _debug("slot %u", slot); - goto found_space; - } + /* We need to try and find one or more consecutive slots to + * hold the entry. + */ + slot = afs_find_contig_bits(block, iter.nr_slots); + if (slot >= 0) { + _debug("slot %u", slot); + goto found_space; } - if (page != page0) { - unlock_page(page); - kunmap(page); - put_page(page); - } + kunmap_local(block); } /* There are no spare slots of sufficient size, yet the operation * succeeded. Download the directory again. */ trace_afs_edit_dir(vnode, why, afs_edit_dir_create_nospc, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_no_slots); goto out_unmap; new_directory: afs_edit_init_block(meta, meta, 0); i_size = AFS_DIR_BLOCK_SIZE; - i_size_write(&vnode->vfs_inode, i_size); + afs_set_i_size(vnode, i_size); slot = AFS_DIR_RESV_BLOCKS0; - page = page0; - block = meta; + block = afs_dir_get_block(&iter, 0); nr_blocks = 1; b = 0; @@ -332,40 +347,39 @@ found_space: de->u.name[name->len] = 0; /* Adjust the bitmap. */ - afs_set_contig_bits(block, slot, need_slots); - if (page != page0) { - unlock_page(page); - kunmap(page); - put_page(page); - } + afs_set_contig_bits(block, slot, iter.nr_slots); /* Adjust the allocation counter. */ if (b < AFS_DIR_BLOCKS_WITH_CTR) - meta->meta.alloc_ctrs[b] -= need_slots; + meta->meta.alloc_ctrs[b] -= iter.nr_slots; + + /* Adjust the hash chain. */ + entry = b * AFS_DIR_SLOTS_PER_BLOCK + slot; + iter.bucket = afs_dir_hash_name(name); + de->u.hash_next = meta->meta.hashtable[iter.bucket]; + meta->meta.hashtable[iter.bucket] = htons(entry); + kunmap_local(block); - inode_inc_iversion_raw(&vnode->vfs_inode); + inode_inc_iversion_raw(&vnode->netfs.inode); afs_stat_v(vnode, n_dir_cr); _debug("Insert %s in %u[%u]", name->name, b, slot); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); + out_unmap: - unlock_page(page0); - kunmap(page0); - put_page(page0); + kunmap_local(meta); _leave(""); return; -invalidated: +already_invalidated: trace_afs_edit_dir(vnode, why, afs_edit_dir_create_inval, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - if (page != page0) { - kunmap(page); - put_page(page); - } + kunmap_local(block); goto out_unmap; +error_too_many_blocks: + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_too_many_blocks); error: trace_afs_edit_dir(vnode, why, afs_edit_dir_create_error, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out_unmap; } @@ -377,129 +391,258 @@ error: * The caller must hold the inode locked. */ void afs_edit_dir_remove(struct afs_vnode *vnode, - struct qstr *name, enum afs_edit_dir_reason why) + const struct qstr *name, enum afs_edit_dir_reason why) { - struct afs_xdr_dir_page *meta_page, *dir_page; - union afs_xdr_dir_block *meta, *block; - union afs_xdr_dirent *de; - struct page *page0, *page; - unsigned int need_slots, nr_blocks, b; - pgoff_t index; + union afs_xdr_dir_block *meta, *block, *pblock; + union afs_xdr_dirent *de, *pde; + struct afs_dir_iter iter = { .dvnode = vnode }; + struct afs_fid fid; + unsigned int b, slot, entry; loff_t i_size; - int slot; + __be16 next; + int found; _enter(",,{%d,%s},", name->len, name->name); - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (i_size < AFS_DIR_BLOCK_SIZE || i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_rem_bad_size); return; } - nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; - page0 = find_lock_page(vnode->vfs_inode.i_mapping, 0); - if (!page0) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - _leave(" [fgp]"); + if (!afs_dir_init_iter(&iter, name)) return; - } - - /* Work out how many slots we're going to discard. */ - need_slots = round_up(12 + name->len + 1 + 4, AFS_DIR_DIRENT_SIZE); - need_slots /= AFS_DIR_DIRENT_SIZE; - - meta_page = kmap(page0); - meta = &meta_page->blocks[0]; - - /* Find a page that has sufficient slots available. Each VM page - * contains two or more directory blocks. - */ - for (b = 0; b < nr_blocks; b++) { - index = b / AFS_DIR_BLOCKS_PER_PAGE; - if (index != 0) { - page = find_lock_page(vnode->vfs_inode.i_mapping, index); - if (!page) - goto error; - dir_page = kmap(page); - } else { - page = page0; - dir_page = meta_page; - } - - /* Abandon the edit if we got a callback break. */ - if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - goto invalidated; - - block = &dir_page->blocks[b % AFS_DIR_BLOCKS_PER_PAGE]; - if (b > AFS_DIR_BLOCKS_WITH_CTR || - meta->meta.alloc_ctrs[b] <= AFS_DIR_SLOTS_PER_BLOCK - 1 - need_slots) { - slot = afs_dir_scan_block(block, name, b); - if (slot >= 0) - goto found_dirent; - } + meta = afs_dir_find_block(&iter, 0); + if (!meta) + return; - if (page != page0) { - unlock_page(page); - kunmap(page); - put_page(page); - } + /* Find the entry in the blob. */ + found = afs_dir_search_bucket(&iter, name, &fid); + if (found < 0) { + /* Didn't find the dirent to clobber. Re-download. */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_noent, + 0, 0, 0, 0, name->name); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_rem_wrong_name); + goto out_unmap; } - /* Didn't find the dirent to clobber. Download the directory again. */ - trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_noent, - 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - goto out_unmap; + entry = found; + b = entry / AFS_DIR_SLOTS_PER_BLOCK; + slot = entry % AFS_DIR_SLOTS_PER_BLOCK; -found_dirent: + block = afs_dir_find_block(&iter, b); + if (!block) + goto error; + if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + goto already_invalidated; + + /* Check and clear the entry. */ de = &block->dirents[slot]; + if (de->u.valid != 1) + goto error_unmap; trace_afs_edit_dir(vnode, why, afs_edit_dir_delete, b, slot, ntohl(de->u.vnode), ntohl(de->u.unique), name->name); - memset(de, 0, sizeof(*de) * need_slots); - /* Adjust the bitmap. */ - afs_clear_contig_bits(block, slot, need_slots); - if (page != page0) { - unlock_page(page); - kunmap(page); - put_page(page); - } + afs_clear_contig_bits(block, slot, iter.nr_slots); /* Adjust the allocation counter. */ if (b < AFS_DIR_BLOCKS_WITH_CTR) - meta->meta.alloc_ctrs[b] += need_slots; + meta->meta.alloc_ctrs[b] += iter.nr_slots; - inode_set_iversion_raw(&vnode->vfs_inode, vnode->status.data_version); + /* Clear the constituent entries. */ + next = de->u.hash_next; + memset(de, 0, sizeof(*de) * iter.nr_slots); + kunmap_local(block); + + /* Adjust the hash chain: if iter->prev_entry is 0, the hashtable head + * index is previous; otherwise it's slot number of the previous entry. + */ + if (!iter.prev_entry) { + __be16 prev_next = meta->meta.hashtable[iter.bucket]; + + if (unlikely(prev_next != htons(entry))) { + pr_warn("%llx:%llx:%x: not head of chain b=%x p=%x,%x e=%x %*s", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, + iter.bucket, iter.prev_entry, prev_next, entry, + name->len, name->name); + goto error; + } + meta->meta.hashtable[iter.bucket] = next; + } else { + unsigned int pb = iter.prev_entry / AFS_DIR_SLOTS_PER_BLOCK; + unsigned int ps = iter.prev_entry % AFS_DIR_SLOTS_PER_BLOCK; + __be16 prev_next; + + pblock = afs_dir_find_block(&iter, pb); + if (!pblock) + goto error; + pde = &pblock->dirents[ps]; + prev_next = pde->u.hash_next; + if (prev_next != htons(entry)) { + kunmap_local(pblock); + pr_warn("%llx:%llx:%x: not prev in chain b=%x p=%x,%x e=%x %*s", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, + iter.bucket, iter.prev_entry, prev_next, entry, + name->len, name->name); + goto error; + } + pde->u.hash_next = next; + kunmap_local(pblock); + } + + netfs_single_mark_inode_dirty(&vnode->netfs.inode); + + inode_set_iversion_raw(&vnode->netfs.inode, vnode->status.data_version); afs_stat_v(vnode, n_dir_rm); _debug("Remove %s from %u[%u]", name->name, b, slot); out_unmap: - unlock_page(page0); - kunmap(page0); - put_page(page0); + kunmap_local(meta); _leave(""); return; -invalidated: +already_invalidated: + kunmap_local(block); trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_inval, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - if (page != page0) { - unlock_page(page); - kunmap(page); - put_page(page); - } goto out_unmap; +error_unmap: + kunmap_local(block); error: trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_error, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out_unmap; } + +/* + * Edit an entry in a directory to update the vnode it refers to. This is also + * used to update the ".." entry in a directory. + */ +void afs_edit_dir_update(struct afs_vnode *vnode, const struct qstr *name, + struct afs_vnode *new_dvnode, enum afs_edit_dir_reason why) +{ + union afs_xdr_dir_block *block; + union afs_xdr_dirent *de; + struct afs_dir_iter iter = { .dvnode = vnode }; + unsigned int nr_blocks, b; + loff_t i_size; + int slot; + + _enter(""); + + i_size = i_size_read(&vnode->netfs.inode); + if (i_size < AFS_DIR_BLOCK_SIZE) { + afs_invalidate_dir(vnode, afs_dir_invalid_edit_upd_bad_size); + return; + } + + nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; + + /* Find a block that has sufficient slots available. Each folio + * contains two or more directory blocks. + */ + for (b = 0; b < nr_blocks; b++) { + block = afs_dir_get_block(&iter, b); + if (!block) + goto error; + + /* Abandon the edit if we got a callback break. */ + if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + goto already_invalidated; + + slot = afs_dir_scan_block(block, name, b); + if (slot >= 0) + goto found_dirent; + + kunmap_local(block); + } + + /* Didn't find the dirent to clobber. Download the directory again. */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_update_nodd, + 0, 0, 0, 0, name->name); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_upd_no_dd); + goto out; + +found_dirent: + de = &block->dirents[slot]; + de->u.vnode = htonl(new_dvnode->fid.vnode); + de->u.unique = htonl(new_dvnode->fid.unique); + + trace_afs_edit_dir(vnode, why, afs_edit_dir_update_dd, b, slot, + ntohl(de->u.vnode), ntohl(de->u.unique), name->name); + + kunmap_local(block); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); + inode_set_iversion_raw(&vnode->netfs.inode, vnode->status.data_version); + +out: + _leave(""); + return; + +already_invalidated: + kunmap_local(block); + trace_afs_edit_dir(vnode, why, afs_edit_dir_update_inval, + 0, 0, 0, 0, name->name); + goto out; + +error: + trace_afs_edit_dir(vnode, why, afs_edit_dir_update_error, + 0, 0, 0, 0, name->name); + goto out; +} + +/* + * Initialise a new directory. We need to fill in the "." and ".." entries. + */ +void afs_mkdir_init_dir(struct afs_vnode *dvnode, struct afs_vnode *parent_dvnode) +{ + union afs_xdr_dir_block *meta; + struct afs_dir_iter iter = { .dvnode = dvnode }; + union afs_xdr_dirent *de; + unsigned int slot = AFS_DIR_RESV_BLOCKS0; + loff_t i_size; + + i_size = i_size_read(&dvnode->netfs.inode); + if (i_size != AFS_DIR_BLOCK_SIZE) { + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_add_bad_size); + return; + } + + meta = afs_dir_get_block(&iter, 0); + if (!meta) + return; + + afs_edit_init_block(meta, meta, 0); + + de = &meta->dirents[slot]; + de->u.valid = 1; + de->u.vnode = htonl(dvnode->fid.vnode); + de->u.unique = htonl(dvnode->fid.unique); + memcpy(de->u.name, ".", 2); + trace_afs_edit_dir(dvnode, afs_edit_dir_for_mkdir, afs_edit_dir_mkdir, 0, slot, + dvnode->fid.vnode, dvnode->fid.unique, "."); + slot++; + + de = &meta->dirents[slot]; + de->u.valid = 1; + de->u.vnode = htonl(parent_dvnode->fid.vnode); + de->u.unique = htonl(parent_dvnode->fid.unique); + memcpy(de->u.name, "..", 3); + trace_afs_edit_dir(dvnode, afs_edit_dir_for_mkdir, afs_edit_dir_mkdir, 0, slot, + parent_dvnode->fid.vnode, parent_dvnode->fid.unique, ".."); + + afs_set_contig_bits(meta, AFS_DIR_RESV_BLOCKS0, 2); + meta->meta.alloc_ctrs[0] -= 2; + kunmap_local(meta); + + netfs_single_mark_inode_dirty(&dvnode->netfs.inode); + set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + set_bit(AFS_VNODE_DIR_READ, &dvnode->flags); +} diff --git a/fs/afs/dir_search.c b/fs/afs/dir_search.c new file mode 100644 index 000000000000..d2516e55b5ed --- /dev/null +++ b/fs/afs/dir_search.c @@ -0,0 +1,227 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Search a directory's hash table. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * https://tools.ietf.org/html/draft-keiser-afs3-directory-object-00 + */ + +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/iversion.h> +#include "internal.h" +#include "afs_fs.h" +#include "xdr_fs.h" + +/* + * Calculate the name hash. + */ +unsigned int afs_dir_hash_name(const struct qstr *name) +{ + const unsigned char *p = name->name; + unsigned int hash = 0, i; + int bucket; + + for (i = 0; i < name->len; i++) + hash = (hash * 173) + p[i]; + bucket = hash & (AFS_DIR_HASHTBL_SIZE - 1); + if (hash > INT_MAX) { + bucket = AFS_DIR_HASHTBL_SIZE - bucket; + bucket &= (AFS_DIR_HASHTBL_SIZE - 1); + } + return bucket; +} + +/* + * Reset a directory iterator. + */ +static bool afs_dir_reset_iter(struct afs_dir_iter *iter) +{ + unsigned long long i_size = i_size_read(&iter->dvnode->netfs.inode); + unsigned int nblocks; + + /* Work out the maximum number of steps we can take. */ + nblocks = umin(i_size / AFS_DIR_BLOCK_SIZE, AFS_DIR_MAX_BLOCKS); + if (!nblocks) + return false; + iter->loop_check = nblocks * (AFS_DIR_SLOTS_PER_BLOCK - AFS_DIR_RESV_BLOCKS); + iter->prev_entry = 0; /* Hash head is previous */ + return true; +} + +/* + * Initialise a directory iterator for looking up a name. + */ +bool afs_dir_init_iter(struct afs_dir_iter *iter, const struct qstr *name) +{ + iter->nr_slots = afs_dir_calc_slots(name->len); + iter->bucket = afs_dir_hash_name(name); + return afs_dir_reset_iter(iter); +} + +/* + * Get a specific block. + */ +union afs_xdr_dir_block *afs_dir_find_block(struct afs_dir_iter *iter, size_t block) +{ + struct folio_queue *fq = iter->fq; + struct afs_vnode *dvnode = iter->dvnode; + struct folio *folio; + size_t blpos = block * AFS_DIR_BLOCK_SIZE; + size_t blend = (block + 1) * AFS_DIR_BLOCK_SIZE, fpos = iter->fpos; + int slot = iter->fq_slot; + + _enter("%zx,%d", block, slot); + + if (iter->block) { + kunmap_local(iter->block); + iter->block = NULL; + } + + if (dvnode->directory_size < blend) + goto fail; + + if (!fq || blpos < fpos) { + fq = dvnode->directory; + slot = 0; + fpos = 0; + } + + /* Search the folio queue for the folio containing the block... */ + for (; fq; fq = fq->next) { + for (; slot < folioq_count(fq); slot++) { + size_t fsize = folioq_folio_size(fq, slot); + + if (blend <= fpos + fsize) { + /* ... and then return the mapped block. */ + folio = folioq_folio(fq, slot); + if (WARN_ON_ONCE(folio_pos(folio) != fpos)) + goto fail; + iter->fq = fq; + iter->fq_slot = slot; + iter->fpos = fpos; + iter->block = kmap_local_folio(folio, blpos - fpos); + return iter->block; + } + fpos += fsize; + } + slot = 0; + } + +fail: + iter->fq = NULL; + iter->fq_slot = 0; + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_get_block); + return NULL; +} + +/* + * Search through a directory bucket. + */ +int afs_dir_search_bucket(struct afs_dir_iter *iter, const struct qstr *name, + struct afs_fid *_fid) +{ + const union afs_xdr_dir_block *meta; + unsigned int entry; + int ret = -ESTALE; + + meta = afs_dir_find_block(iter, 0); + if (!meta) + return -ESTALE; + + entry = ntohs(meta->meta.hashtable[iter->bucket & (AFS_DIR_HASHTBL_SIZE - 1)]); + _enter("%x,%x", iter->bucket, entry); + + while (entry) { + const union afs_xdr_dir_block *block; + const union afs_xdr_dirent *dire; + unsigned int blnum = entry / AFS_DIR_SLOTS_PER_BLOCK; + unsigned int slot = entry % AFS_DIR_SLOTS_PER_BLOCK; + unsigned int resv = (blnum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + + _debug("search %x", entry); + + if (slot < resv) { + kdebug("slot out of range h=%x rs=%2x sl=%2x-%2x", + iter->bucket, resv, slot, slot + iter->nr_slots - 1); + goto bad; + } + + block = afs_dir_find_block(iter, blnum); + if (!block) + goto bad; + dire = &block->dirents[slot]; + + if (slot + iter->nr_slots <= AFS_DIR_SLOTS_PER_BLOCK && + memcmp(dire->u.name, name->name, name->len) == 0 && + dire->u.name[name->len] == '\0') { + _fid->vnode = ntohl(dire->u.vnode); + _fid->unique = ntohl(dire->u.unique); + ret = entry; + goto found; + } + + iter->prev_entry = entry; + entry = ntohs(dire->u.hash_next); + if (!--iter->loop_check) { + kdebug("dir chain loop h=%x", iter->bucket); + goto bad; + } + } + + ret = -ENOENT; +found: + if (iter->block) { + kunmap_local(iter->block); + iter->block = NULL; + } + +bad: + if (ret == -ESTALE) + afs_invalidate_dir(iter->dvnode, afs_dir_invalid_iter_stale); + _leave(" = %d", ret); + return ret; +} + +/* + * Search the appropriate hash chain in the contents of an AFS directory. + */ +int afs_dir_search(struct afs_vnode *dvnode, const struct qstr *name, + struct afs_fid *_fid, afs_dataversion_t *_dir_version) +{ + struct afs_dir_iter iter = { .dvnode = dvnode, }; + int ret, retry_limit = 3; + + _enter("{%lu},,,", dvnode->netfs.inode.i_ino); + + if (!afs_dir_init_iter(&iter, name)) + return -ENOENT; + do { + if (--retry_limit < 0) { + pr_warn("afs_read_dir(): Too many retries\n"); + ret = -ESTALE; + break; + } + ret = afs_read_dir(dvnode, NULL); + if (ret < 0) { + if (ret != -ESTALE) + break; + if (test_bit(AFS_VNODE_DELETED, &dvnode->flags)) { + ret = -ESTALE; + break; + } + continue; + } + *_dir_version = inode_peek_iversion_raw(&dvnode->netfs.inode); + + ret = afs_dir_search_bucket(&iter, name, _fid); + up_read(&dvnode->validate_lock); + if (ret == -ESTALE) + afs_dir_reset_iter(&iter); + } while (ret == -ESTALE); + + _leave(" = %d", ret); + return ret; +} diff --git a/fs/afs/dir_silly.c b/fs/afs/dir_silly.c new file mode 100644 index 000000000000..014495d4b868 --- /dev/null +++ b/fs/afs/dir_silly.c @@ -0,0 +1,291 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* AFS silly rename handling + * + * Copyright (C) 2019 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * - Derived from NFS's sillyrename. + */ + +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/fsnotify.h> +#include "internal.h" + +static void afs_silly_rename_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + + afs_check_dir_conflict(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[0]); +} + +static void afs_silly_rename_edit_dir(struct afs_operation *op) +{ + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + struct afs_vnode *vnode = AFS_FS_I(d_inode(op->dentry)); + struct dentry *old = op->dentry; + struct dentry *new = op->dentry_2; + + spin_lock(&old->d_lock); + old->d_flags |= DCACHE_NFSFS_RENAMED; + spin_unlock(&old->d_lock); + if (dvnode->silly_key != op->key) { + key_put(dvnode->silly_key); + dvnode->silly_key = key_get(op->key); + } + + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) { + afs_edit_dir_remove(dvnode, &old->d_name, + afs_edit_dir_for_silly_0); + afs_edit_dir_add(dvnode, &new->d_name, + &vnode->fid, afs_edit_dir_for_silly_1); + } + up_write(&dvnode->validate_lock); +} + +static const struct afs_operation_ops afs_silly_rename_operation = { + .issue_afs_rpc = afs_fs_rename, + .issue_yfs_rpc = yfs_fs_rename, + .success = afs_silly_rename_success, + .edit_dir = afs_silly_rename_edit_dir, +}; + +/* + * Actually perform the silly rename step. + */ +static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode, + struct dentry *old, struct dentry *new, + struct key *key) +{ + struct afs_operation *op; + + _enter("%pd,%pd", old, new); + + op = afs_alloc_operation(key, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + op->more_files = kvcalloc(2, sizeof(struct afs_vnode_param), GFP_KERNEL); + if (!op->more_files) { + afs_put_operation(op); + return -ENOMEM; + } + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_vnode(op, 1, dvnode); + op->file[0].dv_delta = 1; + op->file[1].dv_delta = 1; + op->file[0].modification = true; + op->file[1].modification = true; + op->file[0].update_ctime = true; + op->file[1].update_ctime = true; + op->more_files[0].vnode = AFS_FS_I(d_inode(old)); + op->more_files[0].speculative = true; + op->more_files[1].vnode = AFS_FS_I(d_inode(new)); + op->more_files[1].speculative = true; + op->nr_files = 4; + + op->dentry = old; + op->dentry_2 = new; + op->ops = &afs_silly_rename_operation; + + trace_afs_silly_rename(vnode, false); + return afs_do_sync_operation(op); +} + +/* + * Perform silly-rename of a dentry. + * + * AFS is stateless and the server doesn't know when the client is holding a + * file open. To prevent application problems when a file is unlinked while + * it's still open, the client performs a "silly-rename". That is, it renames + * the file to a hidden file in the same directory, and only performs the + * unlink once the last reference to it is put. + * + * The final cleanup is done during dentry_iput. + */ +int afs_sillyrename(struct afs_vnode *dvnode, struct afs_vnode *vnode, + struct dentry *dentry, struct key *key) +{ + static unsigned int sillycounter; + struct dentry *sdentry = NULL; + unsigned char silly[16]; + int ret = -EBUSY; + + _enter(""); + + /* We don't allow a dentry to be silly-renamed twice. */ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + return -EBUSY; + + sdentry = NULL; + do { + dput(sdentry); + sillycounter++; + + /* Create a silly name. Note that the ".__afs" prefix is + * understood by the salvager and must not be changed. + */ + scnprintf(silly, sizeof(silly), ".__afs%04X", sillycounter); + sdentry = lookup_noperm(&QSTR(silly), dentry->d_parent); + + /* N.B. Better to return EBUSY here ... it could be dangerous + * to delete the file while it's in use. + */ + if (IS_ERR(sdentry)) + goto out; + } while (!d_is_negative(sdentry)); + + ihold(&vnode->netfs.inode); + + ret = afs_do_silly_rename(dvnode, vnode, dentry, sdentry, key); + switch (ret) { + case 0: + /* The rename succeeded. */ + set_bit(AFS_VNODE_SILLY_DELETED, &vnode->flags); + d_move(dentry, sdentry); + break; + case -ERESTARTSYS: + /* The result of the rename is unknown. Play it safe by forcing + * a new lookup. + */ + d_drop(dentry); + d_drop(sdentry); + } + + iput(&vnode->netfs.inode); + dput(sdentry); +out: + _leave(" = %d", ret); + return ret; +} + +static void afs_silly_unlink_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + afs_check_dir_conflict(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[1]); + afs_update_dentry_version(op, &op->file[0], op->dentry); +} + +static void afs_silly_unlink_edit_dir(struct afs_operation *op) +{ + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_remove(dvnode, &op->dentry->d_name, + afs_edit_dir_for_unlink); + up_write(&dvnode->validate_lock); +} + +static const struct afs_operation_ops afs_silly_unlink_operation = { + .issue_afs_rpc = afs_fs_remove_file, + .issue_yfs_rpc = yfs_fs_remove_file, + .success = afs_silly_unlink_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_silly_unlink_edit_dir, +}; + +/* + * Tell the server to remove a sillyrename file. + */ +static int afs_do_silly_unlink(struct afs_vnode *dvnode, struct afs_vnode *vnode, + struct dentry *dentry, struct key *key) +{ + struct afs_operation *op; + + _enter(""); + + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_vnode(op, 1, vnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + op->file[1].op_unlinked = true; + op->file[1].update_ctime = true; + + op->dentry = dentry; + op->ops = &afs_silly_unlink_operation; + + trace_afs_silly_rename(vnode, true); + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + + /* If there was a conflict with a third party, check the status of the + * unlinked vnode. + */ + if (op->cumul_error.error == 0 && (op->flags & AFS_OPERATION_DIR_CONFLICT)) { + op->file[1].update_ctime = false; + op->fetch_status.which = 1; + op->ops = &afs_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + } + + return afs_put_operation(op); +} + +/* + * Remove sillyrename file on iput. + */ +int afs_silly_iput(struct dentry *dentry, struct inode *inode) +{ + struct afs_vnode *dvnode = AFS_FS_I(d_inode(dentry->d_parent)); + struct afs_vnode *vnode = AFS_FS_I(inode); + struct dentry *alias; + int ret; + + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + + _enter("%p{%pd},%llx", dentry, dentry, vnode->fid.vnode); + + down_read(&dvnode->rmdir_lock); + + alias = d_alloc_parallel(dentry->d_parent, &dentry->d_name, &wq); + if (IS_ERR(alias)) { + up_read(&dvnode->rmdir_lock); + return 0; + } + + if (!d_in_lookup(alias)) { + /* We raced with lookup... See if we need to transfer the + * sillyrename information to the aliased dentry. + */ + ret = 0; + spin_lock(&alias->d_lock); + if (d_really_is_positive(alias) && + !(alias->d_flags & DCACHE_NFSFS_RENAMED)) { + alias->d_flags |= DCACHE_NFSFS_RENAMED; + ret = 1; + } + spin_unlock(&alias->d_lock); + up_read(&dvnode->rmdir_lock); + dput(alias); + return ret; + } + + /* Stop lock-release from complaining. */ + spin_lock(&vnode->lock); + vnode->lock_state = AFS_VNODE_LOCK_DELETED; + trace_afs_flock_ev(vnode, NULL, afs_flock_silly_delete, 0); + spin_unlock(&vnode->lock); + + afs_do_silly_unlink(dvnode, vnode, dentry, dvnode->silly_key); + up_read(&dvnode->rmdir_lock); + d_lookup_done(alias); + dput(alias); + return 1; +} diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c index a9ba81ddf154..aa56e8951e03 100644 --- a/fs/afs/dynroot.c +++ b/fs/afs/dynroot.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS dynamic root handling * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/fs.h> @@ -14,125 +10,127 @@ #include <linux/dns_resolver.h> #include "internal.h" -const struct file_operations afs_dynroot_file_operations = { - .open = dcache_dir_open, - .release = dcache_dir_close, - .iterate_shared = dcache_readdir, - .llseek = dcache_dir_lseek, -}; +#define AFS_MIN_DYNROOT_CELL_INO 4 /* Allow for ., .., @cell, .@cell */ +#define AFS_MAX_DYNROOT_CELL_INO ((unsigned int)INT_MAX) + +static struct dentry *afs_lookup_atcell(struct inode *dir, struct dentry *dentry, ino_t ino); /* - * Probe to see if a cell may exist. This prevents positive dentries from - * being created unnecessarily. + * iget5() comparator for inode created by autocell operations */ -static int afs_probe_cell_name(struct dentry *dentry) +static int afs_iget5_pseudo_test(struct inode *inode, void *opaque) { - struct afs_cell *cell; - const char *name = dentry->d_name.name; - size_t len = dentry->d_name.len; - int ret; + struct afs_fid *fid = opaque; - /* Names prefixed with a dot are R/W mounts. */ - if (name[0] == '.') { - if (len == 1) - return -EINVAL; - name++; - len--; - } - - cell = afs_lookup_cell_rcu(afs_d2net(dentry), name, len); - if (!IS_ERR(cell)) { - afs_put_cell(afs_d2net(dentry), cell); - return 0; - } + return inode->i_ino == fid->vnode; +} - ret = dns_query("afsdb", name, len, "srv=1", NULL, NULL); - if (ret == -ENODATA) - ret = -EDESTADDRREQ; - return ret; +/* + * iget5() inode initialiser + */ +static int afs_iget5_pseudo_set(struct inode *inode, void *opaque) +{ + struct afs_super_info *as = AFS_FS_S(inode->i_sb); + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_fid *fid = opaque; + + vnode->volume = as->volume; + vnode->fid = *fid; + inode->i_ino = fid->vnode; + inode->i_generation = fid->unique; + return 0; } /* - * Try to auto mount the mountpoint with pseudo directory, if the autocell - * operation is setted. + * Create an inode for an autocell dynamic automount dir. */ -struct inode *afs_try_auto_mntpt(struct dentry *dentry, struct inode *dir) +static struct inode *afs_iget_pseudo_dir(struct super_block *sb, ino_t ino) { - struct afs_vnode *vnode = AFS_FS_I(dir); + struct afs_vnode *vnode; struct inode *inode; - int ret = -ENOENT; + struct afs_fid fid = { .vnode = ino, .unique = 1, }; - _enter("%p{%pd}, {%llx:%llu}", - dentry, dentry, vnode->fid.vid, vnode->fid.vnode); + _enter(""); - if (!test_bit(AFS_VNODE_AUTOCELL, &vnode->flags)) - goto out; + inode = iget5_locked(sb, fid.vnode, + afs_iget5_pseudo_test, afs_iget5_pseudo_set, &fid); + if (!inode) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } - ret = afs_probe_cell_name(dentry); - if (ret < 0) - goto out; + _debug("GOT INODE %p { ino=%lu, vl=%llx, vn=%llx, u=%x }", + inode, inode->i_ino, fid.vid, fid.vnode, fid.unique); - inode = afs_iget_pseudo_dir(dir->i_sb, false); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - goto out; - } + vnode = AFS_FS_I(inode); - _leave("= %p", inode); - return inode; + if (inode_state_read_once(inode) & I_NEW) { + netfs_inode_init(&vnode->netfs, NULL, false); + simple_inode_init_ts(inode); + set_nlink(inode, 2); + inode->i_size = 0; + inode->i_mode = S_IFDIR | 0555; + inode->i_op = &afs_autocell_inode_operations; + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_blocks = 0; + inode->i_generation = 0; + inode->i_flags |= S_AUTOMOUNT | S_NOATIME; -out: - _leave("= %d", ret); - return ret == -ENOENT ? NULL : ERR_PTR(ret); + set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags); + set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags); + + unlock_new_inode(inode); + } + _leave(" = %p", inode); + return inode; } /* - * Look up @cell in a dynroot directory. This is a substitution for the - * local cell name for the net namespace. + * Try to automount the mountpoint with pseudo directory, if the autocell + * option is set. */ -static struct dentry *afs_lookup_atcell(struct dentry *dentry) +static struct dentry *afs_dynroot_lookup_cell(struct inode *dir, struct dentry *dentry, + unsigned int flags) { - struct afs_cell *cell; + struct afs_cell *cell = NULL; struct afs_net *net = afs_d2net(dentry); - struct dentry *ret; - unsigned int seq = 0; - char *name; - int len; - - if (!net->ws_cell) - return ERR_PTR(-ENOENT); + struct inode *inode = NULL; + const char *name = dentry->d_name.name; + size_t len = dentry->d_name.len; + bool dotted = false; + int ret = -ENOENT; - ret = ERR_PTR(-ENOMEM); - name = kmalloc(AFS_MAXCELLNAME + 1, GFP_KERNEL); - if (!name) - goto out_p; - - rcu_read_lock(); - do { - read_seqbegin_or_lock(&net->cells_lock, &seq); - cell = rcu_dereference_raw(net->ws_cell); - if (cell) { - len = cell->name_len; - memcpy(name, cell->name, len + 1); - } - } while (need_seqretry(&net->cells_lock, seq)); - done_seqretry(&net->cells_lock, seq); - rcu_read_unlock(); + /* Names prefixed with a dot are R/W mounts. */ + if (name[0] == '.') { + name++; + len--; + dotted = true; + } - ret = ERR_PTR(-ENOENT); - if (!cell) - goto out_n; + cell = afs_lookup_cell(net, name, len, NULL, + AFS_LOOKUP_CELL_DYNROOT, + afs_cell_trace_use_lookup_dynroot); + if (IS_ERR(cell)) { + ret = PTR_ERR(cell); + goto out_no_cell; + } - ret = lookup_one_len(name, dentry->d_parent, len); + inode = afs_iget_pseudo_dir(dir->i_sb, cell->dynroot_ino * 2 + dotted); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + goto out; + } - /* We don't want to d_add() the @cell dentry here as we don't want to - * the cached dentry to hide changes to the local cell name. - */ + dentry->d_fsdata = cell; + return d_splice_alias(inode, dentry); -out_n: - kfree(name); -out_p: - return ret; +out: + afs_unuse_cell(cell, afs_cell_trace_unuse_lookup_dynroot); +out_no_cell: + if (!inode) + return d_splice_alias(inode, dentry); + return ret == -ENOENT ? NULL : ERR_PTR(ret); } /* @@ -143,7 +141,8 @@ static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentr { _enter("%pd", dentry); - ASSERTCMP(d_inode(dentry), ==, NULL); + if (flags & LOOKUP_CREATE) + return ERR_PTR(-EOPNOTSUPP); if (dentry->d_name.len >= AFSNAMEMAX) { _leave(" = -ENAMETOOLONG"); @@ -152,159 +151,256 @@ static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentr if (dentry->d_name.len == 5 && memcmp(dentry->d_name.name, "@cell", 5) == 0) - return afs_lookup_atcell(dentry); + return afs_lookup_atcell(dir, dentry, 2); + + if (dentry->d_name.len == 6 && + memcmp(dentry->d_name.name, ".@cell", 6) == 0) + return afs_lookup_atcell(dir, dentry, 3); - return d_splice_alias(afs_try_auto_mntpt(dentry, dir), dentry); + return afs_dynroot_lookup_cell(dir, dentry, flags); } const struct inode_operations afs_dynroot_inode_operations = { .lookup = afs_dynroot_lookup, }; -/* - * Dirs in the dynamic root don't need revalidation. - */ -static int afs_dynroot_d_revalidate(struct dentry *dentry, unsigned int flags) +static void afs_dynroot_d_release(struct dentry *dentry) { - return 1; + struct afs_cell *cell = dentry->d_fsdata; + + afs_unuse_cell(cell, afs_cell_trace_unuse_dynroot_mntpt); } /* - * Allow the VFS to enquire as to whether a dentry should be unhashed (mustn't - * sleep) - * - called from dput() when d_count is going to 0. - * - return 1 to request dentry be unhashed, 0 otherwise + * Keep @cell symlink dentries around, but only keep cell autodirs when they're + * being used. */ -static int afs_dynroot_d_delete(const struct dentry *dentry) +static int afs_dynroot_delete_dentry(const struct dentry *dentry) { - return d_really_is_positive(dentry); + const struct qstr *name = &dentry->d_name; + + if (name->len == 5 && memcmp(name->name, "@cell", 5) == 0) + return 0; + if (name->len == 6 && memcmp(name->name, ".@cell", 6) == 0) + return 0; + return 1; } const struct dentry_operations afs_dynroot_dentry_operations = { - .d_revalidate = afs_dynroot_d_revalidate, - .d_delete = afs_dynroot_d_delete, - .d_release = afs_d_release, + .d_delete = afs_dynroot_delete_dentry, + .d_release = afs_dynroot_d_release, .d_automount = afs_d_automount, }; +static void afs_atcell_delayed_put_cell(void *arg) +{ + struct afs_cell *cell = arg; + + afs_put_cell(cell, afs_cell_trace_put_atcell); +} + /* - * Create a manually added cell mount directory. - * - The caller must hold net->proc_cells_lock + * Read @cell or .@cell symlinks. */ -int afs_dynroot_mkdir(struct afs_net *net, struct afs_cell *cell) +static const char *afs_atcell_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *done) { - struct super_block *sb = net->dynroot_sb; - struct dentry *root, *subdir; - int ret; + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_cell *cell; + struct afs_net *net = afs_i2net(inode); + const char *name; + bool dotted = vnode->fid.vnode == 3; - if (!sb || atomic_read(&sb->s_active) == 0) - return 0; + if (!rcu_access_pointer(net->ws_cell)) + return ERR_PTR(-ENOENT); - /* Let the ->lookup op do the creation */ - root = sb->s_root; - inode_lock(root->d_inode); - subdir = lookup_one_len(cell->name, root, cell->name_len); - if (IS_ERR(subdir)) { - ret = PTR_ERR(subdir); - goto unlock; + if (!dentry) { + /* We're in RCU-pathwalk. */ + cell = rcu_dereference(net->ws_cell); + if (dotted) + name = cell->name - 1; + else + name = cell->name; + /* Shouldn't need to set a delayed call. */ + return name; } - /* Note that we're retaining an extra ref on the dentry */ - subdir->d_fsdata = (void *)1UL; - ret = 0; -unlock: - inode_unlock(root->d_inode); - return ret; + down_read(&net->cells_lock); + + cell = rcu_dereference_protected(net->ws_cell, lockdep_is_held(&net->cells_lock)); + if (dotted) + name = cell->name - 1; + else + name = cell->name; + afs_get_cell(cell, afs_cell_trace_get_atcell); + set_delayed_call(done, afs_atcell_delayed_put_cell, cell); + + up_read(&net->cells_lock); + return name; } +static const struct inode_operations afs_atcell_inode_operations = { + .get_link = afs_atcell_get_link, +}; + /* - * Remove a manually added cell mount directory. - * - The caller must hold net->proc_cells_lock + * Create an inode for the @cell or .@cell symlinks. */ -void afs_dynroot_rmdir(struct afs_net *net, struct afs_cell *cell) +static struct dentry *afs_lookup_atcell(struct inode *dir, struct dentry *dentry, ino_t ino) { - struct super_block *sb = net->dynroot_sb; - struct dentry *root, *subdir; - - if (!sb || atomic_read(&sb->s_active) == 0) - return; + struct afs_vnode *vnode; + struct inode *inode; + struct afs_fid fid = { .vnode = ino, .unique = 1, }; + + inode = iget5_locked(dir->i_sb, fid.vnode, + afs_iget5_pseudo_test, afs_iget5_pseudo_set, &fid); + if (!inode) + return ERR_PTR(-ENOMEM); + + vnode = AFS_FS_I(inode); + + if (inode_state_read_once(inode) & I_NEW) { + netfs_inode_init(&vnode->netfs, NULL, false); + simple_inode_init_ts(inode); + set_nlink(inode, 1); + inode->i_size = 0; + inode->i_mode = S_IFLNK | 0555; + inode->i_op = &afs_atcell_inode_operations; + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_blocks = 0; + inode->i_generation = 0; + inode->i_flags |= S_NOATIME; + + unlock_new_inode(inode); + } + return d_splice_alias(inode, dentry); +} - root = sb->s_root; - inode_lock(root->d_inode); +/* + * Transcribe the cell database into readdir content under the RCU read lock. + * Each cell produces two entries, one prefixed with a dot and one not. + */ +static int afs_dynroot_readdir_cells(struct afs_net *net, struct dir_context *ctx) +{ + const struct afs_cell *cell; + loff_t newpos; + + _enter("%llu", ctx->pos); + + for (;;) { + unsigned int ix = ctx->pos >> 1; + + cell = idr_get_next(&net->cells_dyn_ino, &ix); + if (!cell) + return 0; + if (READ_ONCE(cell->state) == AFS_CELL_REMOVING || + READ_ONCE(cell->state) == AFS_CELL_DEAD) { + ctx->pos += 2; + ctx->pos &= ~1; + continue; + } - /* Don't want to trigger a lookup call, which will re-add the cell */ - subdir = try_lookup_one_len(cell->name, root, cell->name_len); - if (IS_ERR_OR_NULL(subdir)) { - _debug("lookup %ld", PTR_ERR(subdir)); - goto no_dentry; - } + newpos = ix << 1; + if (newpos > ctx->pos) + ctx->pos = newpos; - _debug("rmdir %pd %u", subdir, d_count(subdir)); + _debug("pos %llu -> cell %u", ctx->pos, cell->dynroot_ino); - if (subdir->d_fsdata) { - _debug("unpin %u", d_count(subdir)); - subdir->d_fsdata = NULL; - dput(subdir); + if ((ctx->pos & 1) == 0) { + if (!dir_emit(ctx, cell->name, cell->name_len, + cell->dynroot_ino, DT_DIR)) + return 0; + ctx->pos++; + } + if ((ctx->pos & 1) == 1) { + if (!dir_emit(ctx, cell->name - 1, cell->name_len + 1, + cell->dynroot_ino + 1, DT_DIR)) + return 0; + ctx->pos++; + } } - dput(subdir); -no_dentry: - inode_unlock(root->d_inode); - _leave(""); + return 0; } /* - * Populate a newly created dynamic root with cell names. + * Read the AFS dynamic root directory. This produces a list of cellnames, + * dotted and undotted, along with @cell and .@cell links if configured. */ -int afs_dynroot_populate(struct super_block *sb) +static int afs_dynroot_readdir(struct file *file, struct dir_context *ctx) { - struct afs_cell *cell; - struct afs_net *net = afs_sb2net(sb); - int ret; + struct afs_net *net = afs_d2net(file->f_path.dentry); + int ret = 0; - if (mutex_lock_interruptible(&net->proc_cells_lock) < 0) - return -ERESTARTSYS; + if (!dir_emit_dots(file, ctx)) + return 0; - net->dynroot_sb = sb; - hlist_for_each_entry(cell, &net->proc_cells, proc_link) { - ret = afs_dynroot_mkdir(net, cell); - if (ret < 0) - goto error; + if (ctx->pos == 2) { + if (rcu_access_pointer(net->ws_cell) && + !dir_emit(ctx, "@cell", 5, 2, DT_LNK)) + return 0; + ctx->pos = 3; + } + if (ctx->pos == 3) { + if (rcu_access_pointer(net->ws_cell) && + !dir_emit(ctx, ".@cell", 6, 3, DT_LNK)) + return 0; + ctx->pos = 4; } - ret = 0; -out: - mutex_unlock(&net->proc_cells_lock); + if ((unsigned long long)ctx->pos <= AFS_MAX_DYNROOT_CELL_INO) { + down_read(&net->cells_lock); + ret = afs_dynroot_readdir_cells(net, ctx); + up_read(&net->cells_lock); + } return ret; - -error: - net->dynroot_sb = NULL; - goto out; } +static const struct file_operations afs_dynroot_file_operations = { + .llseek = generic_file_llseek, + .read = generic_read_dir, + .iterate_shared = afs_dynroot_readdir, + .fsync = noop_fsync, +}; + /* - * When a dynamic root that's in the process of being destroyed, depopulate it - * of pinned directories. + * Create an inode for a dynamic root directory. */ -void afs_dynroot_depopulate(struct super_block *sb) +struct inode *afs_dynroot_iget_root(struct super_block *sb) { - struct afs_net *net = afs_sb2net(sb); - struct dentry *root = sb->s_root, *subdir, *tmp; - - /* Prevent more subdirs from being created */ - mutex_lock(&net->proc_cells_lock); - if (net->dynroot_sb == sb) - net->dynroot_sb = NULL; - mutex_unlock(&net->proc_cells_lock); - - inode_lock(root->d_inode); - - /* Remove all the pins for dirs created for manually added cells */ - list_for_each_entry_safe(subdir, tmp, &root->d_subdirs, d_child) { - if (subdir->d_fsdata) { - subdir->d_fsdata = NULL; - dput(subdir); - } + struct afs_super_info *as = AFS_FS_S(sb); + struct afs_vnode *vnode; + struct inode *inode; + struct afs_fid fid = { .vid = 0, .vnode = 1, .unique = 1,}; + + if (as->volume) + fid.vid = as->volume->vid; + + inode = iget5_locked(sb, fid.vnode, + afs_iget5_pseudo_test, afs_iget5_pseudo_set, &fid); + if (!inode) + return ERR_PTR(-ENOMEM); + + vnode = AFS_FS_I(inode); + + /* there shouldn't be an existing inode */ + if (inode_state_read_once(inode) & I_NEW) { + netfs_inode_init(&vnode->netfs, NULL, false); + simple_inode_init_ts(inode); + set_nlink(inode, 2); + inode->i_size = 0; + inode->i_mode = S_IFDIR | 0555; + inode->i_op = &afs_dynroot_inode_operations; + inode->i_fop = &afs_dynroot_file_operations; + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_blocks = 0; + inode->i_generation = 0; + inode->i_flags |= S_NOATIME; + + set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags); + unlock_new_inode(inode); } - - inode_unlock(root->d_inode); + _leave(" = %p", inode); + return inode; } diff --git a/fs/afs/file.c b/fs/afs/file.c index 323ae9912203..f66a92294284 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS filesystem file handling * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -18,25 +14,30 @@ #include <linux/gfp.h> #include <linux/task_io_accounting_ops.h> #include <linux/mm.h> +#include <linux/swap.h> +#include <linux/netfs.h> +#include <trace/events/netfs.h> #include "internal.h" -static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); -static int afs_readpage(struct file *file, struct page *page); -static void afs_invalidatepage(struct page *page, unsigned int offset, - unsigned int length); -static int afs_releasepage(struct page *page, gfp_t gfp_flags); +static int afs_file_mmap_prepare(struct vm_area_desc *desc); -static int afs_readpages(struct file *filp, struct address_space *mapping, - struct list_head *pages, unsigned nr_pages); +static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter); +static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags); +static void afs_vm_open(struct vm_area_struct *area); +static void afs_vm_close(struct vm_area_struct *area); +static vm_fault_t afs_vm_map_pages(struct vm_fault *vmf, pgoff_t start_pgoff, pgoff_t end_pgoff); const struct file_operations afs_file_operations = { .open = afs_open, .release = afs_release, .llseek = generic_file_llseek, - .read_iter = generic_file_read_iter, - .write_iter = afs_file_write, - .mmap = afs_file_mmap, - .splice_read = generic_file_splice_read, + .read_iter = afs_file_read_iter, + .write_iter = netfs_file_write_iter, + .mmap_prepare = afs_file_mmap_prepare, + .splice_read = afs_file_splice_read, + .splice_write = iter_file_splice_write, .fsync = afs_fsync, .lock = afs_lock, .flock = afs_flock, @@ -46,25 +47,24 @@ const struct inode_operations afs_file_inode_operations = { .getattr = afs_getattr, .setattr = afs_setattr, .permission = afs_permission, - .listxattr = afs_listxattr, }; -const struct address_space_operations afs_fs_aops = { - .readpage = afs_readpage, - .readpages = afs_readpages, - .set_page_dirty = afs_set_page_dirty, - .launder_page = afs_launder_page, - .releasepage = afs_releasepage, - .invalidatepage = afs_invalidatepage, - .write_begin = afs_write_begin, - .write_end = afs_write_end, - .writepage = afs_writepage, +const struct address_space_operations afs_file_aops = { + .direct_IO = noop_direct_IO, + .read_folio = netfs_read_folio, + .readahead = netfs_readahead, + .dirty_folio = netfs_dirty_folio, + .release_folio = netfs_release_folio, + .invalidate_folio = netfs_invalidate_folio, + .migrate_folio = filemap_migrate_folio, .writepages = afs_writepages, }; static const struct vm_operations_struct afs_vm_ops = { + .open = afs_vm_open, + .close = afs_vm_close, .fault = filemap_fault, - .map_pages = filemap_map_pages, + .map_pages = afs_vm_map_pages, .page_mkwrite = afs_page_mkwrite, }; @@ -73,7 +73,7 @@ static const struct vm_operations_struct afs_vm_ops = { */ void afs_put_wb_key(struct afs_wb_key *wbk) { - if (refcount_dec_and_test(&wbk->usage)) { + if (wbk && refcount_dec_and_test(&wbk->usage)) { key_put(wbk->key); kfree(wbk); } @@ -149,7 +149,9 @@ int afs_open(struct inode *inode, struct file *file) if (file->f_flags & O_TRUNC) set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); - + + fscache_use_cookie(afs_vnode_cache(vnode), file->f_mode & FMODE_WRITE); + file->private_data = af; _leave(" = 0"); return 0; @@ -168,508 +170,397 @@ error: */ int afs_release(struct inode *inode, struct file *file) { + struct afs_vnode_cache_aux aux; struct afs_vnode *vnode = AFS_FS_I(inode); struct afs_file *af = file->private_data; + loff_t i_size; + int ret = 0; _enter("{%llx:%llu},", vnode->fid.vid, vnode->fid.vnode); if ((file->f_mode & FMODE_WRITE)) - return vfs_fsync(file, 0); + ret = vfs_fsync(file, 0); file->private_data = NULL; if (af->wb) afs_put_wb_key(af->wb); + + if ((file->f_mode & FMODE_WRITE)) { + i_size = i_size_read(&vnode->netfs.inode); + afs_set_cache_aux(vnode, &aux); + fscache_unuse_cookie(afs_vnode_cache(vnode), &aux, &i_size); + } else { + fscache_unuse_cookie(afs_vnode_cache(vnode), NULL, NULL); + } + key_put(af->key); kfree(af); afs_prune_wb_keys(vnode); - _leave(" = 0"); - return 0; + _leave(" = %d", ret); + return ret; } -/* - * Dispose of a ref to a read record. - */ -void afs_put_read(struct afs_read *req) +static void afs_fetch_data_notify(struct afs_operation *op) { - int i; - - if (refcount_dec_and_test(&req->usage)) { - for (i = 0; i < req->nr_pages; i++) - if (req->pages[i]) - put_page(req->pages[i]); - if (req->pages != req->array) - kfree(req->pages); - kfree(req); - } + struct netfs_io_subrequest *subreq = op->fetch.subreq; + + subreq->error = afs_op_error(op); + netfs_read_subreq_terminated(subreq); } -#ifdef CONFIG_AFS_FSCACHE -/* - * deal with notification that a page was read from the cache - */ -static void afs_file_readpage_read_complete(struct page *page, - void *data, - int error) +static void afs_fetch_data_success(struct afs_operation *op) { - _enter("%p,%p,%d", page, data, error); + struct afs_vnode *vnode = op->file[0].vnode; - /* if the read completes with an error, we just unlock the page and let - * the VM reissue the readpage */ - if (!error) - SetPageUptodate(page); - unlock_page(page); + _enter("op=%08x", op->debug_id); + afs_vnode_commit_status(op, &op->file[0]); + afs_stat_v(vnode, n_fetches); + atomic_long_add(op->fetch.subreq->transferred, &op->net->n_fetch_bytes); + afs_fetch_data_notify(op); } -#endif -/* - * Fetch file data from the volume. - */ -int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *desc) +static void afs_fetch_data_aborted(struct afs_operation *op) { - struct afs_fs_cursor fc; - int ret; - - _enter("%s{%llx:%llu.%u},%x,,,", - vnode->volume->name, - vnode->fid.vid, - vnode->fid.vnode, - vnode->fid.unique, - key_serial(key)); - - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_fetch_data(&fc, desc); - } + afs_check_for_remote_deletion(op); + afs_fetch_data_notify(op); +} - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } +const struct afs_operation_ops afs_fetch_data_operation = { + .issue_afs_rpc = afs_fs_fetch_data, + .issue_yfs_rpc = yfs_fs_fetch_data, + .success = afs_fetch_data_success, + .aborted = afs_fetch_data_aborted, + .failed = afs_fetch_data_notify, +}; - if (ret == 0) { - afs_stat_v(vnode, n_fetches); - atomic_long_add(desc->actual_len, - &afs_v2net(vnode)->n_fetch_bytes); - } +static void afs_issue_read_call(struct afs_operation *op) +{ + op->call_responded = false; + op->call_error = 0; + op->call_abort_code = 0; + if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags)) + yfs_fs_fetch_data(op); + else + afs_fs_fetch_data(op); +} - _leave(" = %d", ret); - return ret; +static void afs_end_read(struct afs_operation *op) +{ + if (op->call_responded && op->server) + set_bit(AFS_SERVER_FL_RESPONDING, &op->server->flags); + + if (!afs_op_error(op)) + afs_fetch_data_success(op); + else if (op->cumul_error.aborted) + afs_fetch_data_aborted(op); + else + afs_fetch_data_notify(op); + + afs_end_vnode_operation(op); + afs_put_operation(op); } /* - * read page from file, directory or symlink, given a key to use + * Perform I/O processing on an asynchronous call. The work item carries a ref + * to the call struct that we either need to release or to pass on. */ -int afs_page_filler(void *data, struct page *page) +static void afs_read_receive(struct afs_call *call) { - struct inode *inode = page->mapping->host; - struct afs_vnode *vnode = AFS_FS_I(inode); - struct afs_read *req; - struct key *key = data; - int ret; - - _enter("{%x},{%lu},{%lu}", key_serial(key), inode->i_ino, page->index); + struct afs_operation *op = call->op; + enum afs_call_state state; - BUG_ON(!PageLocked(page)); - - ret = -ESTALE; - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - goto error; + _enter(""); - /* is it cached? */ -#ifdef CONFIG_AFS_FSCACHE - ret = fscache_read_or_alloc_page(vnode->cache, - page, - afs_file_readpage_read_complete, - NULL, - GFP_KERNEL); -#else - ret = -ENOBUFS; -#endif - switch (ret) { - /* read BIO submitted (page in cache) */ - case 0: - break; + state = READ_ONCE(call->state); + if (state == AFS_CALL_COMPLETE) + return; + trace_afs_read_recv(op, call); - /* page not yet cached */ - case -ENODATA: - _debug("cache said ENODATA"); - goto go_on; + while (state < AFS_CALL_COMPLETE && READ_ONCE(call->need_attention)) { + WRITE_ONCE(call->need_attention, false); + afs_deliver_to_call(call); + state = READ_ONCE(call->state); + } - /* page will not be cached */ - case -ENOBUFS: - _debug("cache said ENOBUFS"); - default: - go_on: - req = kzalloc(sizeof(struct afs_read) + sizeof(struct page *), - GFP_KERNEL); - if (!req) - goto enomem; - - /* We request a full page. If the page is a partial one at the - * end of the file, the server will return a short read and the - * unmarshalling code will clear the unfilled space. - */ - refcount_set(&req->usage, 1); - req->pos = (loff_t)page->index << PAGE_SHIFT; - req->len = PAGE_SIZE; - req->nr_pages = 1; - req->pages = req->array; - req->pages[0] = page; - get_page(page); - - /* read the contents of the file from the server into the - * page */ - ret = afs_fetch_data(vnode, key, req); - afs_put_read(req); - - if (ret < 0) { - if (ret == -ENOENT) { - _debug("got NOENT from server" - " - marking file deleted and stale"); - set_bit(AFS_VNODE_DELETED, &vnode->flags); - ret = -ESTALE; - } - -#ifdef CONFIG_AFS_FSCACHE - fscache_uncache_page(vnode->cache, page); -#endif - BUG_ON(PageFsCache(page)); - - if (ret == -EINTR || - ret == -ENOMEM || - ret == -ERESTARTSYS || - ret == -EAGAIN) - goto error; - goto io_error; - } + if (state < AFS_CALL_COMPLETE) { + netfs_read_subreq_progress(op->fetch.subreq); + if (rxrpc_kernel_check_life(call->net->socket, call->rxcall)) + return; + /* rxrpc terminated the call. */ + afs_set_call_complete(call, call->error, call->abort_code); + } - SetPageUptodate(page); + op->call_abort_code = call->abort_code; + op->call_error = call->error; + op->call_responded = call->responded; + op->call = NULL; + call->op = NULL; + afs_put_call(call); - /* send the page to the cache */ -#ifdef CONFIG_AFS_FSCACHE - if (PageFsCache(page) && - fscache_write_page(vnode->cache, page, vnode->status.size, - GFP_KERNEL) != 0) { - fscache_uncache_page(vnode->cache, page); - BUG_ON(PageFsCache(page)); - } -#endif - unlock_page(page); + /* If the call failed, then we need to crank the server rotation + * handle and try the next. + */ + if (afs_select_fileserver(op)) { + afs_issue_read_call(op); + return; } - _leave(" = 0"); - return 0; - -io_error: - SetPageError(page); - goto error; -enomem: - ret = -ENOMEM; -error: - unlock_page(page); - _leave(" = %d", ret); - return ret; + afs_end_read(op); } -/* - * read page from file, directory or symlink, given a file to nominate the key - * to be used - */ -static int afs_readpage(struct file *file, struct page *page) +void afs_fetch_data_async_rx(struct work_struct *work) { - struct key *key; - int ret; + struct afs_call *call = container_of(work, struct afs_call, async_work); - if (file) { - key = afs_file_key(file); - ASSERT(key != NULL); - ret = afs_page_filler(key, page); - } else { - struct inode *inode = page->mapping->host; - key = afs_request_key(AFS_FS_S(inode->i_sb)->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - } else { - ret = afs_page_filler(key, page); - key_put(key); - } - } - return ret; + afs_read_receive(call); + afs_put_call(call); } -/* - * Make pages available as they're filled. - */ -static void afs_readpages_page_done(struct afs_call *call, struct afs_read *req) +void afs_fetch_data_immediate_cancel(struct afs_call *call) { -#ifdef CONFIG_AFS_FSCACHE - struct afs_vnode *vnode = call->reply[0]; -#endif - struct page *page = req->pages[req->index]; - - req->pages[req->index] = NULL; - SetPageUptodate(page); - - /* send the page to the cache */ -#ifdef CONFIG_AFS_FSCACHE - if (PageFsCache(page) && - fscache_write_page(vnode->cache, page, vnode->status.size, - GFP_KERNEL) != 0) { - fscache_uncache_page(vnode->cache, page); - BUG_ON(PageFsCache(page)); + if (call->async) { + afs_get_call(call, afs_call_trace_wake); + if (!queue_work(afs_async_calls, &call->async_work)) + afs_deferred_put_call(call); + flush_work(&call->async_work); } -#endif - unlock_page(page); - put_page(page); } /* - * Read a contiguous set of pages. + * Fetch file data from the volume. */ -static int afs_readpages_one(struct file *file, struct address_space *mapping, - struct list_head *pages) +static void afs_issue_read(struct netfs_io_subrequest *subreq) { - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - struct afs_read *req; - struct list_head *p; - struct page *first, *page; - struct key *key = afs_file_key(file); - pgoff_t index; - int ret, n, i; - - /* Count the number of contiguous pages at the front of the list. Note - * that the list goes prev-wards rather than next-wards. - */ - first = lru_to_page(pages); - index = first->index + 1; - n = 1; - for (p = first->lru.prev; p != pages; p = p->prev) { - page = list_entry(p, struct page, lru); - if (page->index != index) - break; - index++; - n++; + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); + struct key *key = subreq->rreq->netfs_priv; + + _enter("%s{%llx:%llu.%u},%x,,,", + vnode->volume->name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + op = afs_alloc_operation(key, vnode->volume); + if (IS_ERR(op)) { + subreq->error = PTR_ERR(op); + netfs_read_subreq_terminated(subreq); + return; } - req = kzalloc(sizeof(struct afs_read) + sizeof(struct page *) * n, - GFP_NOFS); - if (!req) - return -ENOMEM; + afs_op_set_vnode(op, 0, vnode); - refcount_set(&req->usage, 1); - req->page_done = afs_readpages_page_done; - req->pos = first->index; - req->pos <<= PAGE_SHIFT; - req->pages = req->array; - - /* Transfer the pages to the request. We add them in until one fails - * to add to the LRU and then we stop (as that'll make a hole in the - * contiguous run. - * - * Note that it's possible for the file size to change whilst we're - * doing this, but we rely on the server returning less than we asked - * for if the file shrank. We also rely on this to deal with a partial - * page at the end of the file. - */ - do { - page = lru_to_page(pages); - list_del(&page->lru); - index = page->index; - if (add_to_page_cache_lru(page, mapping, index, - readahead_gfp_mask(mapping))) { -#ifdef CONFIG_AFS_FSCACHE - fscache_uncache_page(vnode->cache, page); -#endif - put_page(page); - break; + op->fetch.subreq = subreq; + op->ops = &afs_fetch_data_operation; + + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + + if (subreq->rreq->origin == NETFS_READAHEAD || + subreq->rreq->iocb) { + op->flags |= AFS_OPERATION_ASYNC; + + if (!afs_begin_vnode_operation(op)) { + subreq->error = afs_put_operation(op); + netfs_read_subreq_terminated(subreq); + return; } - req->pages[req->nr_pages++] = page; - req->len += PAGE_SIZE; - } while (req->nr_pages < n); + if (!afs_select_fileserver(op)) { + afs_end_read(op); + return; + } - if (req->nr_pages == 0) { - kfree(req); - return 0; + afs_issue_read_call(op); + } else { + afs_do_sync_operation(op); } +} - ret = afs_fetch_data(vnode, key, req); - if (ret < 0) - goto error; +static int afs_init_request(struct netfs_io_request *rreq, struct file *file) +{ + struct afs_vnode *vnode = AFS_FS_I(rreq->inode); - task_io_account_read(PAGE_SIZE * req->nr_pages); - afs_put_read(req); - return 0; + if (file) + rreq->netfs_priv = key_get(afs_file_key(file)); + rreq->rsize = 256 * 1024; + rreq->wsize = 256 * 1024 * 1024; -error: - if (ret == -ENOENT) { - _debug("got NOENT from server" - " - marking file deleted and stale"); - set_bit(AFS_VNODE_DELETED, &vnode->flags); - ret = -ESTALE; - } + switch (rreq->origin) { + case NETFS_READ_SINGLE: + if (!file) { + struct key *key = afs_request_key(vnode->volume->cell); - for (i = 0; i < req->nr_pages; i++) { - page = req->pages[i]; - if (page) { -#ifdef CONFIG_AFS_FSCACHE - fscache_uncache_page(vnode->cache, page); -#endif - SetPageError(page); - unlock_page(page); + if (IS_ERR(key)) + return PTR_ERR(key); + rreq->netfs_priv = key; } + break; + case NETFS_WRITEBACK: + case NETFS_WRITETHROUGH: + case NETFS_UNBUFFERED_WRITE: + case NETFS_DIO_WRITE: + if (S_ISREG(rreq->inode->i_mode)) + rreq->io_streams[0].avail = true; + break; + case NETFS_WRITEBACK_SINGLE: + default: + break; } - - afs_put_read(req); - return ret; + return 0; } -/* - * read a set of pages - */ -static int afs_readpages(struct file *file, struct address_space *mapping, - struct list_head *pages, unsigned nr_pages) +static int afs_check_write_begin(struct file *file, loff_t pos, unsigned len, + struct folio **foliop, void **_fsdata) { - struct key *key = afs_file_key(file); - struct afs_vnode *vnode; - int ret = 0; + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); - _enter("{%d},{%lu},,%d", - key_serial(key), mapping->host->i_ino, nr_pages); + return test_bit(AFS_VNODE_DELETED, &vnode->flags) ? -ESTALE : 0; +} + +static void afs_free_request(struct netfs_io_request *rreq) +{ + key_put(rreq->netfs_priv); + afs_put_wb_key(rreq->netfs_priv2); +} - ASSERT(key != NULL); +static void afs_update_i_size(struct inode *inode, loff_t new_i_size) +{ + struct afs_vnode *vnode = AFS_FS_I(inode); + loff_t i_size; - vnode = AFS_FS_I(mapping->host); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { - _leave(" = -ESTALE"); - return -ESTALE; + write_seqlock(&vnode->cb_lock); + i_size = i_size_read(&vnode->netfs.inode); + if (new_i_size > i_size) { + i_size_write(&vnode->netfs.inode, new_i_size); + inode_set_bytes(&vnode->netfs.inode, new_i_size); } + write_sequnlock(&vnode->cb_lock); + fscache_update_cookie(afs_vnode_cache(vnode), NULL, &new_i_size); +} - /* attempt to read as many of the pages as possible */ -#ifdef CONFIG_AFS_FSCACHE - ret = fscache_read_or_alloc_pages(vnode->cache, - mapping, - pages, - &nr_pages, - afs_file_readpage_read_complete, - NULL, - mapping_gfp_mask(mapping)); -#else - ret = -ENOBUFS; -#endif - - switch (ret) { - /* all pages are being read from the cache */ - case 0: - BUG_ON(!list_empty(pages)); - BUG_ON(nr_pages != 0); - _leave(" = 0 [reading all]"); - return 0; - - /* there were pages that couldn't be read from the cache */ - case -ENODATA: - case -ENOBUFS: - break; +static void afs_netfs_invalidate_cache(struct netfs_io_request *wreq) +{ + struct afs_vnode *vnode = AFS_FS_I(wreq->inode); - /* other error */ - default: - _leave(" = %d", ret); - return ret; - } + afs_invalidate_cache(vnode, 0); +} - while (!list_empty(pages)) { - ret = afs_readpages_one(file, mapping, pages); - if (ret < 0) - break; +const struct netfs_request_ops afs_req_ops = { + .init_request = afs_init_request, + .free_request = afs_free_request, + .check_write_begin = afs_check_write_begin, + .issue_read = afs_issue_read, + .update_i_size = afs_update_i_size, + .invalidate_cache = afs_netfs_invalidate_cache, + .begin_writeback = afs_begin_writeback, + .prepare_write = afs_prepare_write, + .issue_write = afs_issue_write, + .retry_request = afs_retry_request, +}; + +static void afs_add_open_mmap(struct afs_vnode *vnode) +{ + if (atomic_inc_return(&vnode->cb_nr_mmap) == 1) { + down_write(&vnode->volume->open_mmaps_lock); + + if (list_empty(&vnode->cb_mmap_link)) + list_add_tail(&vnode->cb_mmap_link, &vnode->volume->open_mmaps); + + up_write(&vnode->volume->open_mmaps_lock); } +} - _leave(" = %d [netting]", ret); - return ret; +static void afs_drop_open_mmap(struct afs_vnode *vnode) +{ + if (atomic_add_unless(&vnode->cb_nr_mmap, -1, 1)) + return; + + down_write(&vnode->volume->open_mmaps_lock); + + read_seqlock_excl(&vnode->cb_lock); + // the only place where ->cb_nr_mmap may hit 0 + // see __afs_break_callback() for the other side... + if (atomic_dec_and_test(&vnode->cb_nr_mmap)) + list_del_init(&vnode->cb_mmap_link); + read_sequnlock_excl(&vnode->cb_lock); + + up_write(&vnode->volume->open_mmaps_lock); + flush_work(&vnode->cb_work); } /* - * invalidate part or all of a page - * - release a page and clean up its private data if offset is 0 (indicating - * the entire page) + * Handle setting up a memory mapping on an AFS file. */ -static void afs_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) +static int afs_file_mmap_prepare(struct vm_area_desc *desc) { - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - unsigned long priv; + struct afs_vnode *vnode = AFS_FS_I(file_inode(desc->file)); + int ret; - _enter("{%lu},%u,%u", page->index, offset, length); + afs_add_open_mmap(vnode); - BUG_ON(!PageLocked(page)); + ret = generic_file_mmap_prepare(desc); + if (ret == 0) + desc->vm_ops = &afs_vm_ops; + else + afs_drop_open_mmap(vnode); + return ret; +} - /* we clean up only if the entire page is being invalidated */ - if (offset == 0 && length == PAGE_SIZE) { -#ifdef CONFIG_AFS_FSCACHE - if (PageFsCache(page)) { - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - fscache_wait_on_page_write(vnode->cache, page); - fscache_uncache_page(vnode->cache, page); - } -#endif - - if (PagePrivate(page)) { - priv = page_private(page); - trace_afs_page_dirty(vnode, tracepoint_string("inval"), - page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); - } - } +static void afs_vm_open(struct vm_area_struct *vma) +{ + afs_add_open_mmap(AFS_FS_I(file_inode(vma->vm_file))); +} - _leave(""); +static void afs_vm_close(struct vm_area_struct *vma) +{ + afs_drop_open_mmap(AFS_FS_I(file_inode(vma->vm_file))); } -/* - * release a page and clean up its private state if it's not busy - * - return true if the page can now be released, false if not - */ -static int afs_releasepage(struct page *page, gfp_t gfp_flags) +static vm_fault_t afs_vm_map_pages(struct vm_fault *vmf, pgoff_t start_pgoff, pgoff_t end_pgoff) { - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - unsigned long priv; - - _enter("{{%llx:%llu}[%lu],%lx},%x", - vnode->fid.vid, vnode->fid.vnode, page->index, page->flags, - gfp_flags); - - /* deny if page is being written to the cache and the caller hasn't - * elected to wait */ -#ifdef CONFIG_AFS_FSCACHE - if (!fscache_maybe_release_page(vnode->cache, page, gfp_flags)) { - _leave(" = F [cache busy]"); - return 0; - } -#endif - - if (PagePrivate(page)) { - priv = page_private(page); - trace_afs_page_dirty(vnode, tracepoint_string("rel"), - page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); - } + struct afs_vnode *vnode = AFS_FS_I(file_inode(vmf->vma->vm_file)); - /* indicate that the page can be released */ - _leave(" = T"); - return 1; + if (afs_check_validity(vnode)) + return filemap_map_pages(vmf, start_pgoff, end_pgoff); + return 0; } -/* - * Handle setting up a memory mapping on an AFS file. - */ -static int afs_file_mmap(struct file *file, struct vm_area_struct *vma) +static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) { - int ret; + struct inode *inode = file_inode(iocb->ki_filp); + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_file *af = iocb->ki_filp->private_data; + ssize_t ret; + + if (iocb->ki_flags & IOCB_DIRECT) + return netfs_unbuffered_read_iter(iocb, iter); + + ret = netfs_start_io_read(inode); + if (ret < 0) + return ret; + ret = afs_validate(vnode, af->key); + if (ret == 0) + ret = filemap_read(iocb, iter, 0); + netfs_end_io_read(inode); + return ret; +} - ret = generic_file_mmap(file, vma); +static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct inode *inode = file_inode(in); + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_file *af = in->private_data; + ssize_t ret; + + ret = netfs_start_io_read(inode); + if (ret < 0) + return ret; + ret = afs_validate(vnode, af->key); if (ret == 0) - vma->vm_ops = &afs_vm_ops; + ret = filemap_splice_read(in, ppos, pipe, len, flags); + netfs_end_io_read(inode); return ret; } diff --git a/fs/afs/flock.c b/fs/afs/flock.c index e432bd27a2e7..f0e96a35093f 100644 --- a/fs/afs/flock.c +++ b/fs/afs/flock.c @@ -1,21 +1,19 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS file locking support * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include "internal.h" #define AFS_LOCK_GRANTED 0 #define AFS_LOCK_PENDING 1 +#define AFS_LOCK_YOUR_TRY 2 struct workqueue_struct *afs_lock_manager; +static void afs_next_locker(struct afs_vnode *vnode, int error); static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl); static void afs_fl_release_private(struct file_lock *fl); @@ -24,6 +22,14 @@ static const struct file_lock_operations afs_lock_ops = { .fl_release_private = afs_fl_release_private, }; +static inline void afs_set_lock_state(struct afs_vnode *vnode, enum afs_lock_state state) +{ + _debug("STATE %u -> %u", vnode->lock_state, state); + vnode->lock_state = state; +} + +static atomic_t afs_file_lock_debug_id; + /* * if the callback is broken on this vnode, then the lock may now be available */ @@ -31,7 +37,11 @@ void afs_lock_may_be_available(struct afs_vnode *vnode) { _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); - queue_delayed_work(afs_lock_manager, &vnode->lock_work, 0); + spin_lock(&vnode->lock); + if (vnode->lock_state == AFS_VNODE_LOCK_WAITING_FOR_CB) + afs_next_locker(vnode, 0); + trace_afs_flock_ev(vnode, NULL, afs_flock_callback_break, 0); + spin_unlock(&vnode->lock); } /* @@ -40,8 +50,36 @@ void afs_lock_may_be_available(struct afs_vnode *vnode) */ static void afs_schedule_lock_extension(struct afs_vnode *vnode) { - queue_delayed_work(afs_lock_manager, &vnode->lock_work, - AFS_LOCKWAIT * HZ / 2); + ktime_t expires_at, now, duration; + u64 duration_j; + + expires_at = ktime_add_ms(vnode->locked_at, AFS_LOCKWAIT * 1000 / 2); + now = ktime_get_real(); + duration = ktime_sub(expires_at, now); + if (duration <= 0) + duration_j = 0; + else + duration_j = nsecs_to_jiffies(ktime_to_ns(duration)); + + queue_delayed_work(afs_lock_manager, &vnode->lock_work, duration_j); +} + +/* + * In the case of successful completion of a lock operation, record the time + * the reply appeared and start the lock extension timer. + */ +void afs_lock_op_done(struct afs_call *call) +{ + struct afs_operation *op = call->op; + struct afs_vnode *vnode = op->file[0].vnode; + + if (call->error == 0) { + spin_lock(&vnode->lock); + trace_afs_flock_ev(vnode, NULL, afs_flock_timestamp, 0); + vnode->locked_at = call->issue_time; + afs_schedule_lock_extension(vnode); + spin_unlock(&vnode->lock); + } } /* @@ -49,32 +87,111 @@ static void afs_schedule_lock_extension(struct afs_vnode *vnode) * first lock in the queue is itself a readlock) * - the caller must hold the vnode lock */ -static void afs_grant_locks(struct afs_vnode *vnode, struct file_lock *fl) +static void afs_grant_locks(struct afs_vnode *vnode) { struct file_lock *p, *_p; + bool exclusive = (vnode->lock_type == AFS_LOCK_WRITE); - list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); - if (fl->fl_type == F_RDLCK) { - list_for_each_entry_safe(p, _p, &vnode->pending_locks, - fl_u.afs.link) { - if (p->fl_type == F_RDLCK) { - p->fl_u.afs.state = AFS_LOCK_GRANTED; - list_move_tail(&p->fl_u.afs.link, - &vnode->granted_locks); - wake_up(&p->fl_wait); - } + list_for_each_entry_safe(p, _p, &vnode->pending_locks, fl_u.afs.link) { + if (!exclusive && lock_is_write(p)) + continue; + + list_move_tail(&p->fl_u.afs.link, &vnode->granted_locks); + p->fl_u.afs.state = AFS_LOCK_GRANTED; + trace_afs_flock_op(vnode, p, afs_flock_op_grant); + locks_wake_up(p); + } +} + +/* + * If an error is specified, reject every pending lock that matches the + * authentication and type of the lock we failed to get. If there are any + * remaining lockers, try to wake up one of them to have a go. + */ +static void afs_next_locker(struct afs_vnode *vnode, int error) +{ + struct file_lock *p, *_p, *next = NULL; + struct key *key = vnode->lock_key; + unsigned int type = F_RDLCK; + + _enter(""); + + if (vnode->lock_type == AFS_LOCK_WRITE) + type = F_WRLCK; + + list_for_each_entry_safe(p, _p, &vnode->pending_locks, fl_u.afs.link) { + if (error && + p->c.flc_type == type && + afs_file_key(p->c.flc_file) == key) { + list_del_init(&p->fl_u.afs.link); + p->fl_u.afs.state = error; + locks_wake_up(p); } + + /* Select the next locker to hand off to. */ + if (next && (lock_is_write(next) || lock_is_read(p))) + continue; + next = p; + } + + vnode->lock_key = NULL; + key_put(key); + + if (next) { + afs_set_lock_state(vnode, AFS_VNODE_LOCK_SETTING); + next->fl_u.afs.state = AFS_LOCK_YOUR_TRY; + trace_afs_flock_op(vnode, next, afs_flock_op_wake); + locks_wake_up(next); + } else { + afs_set_lock_state(vnode, AFS_VNODE_LOCK_NONE); + trace_afs_flock_ev(vnode, NULL, afs_flock_no_lockers, 0); + } + + _leave(""); +} + +/* + * Kill off all waiters in the the pending lock queue due to the vnode being + * deleted. + */ +static void afs_kill_lockers_enoent(struct afs_vnode *vnode) +{ + struct file_lock *p; + + afs_set_lock_state(vnode, AFS_VNODE_LOCK_DELETED); + + while (!list_empty(&vnode->pending_locks)) { + p = list_entry(vnode->pending_locks.next, + struct file_lock, fl_u.afs.link); + list_del_init(&p->fl_u.afs.link); + p->fl_u.afs.state = -ENOENT; + locks_wake_up(p); } + + key_put(vnode->lock_key); + vnode->lock_key = NULL; } +static void afs_lock_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + afs_vnode_commit_status(op, &op->file[0]); +} + +static const struct afs_operation_ops afs_set_lock_operation = { + .issue_afs_rpc = afs_fs_set_lock, + .issue_yfs_rpc = yfs_fs_set_lock, + .success = afs_lock_success, + .aborted = afs_check_for_remote_deletion, +}; + /* * Get a lock on a file */ static int afs_set_lock(struct afs_vnode *vnode, struct key *key, afs_lock_type_t type) { - struct afs_fs_cursor fc; - int ret; + struct afs_operation *op; _enter("%s{%llx:%llu.%u},%x,%u", vnode->volume->name, @@ -83,29 +200,29 @@ static int afs_set_lock(struct afs_vnode *vnode, struct key *key, vnode->fid.unique, key_serial(key), type); - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_set_lock(&fc, type); - } + op = afs_alloc_operation(key, vnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } + afs_op_set_vnode(op, 0, vnode); - _leave(" = %d", ret); - return ret; + op->lock.type = type; + op->ops = &afs_set_lock_operation; + return afs_do_sync_operation(op); } +static const struct afs_operation_ops afs_extend_lock_operation = { + .issue_afs_rpc = afs_fs_extend_lock, + .issue_yfs_rpc = yfs_fs_extend_lock, + .success = afs_lock_success, +}; + /* * Extend a lock on a file */ static int afs_extend_lock(struct afs_vnode *vnode, struct key *key) { - struct afs_fs_cursor fc; - int ret; + struct afs_operation *op; _enter("%s{%llx:%llu.%u},%x", vnode->volume->name, @@ -114,29 +231,29 @@ static int afs_extend_lock(struct afs_vnode *vnode, struct key *key) vnode->fid.unique, key_serial(key)); - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_current_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_extend_lock(&fc); - } + op = afs_alloc_operation(key, vnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } + afs_op_set_vnode(op, 0, vnode); - _leave(" = %d", ret); - return ret; + op->flags |= AFS_OPERATION_UNINTR; + op->ops = &afs_extend_lock_operation; + return afs_do_sync_operation(op); } +static const struct afs_operation_ops afs_release_lock_operation = { + .issue_afs_rpc = afs_fs_release_lock, + .issue_yfs_rpc = yfs_fs_release_lock, + .success = afs_lock_success, +}; + /* * Release a lock on a file */ static int afs_release_lock(struct afs_vnode *vnode, struct key *key) { - struct afs_fs_cursor fc; - int ret; + struct afs_operation *op; _enter("%s{%llx:%llu.%u},%x", vnode->volume->name, @@ -145,20 +262,15 @@ static int afs_release_lock(struct afs_vnode *vnode, struct key *key) vnode->fid.unique, key_serial(key)); - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_current_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_release_lock(&fc); - } + op = afs_alloc_operation(key, vnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } + afs_op_set_vnode(op, 0, vnode); - _leave(" = %d", ret); - return ret; + op->flags |= AFS_OPERATION_UNINTR; + op->ops = &afs_release_lock_operation; + return afs_do_sync_operation(op); } /* @@ -170,8 +282,6 @@ void afs_lock_work(struct work_struct *work) { struct afs_vnode *vnode = container_of(work, struct afs_vnode, lock_work.work); - struct file_lock *fl, *next; - afs_lock_type_t type; struct key *key; int ret; @@ -183,35 +293,28 @@ again: _debug("wstate %u for %p", vnode->lock_state, vnode); switch (vnode->lock_state) { case AFS_VNODE_LOCK_NEED_UNLOCK: - _debug("unlock"); - vnode->lock_state = AFS_VNODE_LOCK_UNLOCKING; + afs_set_lock_state(vnode, AFS_VNODE_LOCK_UNLOCKING); + trace_afs_flock_ev(vnode, NULL, afs_flock_work_unlocking, 0); spin_unlock(&vnode->lock); /* attempt to release the server lock; if it fails, we just * wait 5 minutes and it'll expire anyway */ ret = afs_release_lock(vnode, vnode->lock_key); - if (ret < 0) + if (ret < 0 && vnode->lock_state != AFS_VNODE_LOCK_DELETED) { + trace_afs_flock_ev(vnode, NULL, afs_flock_release_fail, + ret); printk(KERN_WARNING "AFS:" " Failed to release lock on {%llx:%llx} error %d\n", vnode->fid.vid, vnode->fid.vnode, ret); - - spin_lock(&vnode->lock); - key_put(vnode->lock_key); - vnode->lock_key = NULL; - vnode->lock_state = AFS_VNODE_LOCK_NONE; - - if (list_empty(&vnode->pending_locks)) { - spin_unlock(&vnode->lock); - return; } - /* The new front of the queue now owns the state variables. */ - next = list_entry(vnode->pending_locks.next, - struct file_lock, fl_u.afs.link); - vnode->lock_key = key_get(afs_file_key(next->fl_file)); - vnode->lock_type = (next->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; - vnode->lock_state = AFS_VNODE_LOCK_WAITING_FOR_CB; - goto again; + spin_lock(&vnode->lock); + if (ret == -ENOENT) + afs_kill_lockers_enoent(vnode); + else + afs_next_locker(vnode, 0); + spin_unlock(&vnode->lock); + return; /* If we've already got a lock, then it must be time to extend that * lock as AFS locks time out after 5 minutes. @@ -222,86 +325,55 @@ again: ASSERT(!list_empty(&vnode->granted_locks)); key = key_get(vnode->lock_key); - vnode->lock_state = AFS_VNODE_LOCK_EXTENDING; + afs_set_lock_state(vnode, AFS_VNODE_LOCK_EXTENDING); + trace_afs_flock_ev(vnode, NULL, afs_flock_work_extending, 0); spin_unlock(&vnode->lock); ret = afs_extend_lock(vnode, key); /* RPC */ key_put(key); - if (ret < 0) - pr_warning("AFS: Failed to extend lock on {%llx:%llx} error %d\n", - vnode->fid.vid, vnode->fid.vnode, ret); + if (ret < 0) { + trace_afs_flock_ev(vnode, NULL, afs_flock_extend_fail, + ret); + pr_warn("AFS: Failed to extend lock on {%llx:%llx} error %d\n", + vnode->fid.vid, vnode->fid.vnode, ret); + } spin_lock(&vnode->lock); + if (ret == -ENOENT) { + afs_kill_lockers_enoent(vnode); + spin_unlock(&vnode->lock); + return; + } + if (vnode->lock_state != AFS_VNODE_LOCK_EXTENDING) goto again; - vnode->lock_state = AFS_VNODE_LOCK_GRANTED; + afs_set_lock_state(vnode, AFS_VNODE_LOCK_GRANTED); - if (ret == 0) - afs_schedule_lock_extension(vnode); - else + if (ret != 0) queue_delayed_work(afs_lock_manager, &vnode->lock_work, HZ * 10); spin_unlock(&vnode->lock); _leave(" [ext]"); return; - /* If we don't have a granted lock, then we must've been called - * back by the server, and so if might be possible to get a - * lock we're currently waiting for. - */ + /* If we're waiting for a callback to indicate lock release, we can't + * actually rely on this, so need to recheck at regular intervals. The + * problem is that the server might not notify us if the lock just + * expires (say because a client died) rather than being explicitly + * released. + */ case AFS_VNODE_LOCK_WAITING_FOR_CB: - _debug("get"); - - key = key_get(vnode->lock_key); - type = vnode->lock_type; - vnode->lock_state = AFS_VNODE_LOCK_SETTING; + _debug("retry"); + afs_next_locker(vnode, 0); spin_unlock(&vnode->lock); + return; - ret = afs_set_lock(vnode, key, type); /* RPC */ - key_put(key); - - spin_lock(&vnode->lock); - switch (ret) { - case -EWOULDBLOCK: - _debug("blocked"); - break; - case 0: - _debug("acquired"); - vnode->lock_state = AFS_VNODE_LOCK_GRANTED; - /* Fall through */ - default: - /* Pass the lock or the error onto the first locker in - * the list - if they're looking for this type of lock. - * If they're not, we assume that whoever asked for it - * took a signal. - */ - if (list_empty(&vnode->pending_locks)) { - _debug("withdrawn"); - vnode->lock_state = AFS_VNODE_LOCK_NEED_UNLOCK; - goto again; - } - - fl = list_entry(vnode->pending_locks.next, - struct file_lock, fl_u.afs.link); - type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; - if (vnode->lock_type != type) { - _debug("changed"); - vnode->lock_state = AFS_VNODE_LOCK_NEED_UNLOCK; - goto again; - } - - fl->fl_u.afs.state = ret; - if (ret == 0) - afs_grant_locks(vnode, fl); - else - list_del_init(&fl->fl_u.afs.link); - wake_up(&fl->fl_wait); - spin_unlock(&vnode->lock); - _leave(" [granted]"); - return; - } + case AFS_VNODE_LOCK_DELETED: + afs_kill_lockers_enoent(vnode); + spin_unlock(&vnode->lock); + return; default: /* Looks like a lock request was withdrawn. */ @@ -319,14 +391,16 @@ again: */ static void afs_defer_unlock(struct afs_vnode *vnode) { - _enter(""); + _enter("%u", vnode->lock_state); - if (vnode->lock_state == AFS_VNODE_LOCK_GRANTED || - vnode->lock_state == AFS_VNODE_LOCK_EXTENDING) { + if (list_empty(&vnode->granted_locks) && + (vnode->lock_state == AFS_VNODE_LOCK_GRANTED || + vnode->lock_state == AFS_VNODE_LOCK_EXTENDING)) { cancel_delayed_work(&vnode->lock_work); - vnode->lock_state = AFS_VNODE_LOCK_NEED_UNLOCK; - afs_lock_may_be_available(vnode); + afs_set_lock_state(vnode, AFS_VNODE_LOCK_NEED_UNLOCK); + trace_afs_flock_ev(vnode, NULL, afs_flock_defer_unlock, 0); + queue_delayed_work(afs_lock_manager, &vnode->lock_work, 0); } } @@ -335,7 +409,7 @@ static void afs_defer_unlock(struct afs_vnode *vnode) * whether we think that we have a locking permit. */ static int afs_do_setlk_check(struct afs_vnode *vnode, struct key *key, - afs_lock_type_t type, bool can_sleep) + enum afs_flock_mode mode, afs_lock_type_t type) { afs_access_t access; int ret; @@ -363,160 +437,177 @@ static int afs_do_setlk_check(struct afs_vnode *vnode, struct key *key, if (type == AFS_LOCK_READ) { if (!(access & (AFS_ACE_INSERT | AFS_ACE_WRITE | AFS_ACE_LOCK))) return -EACCES; - if (vnode->status.lock_count == -1 && !can_sleep) - return -EAGAIN; /* Write locked */ } else { if (!(access & (AFS_ACE_INSERT | AFS_ACE_WRITE))) return -EACCES; - if (vnode->status.lock_count != 0 && !can_sleep) - return -EAGAIN; /* Locked */ } return 0; } /* - * Remove the front runner from the pending queue. - * - The caller must hold vnode->lock. - */ -static void afs_dequeue_lock(struct afs_vnode *vnode, struct file_lock *fl) -{ - struct file_lock *next; - - _enter(""); - - /* ->lock_type, ->lock_key and ->lock_state only belong to this - * file_lock if we're at the front of the pending queue or if we have - * the lock granted or if the lock_state is NEED_UNLOCK or UNLOCKING. - */ - if (vnode->granted_locks.next == &fl->fl_u.afs.link && - vnode->granted_locks.prev == &fl->fl_u.afs.link) { - list_del_init(&fl->fl_u.afs.link); - afs_defer_unlock(vnode); - return; - } - - if (!list_empty(&vnode->granted_locks) || - vnode->pending_locks.next != &fl->fl_u.afs.link) { - list_del_init(&fl->fl_u.afs.link); - return; - } - - list_del_init(&fl->fl_u.afs.link); - key_put(vnode->lock_key); - vnode->lock_key = NULL; - vnode->lock_state = AFS_VNODE_LOCK_NONE; - - if (list_empty(&vnode->pending_locks)) - return; - - /* The new front of the queue now owns the state variables. */ - next = list_entry(vnode->pending_locks.next, - struct file_lock, fl_u.afs.link); - vnode->lock_key = key_get(afs_file_key(next->fl_file)); - vnode->lock_type = (next->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; - vnode->lock_state = AFS_VNODE_LOCK_WAITING_FOR_CB; - afs_lock_may_be_available(vnode); -} - -/* * request a lock on a file on the server */ static int afs_do_setlk(struct file *file, struct file_lock *fl) { - struct inode *inode = locks_inode(file); + struct inode *inode = file_inode(file); struct afs_vnode *vnode = AFS_FS_I(inode); + enum afs_flock_mode mode = AFS_FS_S(inode->i_sb)->flock_mode; afs_lock_type_t type; struct key *key = afs_file_key(file); + bool partial, no_server_lock = false; int ret; - _enter("{%llx:%llu},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type); + if (mode == afs_flock_mode_unset) + mode = afs_flock_mode_openafs; - /* only whole-file locks are supported */ - if (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX) - return -EINVAL; + _enter("{%llx:%llu},%llu-%llu,%u,%u", + vnode->fid.vid, vnode->fid.vnode, + fl->fl_start, fl->fl_end, fl->c.flc_type, mode); fl->fl_ops = &afs_lock_ops; INIT_LIST_HEAD(&fl->fl_u.afs.link); fl->fl_u.afs.state = AFS_LOCK_PENDING; - type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; + partial = (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX); + type = lock_is_read(fl) ? AFS_LOCK_READ : AFS_LOCK_WRITE; + if (mode == afs_flock_mode_write && partial) + type = AFS_LOCK_WRITE; - ret = afs_do_setlk_check(vnode, key, type, fl->fl_flags & FL_SLEEP); + ret = afs_do_setlk_check(vnode, key, mode, type); if (ret < 0) return ret; - spin_lock(&vnode->lock); + trace_afs_flock_op(vnode, fl, afs_flock_op_set_lock); - /* If we've already got a readlock on the server then we instantly - * grant another readlock, irrespective of whether there are any - * pending writelocks. + /* AFS3 protocol only supports full-file locks and doesn't provide any + * method of upgrade/downgrade, so we need to emulate for partial-file + * locks. + * + * The OpenAFS client only gets a server lock for a full-file lock and + * keeps partial-file locks local. Allow this behaviour to be emulated + * (as the default). */ - if (type == AFS_LOCK_READ && - vnode->lock_state == AFS_VNODE_LOCK_GRANTED && - vnode->lock_type == AFS_LOCK_READ) { - _debug("instant readlock"); - ASSERT(!list_empty(&vnode->granted_locks)); - goto share_existing_lock; + if (mode == afs_flock_mode_local || + (partial && mode == afs_flock_mode_openafs)) { + no_server_lock = true; + goto skip_server_lock; } + spin_lock(&vnode->lock); list_add_tail(&fl->fl_u.afs.link, &vnode->pending_locks); + ret = -ENOENT; + if (vnode->lock_state == AFS_VNODE_LOCK_DELETED) + goto error_unlock; + + /* If we've already got a lock on the server then try to move to having + * the VFS grant the requested lock. Note that this means that other + * clients may get starved out. + */ + _debug("try %u", vnode->lock_state); + if (vnode->lock_state == AFS_VNODE_LOCK_GRANTED) { + if (type == AFS_LOCK_READ) { + _debug("instant readlock"); + list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); + fl->fl_u.afs.state = AFS_LOCK_GRANTED; + goto vnode_is_locked_u; + } + + if (vnode->lock_type == AFS_LOCK_WRITE) { + _debug("instant writelock"); + list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); + fl->fl_u.afs.state = AFS_LOCK_GRANTED; + goto vnode_is_locked_u; + } + } + + if (vnode->lock_state == AFS_VNODE_LOCK_NONE && + !(fl->c.flc_flags & FL_SLEEP)) { + ret = -EAGAIN; + if (type == AFS_LOCK_READ) { + if (vnode->status.lock_count == -1) + goto lock_is_contended; /* Write locked */ + } else { + if (vnode->status.lock_count != 0) + goto lock_is_contended; /* Locked */ + } + } + if (vnode->lock_state != AFS_VNODE_LOCK_NONE) goto need_to_wait; +try_to_lock: /* We don't have a lock on this vnode and we aren't currently waiting * for one either, so ask the server for a lock. * * Note that we need to be careful if we get interrupted by a signal * after dispatching the request as we may still get the lock, even * though we don't wait for the reply (it's not too bad a problem - the - * lock will expire in 10 mins anyway). + * lock will expire in 5 mins anyway). */ - _debug("not locked"); + trace_afs_flock_ev(vnode, fl, afs_flock_try_to_lock, 0); vnode->lock_key = key_get(key); vnode->lock_type = type; - vnode->lock_state = AFS_VNODE_LOCK_SETTING; + afs_set_lock_state(vnode, AFS_VNODE_LOCK_SETTING); spin_unlock(&vnode->lock); ret = afs_set_lock(vnode, key, type); /* RPC */ spin_lock(&vnode->lock); switch (ret) { + case -EKEYREJECTED: + case -EKEYEXPIRED: + case -EKEYREVOKED: + case -EPERM: + case -EACCES: + fl->fl_u.afs.state = ret; + trace_afs_flock_ev(vnode, fl, afs_flock_fail_perm, ret); + list_del_init(&fl->fl_u.afs.link); + afs_next_locker(vnode, ret); + goto error_unlock; + + case -ENOENT: + fl->fl_u.afs.state = ret; + trace_afs_flock_ev(vnode, fl, afs_flock_fail_other, ret); + list_del_init(&fl->fl_u.afs.link); + afs_kill_lockers_enoent(vnode); + goto error_unlock; + default: - goto abort_attempt; + fl->fl_u.afs.state = ret; + trace_afs_flock_ev(vnode, fl, afs_flock_fail_other, ret); + list_del_init(&fl->fl_u.afs.link); + afs_next_locker(vnode, 0); + goto error_unlock; case -EWOULDBLOCK: /* The server doesn't have a lock-waiting queue, so the client * will have to retry. The server will break the outstanding * callbacks on a file when a lock is released. */ - _debug("would block"); ASSERT(list_empty(&vnode->granted_locks)); ASSERTCMP(vnode->pending_locks.next, ==, &fl->fl_u.afs.link); - vnode->lock_state = AFS_VNODE_LOCK_WAITING_FOR_CB; - goto need_to_wait; + goto lock_is_contended; case 0: - _debug("acquired"); - break; + afs_set_lock_state(vnode, AFS_VNODE_LOCK_GRANTED); + trace_afs_flock_ev(vnode, fl, afs_flock_acquired, type); + afs_grant_locks(vnode); + goto vnode_is_locked_u; } - /* we've acquired a server lock, but it needs to be renewed after 5 - * mins */ - vnode->lock_state = AFS_VNODE_LOCK_GRANTED; - afs_schedule_lock_extension(vnode); - -share_existing_lock: - /* the lock has been granted as far as we're concerned... */ - fl->fl_u.afs.state = AFS_LOCK_GRANTED; - list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); - -given_lock: - /* ... but we do still need to get the VFS's blessing */ +vnode_is_locked_u: spin_unlock(&vnode->lock); - - ret = posix_lock_file(file, fl, NULL); +vnode_is_locked: + /* the lock has been granted by the server... */ + ASSERTCMP(fl->fl_u.afs.state, ==, AFS_LOCK_GRANTED); + +skip_server_lock: + /* ... but the VFS still needs to distribute access on this client. */ + trace_afs_flock_ev(vnode, fl, afs_flock_vfs_locking, 0); + ret = locks_lock_file_wait(file, fl); + trace_afs_flock_ev(vnode, fl, afs_flock_vfs_lock, ret); if (ret < 0) goto vfs_rejected_lock; @@ -528,38 +619,62 @@ given_lock: _leave(" = 0"); return 0; +lock_is_contended: + if (!(fl->c.flc_flags & FL_SLEEP)) { + list_del_init(&fl->fl_u.afs.link); + afs_next_locker(vnode, 0); + ret = -EAGAIN; + goto error_unlock; + } + + afs_set_lock_state(vnode, AFS_VNODE_LOCK_WAITING_FOR_CB); + trace_afs_flock_ev(vnode, fl, afs_flock_would_block, ret); + queue_delayed_work(afs_lock_manager, &vnode->lock_work, HZ * 5); + need_to_wait: /* We're going to have to wait. Either this client doesn't have a lock * on the server yet and we need to wait for a callback to occur, or - * the client does have a lock on the server, but it belongs to some - * other process(es) and is incompatible with the lock we want. + * the client does have a lock on the server, but it's shared and we + * need an exclusive lock. */ - ret = -EAGAIN; - if (fl->fl_flags & FL_SLEEP) { - spin_unlock(&vnode->lock); + spin_unlock(&vnode->lock); - _debug("sleep"); - ret = wait_event_interruptible(fl->fl_wait, - fl->fl_u.afs.state != AFS_LOCK_PENDING); + trace_afs_flock_ev(vnode, fl, afs_flock_waiting, 0); + ret = wait_event_interruptible(fl->c.flc_wait, + fl->fl_u.afs.state != AFS_LOCK_PENDING); + trace_afs_flock_ev(vnode, fl, afs_flock_waited, ret); + if (fl->fl_u.afs.state >= 0 && fl->fl_u.afs.state != AFS_LOCK_GRANTED) { spin_lock(&vnode->lock); - } - if (fl->fl_u.afs.state == AFS_LOCK_GRANTED) - goto given_lock; - if (fl->fl_u.afs.state < 0) - ret = fl->fl_u.afs.state; + switch (fl->fl_u.afs.state) { + case AFS_LOCK_YOUR_TRY: + fl->fl_u.afs.state = AFS_LOCK_PENDING; + goto try_to_lock; + case AFS_LOCK_PENDING: + if (ret > 0) { + /* We need to retry the lock. We may not be + * notified by the server if it just expired + * rather than being released. + */ + ASSERTCMP(vnode->lock_state, ==, AFS_VNODE_LOCK_WAITING_FOR_CB); + afs_set_lock_state(vnode, AFS_VNODE_LOCK_SETTING); + fl->fl_u.afs.state = AFS_LOCK_PENDING; + goto try_to_lock; + } + goto error_unlock; + case AFS_LOCK_GRANTED: + default: + break; + } -abort_attempt: - /* we aren't going to get the lock, either because we're unwilling to - * wait, or because some signal happened */ - _debug("abort"); - afs_dequeue_lock(vnode, fl); + spin_unlock(&vnode->lock); + } -error_unlock: - spin_unlock(&vnode->lock); - _leave(" = %d", ret); - return ret; + if (fl->fl_u.afs.state == AFS_LOCK_GRANTED) + goto vnode_is_locked; + ret = fl->fl_u.afs.state; + goto error; vfs_rejected_lock: /* The VFS rejected the lock we just obtained, so we have to discard @@ -567,11 +682,17 @@ vfs_rejected_lock: * deal with. */ _debug("vfs refused %d", ret); + if (no_server_lock) + goto error; spin_lock(&vnode->lock); list_del_init(&fl->fl_u.afs.link); - if (list_empty(&vnode->granted_locks)) - afs_defer_unlock(vnode); - goto error_unlock; + afs_defer_unlock(vnode); + +error_unlock: + spin_unlock(&vnode->lock); +error: + _leave(" = %d", ret); + return ret; } /* @@ -579,19 +700,18 @@ vfs_rejected_lock: */ static int afs_do_unlk(struct file *file, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); int ret; - _enter("{%llx:%llu},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type); + _enter("{%llx:%llu},%u", vnode->fid.vid, vnode->fid.vnode, + fl->c.flc_type); + + trace_afs_flock_op(vnode, fl, afs_flock_op_unlock); /* Flush all pending writes before doing anything with locks. */ vfs_fsync(file, 0); - /* only whole-file unlocks are supported */ - if (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX) - return -EINVAL; - - ret = posix_lock_file(file, fl, NULL); + ret = locks_lock_file_wait(file, fl); _leave(" = %d [%u]", ret, vnode->lock_state); return ret; } @@ -601,34 +721,40 @@ static int afs_do_unlk(struct file *file, struct file_lock *fl) */ static int afs_do_getlk(struct file *file, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); struct key *key = afs_file_key(file); int ret, lock_count; _enter(""); - fl->fl_type = F_UNLCK; + if (vnode->lock_state == AFS_VNODE_LOCK_DELETED) + return -ENOENT; + + fl->c.flc_type = F_UNLCK; /* check local lock records first */ posix_test_lock(file, fl); - if (fl->fl_type == F_UNLCK) { + if (lock_is_unlock(fl)) { /* no local locks; consult the server */ - ret = afs_fetch_status(vnode, key, false); + ret = afs_fetch_status(vnode, key, false, NULL); if (ret < 0) goto error; lock_count = READ_ONCE(vnode->status.lock_count); - if (lock_count > 0) - fl->fl_type = F_RDLCK; - else - fl->fl_type = F_WRLCK; - fl->fl_start = 0; - fl->fl_end = OFFSET_MAX; + if (lock_count != 0) { + if (lock_count > 0) + fl->c.flc_type = F_RDLCK; + else + fl->c.flc_type = F_WRLCK; + fl->fl_start = 0; + fl->fl_end = OFFSET_MAX; + fl->c.flc_pid = 0; + } } ret = 0; error: - _leave(" = %d [%hd]", ret, fl->fl_type); + _leave(" = %d [%hd]", ret, fl->c.flc_type); return ret; } @@ -637,22 +763,34 @@ error: */ int afs_lock(struct file *file, int cmd, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + enum afs_flock_operation op; + int ret; _enter("{%llx:%llu},%d,{t=%x,fl=%x,r=%Ld:%Ld}", vnode->fid.vid, vnode->fid.vnode, cmd, - fl->fl_type, fl->fl_flags, + fl->c.flc_type, fl->c.flc_flags, (long long) fl->fl_start, (long long) fl->fl_end); - /* AFS doesn't support mandatory locks */ - if (__mandatory_lock(&vnode->vfs_inode) && fl->fl_type != F_UNLCK) - return -ENOLCK; - if (IS_GETLK(cmd)) return afs_do_getlk(file, fl); - if (fl->fl_type == F_UNLCK) - return afs_do_unlk(file, fl); - return afs_do_setlk(file, fl); + + fl->fl_u.afs.debug_id = atomic_inc_return(&afs_file_lock_debug_id); + trace_afs_flock_op(vnode, fl, afs_flock_op_lock); + + if (lock_is_unlock(fl)) + ret = afs_do_unlk(file, fl); + else + ret = afs_do_setlk(file, fl); + + switch (ret) { + case 0: op = afs_flock_op_return_ok; break; + case -EAGAIN: op = afs_flock_op_return_eagain; break; + case -EDEADLK: op = afs_flock_op_return_edeadlk; break; + default: op = afs_flock_op_return_error; break; + } + trace_afs_flock_op(vnode, fl, op); + return ret; } /* @@ -660,11 +798,13 @@ int afs_lock(struct file *file, int cmd, struct file_lock *fl) */ int afs_flock(struct file *file, int cmd, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + enum afs_flock_operation op; + int ret; _enter("{%llx:%llu},%d,{t=%x,fl=%x}", vnode->fid.vid, vnode->fid.vnode, cmd, - fl->fl_type, fl->fl_flags); + fl->c.flc_type, fl->c.flc_flags); /* * No BSD flocks over NFS allowed. @@ -673,13 +813,26 @@ int afs_flock(struct file *file, int cmd, struct file_lock *fl) * Not sure whether that would be unique, though, or whether * that would break in other places. */ - if (!(fl->fl_flags & FL_FLOCK)) + if (!(fl->c.flc_flags & FL_FLOCK)) return -ENOLCK; + fl->fl_u.afs.debug_id = atomic_inc_return(&afs_file_lock_debug_id); + trace_afs_flock_op(vnode, fl, afs_flock_op_flock); + /* we're simulating flock() locks using posix locks on the server */ - if (fl->fl_type == F_UNLCK) - return afs_do_unlk(file, fl); - return afs_do_setlk(file, fl); + if (lock_is_unlock(fl)) + ret = afs_do_unlk(file, fl); + else + ret = afs_do_setlk(file, fl); + + switch (ret) { + case 0: op = afs_flock_op_return_ok; break; + case -EAGAIN: op = afs_flock_op_return_eagain; break; + case -EDEADLK: op = afs_flock_op_return_edeadlk; break; + default: op = afs_flock_op_return_error; break; + } + trace_afs_flock_op(vnode, fl, op); + return ret; } /* @@ -690,11 +843,14 @@ int afs_flock(struct file *file, int cmd, struct file_lock *fl) */ static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(fl->fl_file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(fl->c.flc_file)); _enter(""); + new->fl_u.afs.debug_id = atomic_inc_return(&afs_file_lock_debug_id); + spin_lock(&vnode->lock); + trace_afs_flock_op(vnode, new, afs_flock_op_copy_lock); list_add(&new->fl_u.afs.link, &fl->fl_u.afs.link); spin_unlock(&vnode->lock); } @@ -705,12 +861,17 @@ static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl) */ static void afs_fl_release_private(struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(locks_inode(fl->fl_file)); + struct afs_vnode *vnode = AFS_FS_I(file_inode(fl->c.flc_file)); _enter(""); spin_lock(&vnode->lock); - afs_dequeue_lock(vnode, fl); + + trace_afs_flock_op(vnode, fl, afs_flock_op_release_lock); + list_del_init(&fl->fl_u.afs.link); + if (list_empty(&vnode->granted_locks)) + afs_defer_unlock(vnode); + _debug("state %u for %p", vnode->lock_state, vnode); spin_unlock(&vnode->lock); } diff --git a/fs/afs/fs_operation.c b/fs/afs/fs_operation.c new file mode 100644 index 000000000000..8418813ee043 --- /dev/null +++ b/fs/afs/fs_operation.c @@ -0,0 +1,377 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Fileserver-directed operation handling. + * + * Copyright (C) 2020 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/fs.h> +#include "internal.h" + +static atomic_t afs_operation_debug_counter; + +/* + * Create an operation against a volume. + */ +struct afs_operation *afs_alloc_operation(struct key *key, struct afs_volume *volume) +{ + struct afs_operation *op; + + _enter(""); + + op = kzalloc(sizeof(*op), GFP_KERNEL); + if (!op) + return ERR_PTR(-ENOMEM); + + if (!key) { + key = afs_request_key(volume->cell); + if (IS_ERR(key)) { + kfree(op); + return ERR_CAST(key); + } + } else { + key_get(key); + } + + op->key = key; + op->volume = afs_get_volume(volume, afs_volume_trace_get_new_op); + op->net = volume->cell->net; + op->cb_v_break = atomic_read(&volume->cb_v_break); + op->pre_volsync.creation = volume->creation_time; + op->pre_volsync.update = volume->update_time; + op->debug_id = atomic_inc_return(&afs_operation_debug_counter); + op->nr_iterations = -1; + afs_op_set_error(op, -EDESTADDRREQ); + + _leave(" = [op=%08x]", op->debug_id); + return op; +} + +struct afs_io_locker { + struct list_head link; + struct task_struct *task; + unsigned long have_lock; +}; + +/* + * Unlock the I/O lock on a vnode. + */ +static void afs_unlock_for_io(struct afs_vnode *vnode) +{ + struct afs_io_locker *locker; + + spin_lock(&vnode->lock); + locker = list_first_entry_or_null(&vnode->io_lock_waiters, + struct afs_io_locker, link); + if (locker) { + list_del(&locker->link); + smp_store_release(&locker->have_lock, 1); /* The unlock barrier. */ + smp_mb__after_atomic(); /* Store have_lock before task state */ + wake_up_process(locker->task); + } else { + clear_bit(AFS_VNODE_IO_LOCK, &vnode->flags); + } + spin_unlock(&vnode->lock); +} + +/* + * Lock the I/O lock on a vnode uninterruptibly. We can't use an ordinary + * mutex as lockdep will complain if we unlock it in the wrong thread. + */ +static void afs_lock_for_io(struct afs_vnode *vnode) +{ + struct afs_io_locker myself = { .task = current, }; + + spin_lock(&vnode->lock); + + if (!test_and_set_bit(AFS_VNODE_IO_LOCK, &vnode->flags)) { + spin_unlock(&vnode->lock); + return; + } + + list_add_tail(&myself.link, &vnode->io_lock_waiters); + spin_unlock(&vnode->lock); + + for (;;) { + set_current_state(TASK_UNINTERRUPTIBLE); + if (smp_load_acquire(&myself.have_lock)) /* The lock barrier */ + break; + schedule(); + } + __set_current_state(TASK_RUNNING); +} + +/* + * Lock the I/O lock on a vnode interruptibly. We can't use an ordinary mutex + * as lockdep will complain if we unlock it in the wrong thread. + */ +static int afs_lock_for_io_interruptible(struct afs_vnode *vnode) +{ + struct afs_io_locker myself = { .task = current, }; + int ret = 0; + + spin_lock(&vnode->lock); + + if (!test_and_set_bit(AFS_VNODE_IO_LOCK, &vnode->flags)) { + spin_unlock(&vnode->lock); + return 0; + } + + list_add_tail(&myself.link, &vnode->io_lock_waiters); + spin_unlock(&vnode->lock); + + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (smp_load_acquire(&myself.have_lock) || /* The lock barrier */ + signal_pending(current)) + break; + schedule(); + } + __set_current_state(TASK_RUNNING); + + /* If we got a signal, try to transfer the lock onto the next + * waiter. + */ + if (unlikely(signal_pending(current))) { + spin_lock(&vnode->lock); + if (myself.have_lock) { + spin_unlock(&vnode->lock); + afs_unlock_for_io(vnode); + } else { + list_del(&myself.link); + spin_unlock(&vnode->lock); + } + ret = -ERESTARTSYS; + } + return ret; +} + +/* + * Lock the vnode(s) being operated upon. + */ +static bool afs_get_io_locks(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->file[0].vnode; + struct afs_vnode *vnode2 = op->file[1].vnode; + + _enter(""); + + if (op->flags & AFS_OPERATION_UNINTR) { + afs_lock_for_io(vnode); + op->flags |= AFS_OPERATION_LOCK_0; + _leave(" = t [1]"); + return true; + } + + if (!vnode2 || !op->file[1].need_io_lock || vnode == vnode2) + vnode2 = NULL; + + if (vnode2 > vnode) + swap(vnode, vnode2); + + if (afs_lock_for_io_interruptible(vnode) < 0) { + afs_op_set_error(op, -ERESTARTSYS); + op->flags |= AFS_OPERATION_STOP; + _leave(" = f [I 0]"); + return false; + } + op->flags |= AFS_OPERATION_LOCK_0; + + if (vnode2) { + if (afs_lock_for_io_interruptible(vnode2) < 0) { + afs_op_set_error(op, -ERESTARTSYS); + op->flags |= AFS_OPERATION_STOP; + afs_unlock_for_io(vnode); + op->flags &= ~AFS_OPERATION_LOCK_0; + _leave(" = f [I 1]"); + return false; + } + op->flags |= AFS_OPERATION_LOCK_1; + } + + _leave(" = t [2]"); + return true; +} + +static void afs_drop_io_locks(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->file[0].vnode; + struct afs_vnode *vnode2 = op->file[1].vnode; + + _enter(""); + + if (op->flags & AFS_OPERATION_LOCK_1) + afs_unlock_for_io(vnode2); + if (op->flags & AFS_OPERATION_LOCK_0) + afs_unlock_for_io(vnode); +} + +static void afs_prepare_vnode(struct afs_operation *op, struct afs_vnode_param *vp, + unsigned int index) +{ + struct afs_vnode *vnode = vp->vnode; + + if (vnode) { + vp->fid = vnode->fid; + vp->dv_before = vnode->status.data_version; + vp->cb_break_before = afs_calc_vnode_cb_break(vnode); + if (vnode->lock_state != AFS_VNODE_LOCK_NONE) + op->flags |= AFS_OPERATION_CUR_ONLY; + if (vp->modification) + set_bit(AFS_VNODE_MODIFYING, &vnode->flags); + } + + if (vp->fid.vnode) + _debug("PREP[%u] {%llx:%llu.%u}", + index, vp->fid.vid, vp->fid.vnode, vp->fid.unique); +} + +/* + * Begin an operation on the fileserver. + * + * Fileserver operations are serialised on the server by vnode, so we serialise + * them here also using the io_lock. + */ +bool afs_begin_vnode_operation(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->file[0].vnode; + + ASSERT(vnode); + + _enter(""); + + if (op->file[0].need_io_lock) + if (!afs_get_io_locks(op)) + return false; + + afs_prepare_vnode(op, &op->file[0], 0); + afs_prepare_vnode(op, &op->file[1], 1); + op->cb_v_break = atomic_read(&op->volume->cb_v_break); + _leave(" = true"); + return true; +} + +/* + * Tidy up a filesystem cursor and unlock the vnode. + */ +void afs_end_vnode_operation(struct afs_operation *op) +{ + _enter(""); + + switch (afs_op_error(op)) { + case -EDESTADDRREQ: + case -EADDRNOTAVAIL: + case -ENETUNREACH: + case -EHOSTUNREACH: + afs_dump_edestaddrreq(op); + break; + } + + afs_drop_io_locks(op); +} + +/* + * Wait for an in-progress operation to complete. + */ +void afs_wait_for_operation(struct afs_operation *op) +{ + _enter(""); + + while (afs_select_fileserver(op)) { + op->call_responded = false; + op->call_error = 0; + op->call_abort_code = 0; + if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags) && + op->ops->issue_yfs_rpc) + op->ops->issue_yfs_rpc(op); + else if (op->ops->issue_afs_rpc) + op->ops->issue_afs_rpc(op); + else + op->call_error = -ENOTSUPP; + + if (op->call) { + afs_wait_for_call_to_complete(op->call); + op->call_abort_code = op->call->abort_code; + op->call_error = op->call->error; + op->call_responded = op->call->responded; + afs_put_call(op->call); + } + } + + if (op->call_responded && op->server) + set_bit(AFS_SERVER_FL_RESPONDING, &op->server->flags); + + if (!afs_op_error(op)) { + _debug("success"); + op->ops->success(op); + } else if (op->cumul_error.aborted) { + if (op->ops->aborted) + op->ops->aborted(op); + } else { + if (op->ops->failed) + op->ops->failed(op); + } + + afs_end_vnode_operation(op); + + if (!afs_op_error(op) && op->ops->edit_dir) { + _debug("edit_dir"); + op->ops->edit_dir(op); + } + _leave(""); +} + +/* + * Dispose of an operation. + */ +int afs_put_operation(struct afs_operation *op) +{ + struct afs_addr_list *alist; + int i, ret = afs_op_error(op); + + _enter("op=%08x,%d", op->debug_id, ret); + + if (op->ops && op->ops->put) + op->ops->put(op); + if (op->file[0].modification) + clear_bit(AFS_VNODE_MODIFYING, &op->file[0].vnode->flags); + if (op->file[1].modification && op->file[1].vnode != op->file[0].vnode) + clear_bit(AFS_VNODE_MODIFYING, &op->file[1].vnode->flags); + if (op->file[0].put_vnode) + iput(&op->file[0].vnode->netfs.inode); + if (op->file[1].put_vnode) + iput(&op->file[1].vnode->netfs.inode); + + if (op->more_files) { + for (i = 0; i < op->nr_files - 2; i++) + if (op->more_files[i].put_vnode) + iput(&op->more_files[i].vnode->netfs.inode); + kfree(op->more_files); + } + + if (op->estate) { + alist = op->estate->addresses; + if (alist) { + if (op->call_responded && + op->addr_index != alist->preferred && + test_bit(alist->preferred, &op->addr_tried)) + WRITE_ONCE(alist->preferred, op->addr_index); + } + } + + afs_clear_server_states(op); + afs_put_serverlist(op->net, op->server_list); + afs_put_volume(op->volume, afs_volume_trace_put_put_op); + key_put(op->key); + kfree(op); + return ret; +} + +int afs_do_sync_operation(struct afs_operation *op) +{ + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + return afs_put_operation(op); +} diff --git a/fs/afs/fs_probe.c b/fs/afs/fs_probe.c index 3a9eaec06756..e0030ac74ea0 100644 --- a/fs/afs/fs_probe.c +++ b/fs/afs/fs_probe.c @@ -1,29 +1,134 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS fileserver probing * - * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2018, 2020 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/sched.h> #include <linux/slab.h> #include "afs_fs.h" #include "internal.h" +#include "protocol_afs.h" #include "protocol_yfs.h" -static bool afs_fs_probe_done(struct afs_server *server) +static unsigned int afs_fs_probe_fast_poll_interval = 30 * HZ; +static unsigned int afs_fs_probe_slow_poll_interval = 5 * 60 * HZ; + +struct afs_endpoint_state *afs_get_endpoint_state(struct afs_endpoint_state *estate, + enum afs_estate_trace where) +{ + if (estate) { + int r; + + __refcount_inc(&estate->ref, &r); + trace_afs_estate(estate->server_id, estate->probe_seq, r, where); + } + return estate; +} + +static void afs_endpoint_state_rcu(struct rcu_head *rcu) +{ + struct afs_endpoint_state *estate = container_of(rcu, struct afs_endpoint_state, rcu); + + trace_afs_estate(estate->server_id, estate->probe_seq, refcount_read(&estate->ref), + afs_estate_trace_free); + afs_put_addrlist(estate->addresses, afs_alist_trace_put_estate); + kfree(estate); +} + +void afs_put_endpoint_state(struct afs_endpoint_state *estate, enum afs_estate_trace where) +{ + if (estate) { + unsigned int server_id = estate->server_id, probe_seq = estate->probe_seq; + bool dead; + int r; + + dead = __refcount_dec_and_test(&estate->ref, &r); + trace_afs_estate(server_id, probe_seq, r, where); + if (dead) + call_rcu(&estate->rcu, afs_endpoint_state_rcu); + } +} + +/* + * Start the probe polling timer. We have to supply it with an inc on the + * outstanding server count. + */ +static void afs_schedule_fs_probe(struct afs_net *net, + struct afs_server *server, bool fast) +{ + unsigned long atj; + + if (!net->live) + return; + + atj = server->probed_at; + atj += fast ? afs_fs_probe_fast_poll_interval : afs_fs_probe_slow_poll_interval; + + afs_inc_servers_outstanding(net); + if (timer_reduce(&net->fs_probe_timer, atj)) + afs_dec_servers_outstanding(net); +} + +/* + * Handle the completion of a set of probes. + */ +static void afs_finished_fs_probe(struct afs_net *net, struct afs_server *server, + struct afs_endpoint_state *estate) +{ + bool responded = test_bit(AFS_ESTATE_RESPONDED, &estate->flags); + + write_seqlock(&net->fs_lock); + if (responded) { + list_add_tail(&server->probe_link, &net->fs_probe_slow); + } else { + server->rtt = UINT_MAX; + clear_bit(AFS_SERVER_FL_RESPONDING, &server->flags); + list_add_tail(&server->probe_link, &net->fs_probe_fast); + } + + write_sequnlock(&net->fs_lock); + + afs_schedule_fs_probe(net, server, !responded); +} + +/* + * Handle the completion of a probe. + */ +static void afs_done_one_fs_probe(struct afs_net *net, struct afs_server *server, + struct afs_endpoint_state *estate) +{ + _enter(""); + + if (atomic_dec_and_test(&estate->nr_probing)) + afs_finished_fs_probe(net, server, estate); + + wake_up_all(&server->probe_wq); +} + +/* + * Handle inability to send a probe due to ENOMEM when trying to allocate a + * call struct. + */ +static void afs_fs_probe_not_done(struct afs_net *net, + struct afs_server *server, + struct afs_endpoint_state *estate, + int index) { - if (!atomic_dec_and_test(&server->probe_outstanding)) - return false; + _enter(""); + + trace_afs_io_error(0, -ENOMEM, afs_io_error_fs_probe_fail); + spin_lock(&server->probe_lock); - wake_up_var(&server->probe_outstanding); - clear_bit_unlock(AFS_SERVER_FL_PROBING, &server->flags); - wake_up_bit(&server->flags, AFS_SERVER_FL_PROBING); - return true; + set_bit(AFS_ESTATE_LOCAL_FAILURE, &estate->flags); + if (estate->error == 0) + estate->error = -ENOMEM; + + set_bit(index, &estate->failed_set); + + spin_unlock(&server->probe_lock); + return afs_done_one_fs_probe(net, server, estate); } /* @@ -32,33 +137,35 @@ static bool afs_fs_probe_done(struct afs_server *server) */ void afs_fileserver_probe_result(struct afs_call *call) { - struct afs_addr_list *alist = call->alist; - struct afs_server *server = call->reply[0]; - unsigned int server_index = (long)call->reply[1]; - unsigned int index = call->addr_ix; - unsigned int rtt = UINT_MAX; - bool have_result = false; - u64 _rtt; + struct afs_endpoint_state *estate = call->probe; + struct afs_addr_list *alist = estate->addresses; + struct afs_address *addr = &alist->addrs[call->probe_index]; + struct afs_server *server = call->server; + unsigned int index = call->probe_index; + unsigned int rtt_us = -1, cap0; int ret = call->error; _enter("%pU,%u", &server->uuid, index); + WRITE_ONCE(addr->last_error, ret); + spin_lock(&server->probe_lock); switch (ret) { case 0: - server->probe.error = 0; + estate->error = 0; goto responded; case -ECONNABORTED: - if (!server->probe.responded) { - server->probe.abort_code = call->abort_code; - server->probe.error = ret; + if (!test_bit(AFS_ESTATE_RESPONDED, &estate->flags)) { + estate->abort_code = call->abort_code; + estate->error = ret; } goto responded; case -ENOMEM: case -ENONET: - server->probe.local_failure = true; - afs_io_error(call, afs_io_error_fs_probe_fail); + clear_bit(index, &estate->responsive_set); + set_bit(AFS_ESTATE_LOCAL_FAILURE, &estate->flags); + trace_afs_io_error(call->debug_id, ret, afs_io_error_fs_probe_fail); goto out; case -ECONNRESET: /* Responded, but call expired. */ case -ERFKILL: @@ -70,180 +177,179 @@ void afs_fileserver_probe_result(struct afs_call *call) case -ETIMEDOUT: case -ETIME: default: - clear_bit(index, &alist->responded); - set_bit(index, &alist->failed); - if (!server->probe.responded && - (server->probe.error == 0 || - server->probe.error == -ETIMEDOUT || - server->probe.error == -ETIME)) - server->probe.error = ret; - afs_io_error(call, afs_io_error_fs_probe_fail); + clear_bit(index, &estate->responsive_set); + set_bit(index, &estate->failed_set); + if (!test_bit(AFS_ESTATE_RESPONDED, &estate->flags) && + (estate->error == 0 || + estate->error == -ETIMEDOUT || + estate->error == -ETIME)) + estate->error = ret; + trace_afs_io_error(call->debug_id, ret, afs_io_error_fs_probe_fail); goto out; } responded: - set_bit(index, &alist->responded); - clear_bit(index, &alist->failed); + clear_bit(index, &estate->failed_set); if (call->service_id == YFS_FS_SERVICE) { - server->probe.is_yfs = true; + set_bit(AFS_ESTATE_IS_YFS, &estate->flags); set_bit(AFS_SERVER_FL_IS_YFS, &server->flags); - alist->addrs[index].srx_service = call->service_id; + server->service_id = call->service_id; } else { - server->probe.not_yfs = true; - if (!server->probe.is_yfs) { + set_bit(AFS_ESTATE_NOT_YFS, &estate->flags); + if (!test_bit(AFS_ESTATE_IS_YFS, &estate->flags)) { clear_bit(AFS_SERVER_FL_IS_YFS, &server->flags); - alist->addrs[index].srx_service = call->service_id; + server->service_id = call->service_id; } + cap0 = ntohl(call->tmp); + if (cap0 & AFS3_VICED_CAPABILITY_64BITFILES) + set_bit(AFS_SERVER_FL_HAS_FS64, &server->flags); + else + clear_bit(AFS_SERVER_FL_HAS_FS64, &server->flags); } - /* Get the RTT and scale it to fit into a 32-bit value that represents - * over a minute of time so that we can access it with one instruction - * on a 32-bit system. - */ - _rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall); - _rtt /= 64; - rtt = (_rtt > UINT_MAX) ? UINT_MAX : _rtt; - if (rtt < server->probe.rtt) { - server->probe.rtt = rtt; + rtt_us = rxrpc_kernel_get_srtt(addr->peer); + if (rtt_us < estate->rtt) { + estate->rtt = rtt_us; + server->rtt = rtt_us; alist->preferred = index; - have_result = true; } smp_wmb(); /* Set rtt before responded. */ - server->probe.responded = true; - set_bit(AFS_SERVER_FL_PROBED, &server->flags); + set_bit(AFS_ESTATE_RESPONDED, &estate->flags); + set_bit(index, &estate->responsive_set); + set_bit(AFS_SERVER_FL_RESPONDING, &server->flags); out: spin_unlock(&server->probe_lock); - _debug("probe [%u][%u] %pISpc rtt=%u ret=%d", - server_index, index, &alist->addrs[index].transport, - (unsigned int)rtt, ret); + trace_afs_fs_probe(server, false, estate, index, call->error, call->abort_code, rtt_us); + _debug("probe[%x] %pU [%u] %pISpc rtt=%d ret=%d", + estate->probe_seq, &server->uuid, index, + rxrpc_kernel_remote_addr(alist->addrs[index].peer), + rtt_us, ret); - have_result |= afs_fs_probe_done(server); - if (have_result) { - server->probe.have_result = true; - wake_up_var(&server->probe.have_result); - wake_up_all(&server->probe_wq); - } + return afs_done_one_fs_probe(call->net, server, estate); } /* * Probe all of a fileserver's addresses to find out the best route and to * query its capabilities. */ -static int afs_do_probe_fileserver(struct afs_net *net, - struct afs_server *server, - struct key *key, - unsigned int server_index, - struct afs_error *_e) +int afs_fs_probe_fileserver(struct afs_net *net, struct afs_server *server, + struct afs_addr_list *new_alist, struct key *key) { - struct afs_addr_cursor ac = { - .index = 0, - }; - bool in_progress = false; - int err; + struct afs_endpoint_state *estate, *old; + struct afs_addr_list *old_alist = NULL, *alist; + unsigned long unprobed; _enter("%pU", &server->uuid); - read_lock(&server->fs_lock); - ac.alist = rcu_dereference_protected(server->addresses, - lockdep_is_held(&server->fs_lock)); - read_unlock(&server->fs_lock); + estate = kzalloc(sizeof(*estate), GFP_KERNEL); + if (!estate) + return -ENOMEM; + + refcount_set(&estate->ref, 2); + estate->server_id = server->debug_id; + estate->rtt = UINT_MAX; - atomic_set(&server->probe_outstanding, ac.alist->nr_addrs); - memset(&server->probe, 0, sizeof(server->probe)); - server->probe.rtt = UINT_MAX; + write_lock(&server->fs_lock); - for (ac.index = 0; ac.index < ac.alist->nr_addrs; ac.index++) { - err = afs_fs_get_capabilities(net, server, &ac, key, server_index, - true); - if (err == -EINPROGRESS) - in_progress = true; - else - afs_prioritise_error(_e, err, ac.abort_code); + old = rcu_dereference_protected(server->endpoint_state, + lockdep_is_held(&server->fs_lock)); + if (old) { + estate->responsive_set = old->responsive_set; + if (!new_alist) + new_alist = old->addresses; } - if (!in_progress) - afs_fs_probe_done(server); - return in_progress; -} + if (old_alist != new_alist) + afs_set_peer_appdata(server, old_alist, new_alist); + + estate->addresses = afs_get_addrlist(new_alist, afs_alist_trace_get_estate); + alist = estate->addresses; + estate->probe_seq = ++server->probe_counter; + atomic_set(&estate->nr_probing, alist->nr_addrs); + + if (new_alist) + server->addr_version = new_alist->version; + rcu_assign_pointer(server->endpoint_state, estate); + write_unlock(&server->fs_lock); + if (old) + set_bit(AFS_ESTATE_SUPERSEDED, &old->flags); + + trace_afs_estate(estate->server_id, estate->probe_seq, refcount_read(&estate->ref), + afs_estate_trace_alloc_probe); + + afs_get_address_preferences(net, new_alist); + + server->probed_at = jiffies; + unprobed = (1UL << alist->nr_addrs) - 1; + while (unprobed) { + unsigned int index = 0, i; + int best_prio = -1; + + for (i = 0; i < alist->nr_addrs; i++) { + if (test_bit(i, &unprobed) && + alist->addrs[i].prio > best_prio) { + index = i; + best_prio = alist->addrs[i].prio; + } + } + __clear_bit(index, &unprobed); -/* - * Send off probes to all unprobed servers. - */ -int afs_probe_fileservers(struct afs_net *net, struct key *key, - struct afs_server_list *list) -{ - struct afs_server *server; - struct afs_error e; - bool in_progress = false; - int i; - - e.error = 0; - e.responded = false; - for (i = 0; i < list->nr_servers; i++) { - server = list->servers[i].server; - if (test_bit(AFS_SERVER_FL_PROBED, &server->flags)) - continue; - - if (!test_and_set_bit_lock(AFS_SERVER_FL_PROBING, &server->flags) && - afs_do_probe_fileserver(net, server, key, i, &e)) - in_progress = true; + trace_afs_fs_probe(server, true, estate, index, 0, 0, 0); + if (!afs_fs_get_capabilities(net, server, estate, index, key)) + afs_fs_probe_not_done(net, server, estate, index); } - return in_progress ? 0 : e.error; + afs_put_endpoint_state(old, afs_estate_trace_put_probe); + afs_put_endpoint_state(estate, afs_estate_trace_put_probe); + return 0; } /* - * Wait for the first as-yet untried fileserver to respond. + * Wait for the first as-yet untried fileserver to respond, for the probe state + * to be superseded or for all probes to finish. */ -int afs_wait_for_fs_probes(struct afs_server_list *slist, unsigned long untried) +int afs_wait_for_fs_probes(struct afs_operation *op, struct afs_server_state *states, bool intr) { - struct wait_queue_entry *waits; - struct afs_server *server; - unsigned int rtt = UINT_MAX; - bool have_responders = false; - int pref = -1, i; + struct afs_endpoint_state *estate; + struct afs_server_list *slist = op->server_list; + bool still_probing = true; + int ret = 0, i; - _enter("%u,%lx", slist->nr_servers, untried); + _enter("%u", slist->nr_servers); - /* Only wait for servers that have a probe outstanding. */ for (i = 0; i < slist->nr_servers; i++) { - if (test_bit(i, &untried)) { - server = slist->servers[i].server; - if (!test_bit(AFS_SERVER_FL_PROBING, &server->flags)) - __clear_bit(i, &untried); - if (server->probe.responded) - have_responders = true; - } + estate = states[i].endpoint_state; + if (test_bit(AFS_ESTATE_SUPERSEDED, &estate->flags)) + return 2; + if (atomic_read(&estate->nr_probing)) + still_probing = true; + if (estate->responsive_set & states[i].untried_addrs) + return 1; } - if (have_responders || !untried) + if (!still_probing) return 0; - waits = kmalloc(array_size(slist->nr_servers, sizeof(*waits)), GFP_KERNEL); - if (!waits) - return -ENOMEM; - - for (i = 0; i < slist->nr_servers; i++) { - if (test_bit(i, &untried)) { - server = slist->servers[i].server; - init_waitqueue_entry(&waits[i], current); - add_wait_queue(&server->probe_wq, &waits[i]); - } - } + for (i = 0; i < slist->nr_servers; i++) + add_wait_queue(&slist->servers[i].server->probe_wq, &states[i].probe_waiter); for (;;) { - bool still_probing = false; + still_probing = false; - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(intr ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE); for (i = 0; i < slist->nr_servers; i++) { - if (test_bit(i, &untried)) { - server = slist->servers[i].server; - if (server->probe.responded) - goto stop; - if (test_bit(AFS_SERVER_FL_PROBING, &server->flags)) - still_probing = true; + estate = states[i].endpoint_state; + if (test_bit(AFS_ESTATE_SUPERSEDED, &estate->flags)) { + ret = 2; + goto stop; + } + if (atomic_read(&estate->nr_probing)) + still_probing = true; + if (estate->responsive_set & states[i].untried_addrs) { + ret = 1; + goto stop; } } @@ -255,25 +361,179 @@ int afs_wait_for_fs_probes(struct afs_server_list *slist, unsigned long untried) stop: set_current_state(TASK_RUNNING); - for (i = 0; i < slist->nr_servers; i++) { - if (test_bit(i, &untried)) { - server = slist->servers[i].server; - if (server->probe.responded && - server->probe.rtt < rtt) { - pref = i; - rtt = server->probe.rtt; - } + for (i = 0; i < slist->nr_servers; i++) + remove_wait_queue(&slist->servers[i].server->probe_wq, &states[i].probe_waiter); + + if (!ret && signal_pending(current)) + ret = -ERESTARTSYS; + return ret; +} + +/* + * Probe timer. We have an increment on fs_outstanding that we need to pass + * along to the work item. + */ +void afs_fs_probe_timer(struct timer_list *timer) +{ + struct afs_net *net = container_of(timer, struct afs_net, fs_probe_timer); + + if (!net->live || !queue_work(afs_wq, &net->fs_prober)) + afs_dec_servers_outstanding(net); +} + +/* + * Dispatch a probe to a server. + */ +static void afs_dispatch_fs_probe(struct afs_net *net, struct afs_server *server) + __releases(&net->fs_lock) +{ + struct key *key = NULL; + + /* We remove it from the queues here - it will be added back to + * one of the queues on the completion of the probe. + */ + list_del_init(&server->probe_link); + + afs_get_server(server, afs_server_trace_get_probe); + write_sequnlock(&net->fs_lock); + + afs_fs_probe_fileserver(net, server, NULL, key); + afs_put_server(net, server, afs_server_trace_put_probe); +} + +/* + * Probe a server immediately without waiting for its due time to come + * round. This is used when all of the addresses have been tried. + */ +void afs_probe_fileserver(struct afs_net *net, struct afs_server *server) +{ + write_seqlock(&net->fs_lock); + if (!list_empty(&server->probe_link)) + return afs_dispatch_fs_probe(net, server); + write_sequnlock(&net->fs_lock); +} - remove_wait_queue(&server->probe_wq, &waits[i]); +/* + * Probe dispatcher to regularly dispatch probes to keep NAT alive. + */ +void afs_fs_probe_dispatcher(struct work_struct *work) +{ + struct afs_net *net = container_of(work, struct afs_net, fs_prober); + struct afs_server *fast, *slow, *server; + unsigned long nowj, timer_at, poll_at; + bool first_pass = true, set_timer = false; + + if (!net->live) { + afs_dec_servers_outstanding(net); + return; + } + + _enter(""); + + if (list_empty(&net->fs_probe_fast) && list_empty(&net->fs_probe_slow)) { + afs_dec_servers_outstanding(net); + _leave(" [none]"); + return; + } + +again: + write_seqlock(&net->fs_lock); + + fast = slow = server = NULL; + nowj = jiffies; + timer_at = nowj + MAX_JIFFY_OFFSET; + + if (!list_empty(&net->fs_probe_fast)) { + fast = list_first_entry(&net->fs_probe_fast, struct afs_server, probe_link); + poll_at = fast->probed_at + afs_fs_probe_fast_poll_interval; + if (time_before(nowj, poll_at)) { + timer_at = poll_at; + set_timer = true; + fast = NULL; + } + } + + if (!list_empty(&net->fs_probe_slow)) { + slow = list_first_entry(&net->fs_probe_slow, struct afs_server, probe_link); + poll_at = slow->probed_at + afs_fs_probe_slow_poll_interval; + if (time_before(nowj, poll_at)) { + if (time_before(poll_at, timer_at)) + timer_at = poll_at; + set_timer = true; + slow = NULL; } } - kfree(waits); + server = fast ?: slow; + if (server) + _debug("probe %pU", &server->uuid); + + if (server && (first_pass || !need_resched())) { + afs_dispatch_fs_probe(net, server); + first_pass = false; + goto again; + } + + write_sequnlock(&net->fs_lock); + + if (server) { + if (!queue_work(afs_wq, &net->fs_prober)) + afs_dec_servers_outstanding(net); + _leave(" [requeue]"); + } else if (set_timer) { + if (timer_reduce(&net->fs_probe_timer, timer_at)) + afs_dec_servers_outstanding(net); + _leave(" [timer]"); + } else { + afs_dec_servers_outstanding(net); + _leave(" [quiesce]"); + } +} - if (pref == -1 && signal_pending(current)) +/* + * Wait for a probe on a particular fileserver to complete for 2s. + */ +int afs_wait_for_one_fs_probe(struct afs_server *server, struct afs_endpoint_state *estate, + unsigned long exclude, bool is_intr) +{ + struct wait_queue_entry wait; + unsigned long timo = 2 * HZ; + + if (atomic_read(&estate->nr_probing) == 0) + goto dont_wait; + + init_wait_entry(&wait, 0); + for (;;) { + prepare_to_wait_event(&server->probe_wq, &wait, + is_intr ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE); + if (timo == 0 || + test_bit(AFS_ESTATE_SUPERSEDED, &estate->flags) || + (estate->responsive_set & ~exclude) || + atomic_read(&estate->nr_probing) == 0 || + (is_intr && signal_pending(current))) + break; + timo = schedule_timeout(timo); + } + + finish_wait(&server->probe_wq, &wait); + +dont_wait: + if (test_bit(AFS_ESTATE_SUPERSEDED, &estate->flags)) + return 0; + if (estate->responsive_set & ~exclude) + return 1; + if (is_intr && signal_pending(current)) return -ERESTARTSYS; + if (timo == 0) + return -ETIME; + return -EDESTADDRREQ; +} - if (pref >= 0) - slist->preferred = pref; - return 0; +/* + * Clean up the probing when the namespace is killed off. + */ +void afs_fs_probe_cleanup(struct afs_net *net) +{ + if (timer_delete_sync(&net->fs_probe_timer)) + afs_dec_servers_outstanding(net); } diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index ca08c83168f5..bc9556991d7c 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS File Server client stubs * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/init.h> @@ -14,17 +10,10 @@ #include <linux/sched.h> #include <linux/circ_buf.h> #include <linux/iversion.h> +#include <linux/netfs.h> #include "internal.h" #include "afs_fs.h" #include "xdr_fs.h" -#include "protocol_yfs.h" - -static const struct afs_fid afs_zero_fid; - -static inline void afs_use_fs_server(struct afs_call *call, struct afs_cb_interest *cbi) -{ - call->cbi = afs_get_cb_interest(cbi); -} /* * decode an AFSFid block @@ -60,78 +49,17 @@ static void xdr_dump_bad(const __be32 *bp) } /* - * Update the core inode struct from a returned status record. - */ -void afs_update_inode_from_status(struct afs_vnode *vnode, - struct afs_file_status *status, - const afs_dataversion_t *expected_version, - u8 flags) -{ - struct timespec64 t; - umode_t mode; - - t = status->mtime_client; - vnode->vfs_inode.i_ctime = t; - vnode->vfs_inode.i_mtime = t; - vnode->vfs_inode.i_atime = t; - - if (flags & (AFS_VNODE_META_CHANGED | AFS_VNODE_NOT_YET_SET)) { - vnode->vfs_inode.i_uid = make_kuid(&init_user_ns, status->owner); - vnode->vfs_inode.i_gid = make_kgid(&init_user_ns, status->group); - set_nlink(&vnode->vfs_inode, status->nlink); - - mode = vnode->vfs_inode.i_mode; - mode &= ~S_IALLUGO; - mode |= status->mode; - barrier(); - vnode->vfs_inode.i_mode = mode; - } - - if (!(flags & AFS_VNODE_NOT_YET_SET)) { - if (expected_version && - *expected_version != status->data_version) { - _debug("vnode modified %llx on {%llx:%llu} [exp %llx]", - (unsigned long long) status->data_version, - vnode->fid.vid, vnode->fid.vnode, - (unsigned long long) *expected_version); - vnode->invalid_before = status->data_version; - if (vnode->status.type == AFS_FTYPE_DIR) { - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - afs_stat_v(vnode, n_inval); - } else { - set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); - } - } else if (vnode->status.type == AFS_FTYPE_DIR) { - /* Expected directory change is handled elsewhere so - * that we can locally edit the directory and save on a - * download. - */ - if (test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - flags &= ~AFS_VNODE_DATA_CHANGED; - } - } - - if (flags & (AFS_VNODE_DATA_CHANGED | AFS_VNODE_NOT_YET_SET)) { - inode_set_iversion_raw(&vnode->vfs_inode, status->data_version); - i_size_write(&vnode->vfs_inode, status->size); - } -} - -/* * decode an AFSFetchStatus block */ -static int xdr_decode_AFSFetchStatus(struct afs_call *call, - const __be32 **_bp, - struct afs_file_status *status, - struct afs_vnode *vnode, - const afs_dataversion_t *expected_version, - struct afs_read *read_req) +static void xdr_decode_AFSFetchStatus(const __be32 **_bp, + struct afs_call *call, + struct afs_status_cb *scb) { const struct afs_xdr_AFSFetchStatus *xdr = (const void *)*_bp; + struct afs_file_status *status = &scb->status; bool inline_error = (call->operation_ID == afs_FS_InlineBulkStatus); u64 data_version, size; u32 type, abort_code; - u8 flags = 0; abort_code = ntohl(xdr->abort_code); @@ -144,7 +72,8 @@ static int xdr_decode_AFSFetchStatus(struct afs_call *call, * case. */ status->abort_code = abort_code; - return 0; + scb->have_error = true; + goto advance; } pr_warn("Unknown AFSFetchStatus version %u\n", ntohl(xdr->if_version)); @@ -153,7 +82,8 @@ static int xdr_decode_AFSFetchStatus(struct afs_call *call, if (abort_code != 0 && inline_error) { status->abort_code = abort_code; - return 0; + scb->have_error = true; + goto advance; } type = ntohl(xdr->type); @@ -161,44 +91,25 @@ static int xdr_decode_AFSFetchStatus(struct afs_call *call, case AFS_FTYPE_FILE: case AFS_FTYPE_DIR: case AFS_FTYPE_SYMLINK: - if (type != status->type && - vnode && - !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { - pr_warning("Vnode %llx:%llx:%x changed type %u to %u\n", - vnode->fid.vid, - vnode->fid.vnode, - vnode->fid.unique, - status->type, type); - goto bad; - } status->type = type; break; default: goto bad; } -#define EXTRACT_M(FIELD) \ - do { \ - u32 x = ntohl(xdr->FIELD); \ - if (status->FIELD != x) { \ - flags |= AFS_VNODE_META_CHANGED; \ - status->FIELD = x; \ - } \ - } while (0) - - EXTRACT_M(nlink); - EXTRACT_M(author); - EXTRACT_M(owner); - EXTRACT_M(caller_access); /* call ticket dependent */ - EXTRACT_M(anon_access); - EXTRACT_M(mode); - EXTRACT_M(group); + status->nlink = ntohl(xdr->nlink); + status->author = ntohl(xdr->author); + status->owner = ntohl(xdr->owner); + status->caller_access = ntohl(xdr->caller_access); /* Ticket dependent */ + status->anon_access = ntohl(xdr->anon_access); + status->mode = ntohl(xdr->mode) & S_IALLUGO; + status->group = ntohl(xdr->group); + status->lock_count = ntohl(xdr->lock_count); status->mtime_client.tv_sec = ntohl(xdr->mtime_client); status->mtime_client.tv_nsec = 0; status->mtime_server.tv_sec = ntohl(xdr->mtime_server); status->mtime_server.tv_nsec = 0; - status->lock_count = ntohl(xdr->lock_count); size = (u64)ntohl(xdr->size_lo); size |= (u64)ntohl(xdr->size_hi) << 32; @@ -206,103 +117,34 @@ static int xdr_decode_AFSFetchStatus(struct afs_call *call, data_version = (u64)ntohl(xdr->data_version_lo); data_version |= (u64)ntohl(xdr->data_version_hi) << 32; - if (data_version != status->data_version) { - status->data_version = data_version; - flags |= AFS_VNODE_DATA_CHANGED; - } - - if (read_req) { - read_req->data_version = data_version; - read_req->file_size = size; - } - + status->data_version = data_version; + scb->have_status = true; +advance: *_bp = (const void *)*_bp + sizeof(*xdr); - - if (vnode) { - if (test_bit(AFS_VNODE_UNSET, &vnode->flags)) - flags |= AFS_VNODE_NOT_YET_SET; - afs_update_inode_from_status(vnode, status, expected_version, - flags); - } - - return 0; + return; bad: xdr_dump_bad(*_bp); - return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status); + afs_protocol_error(call, afs_eproto_bad_status); + goto advance; } -/* - * Decode the file status. We need to lock the target vnode if we're going to - * update its status so that stat() sees the attributes update atomically. - */ -static int afs_decode_status(struct afs_call *call, - const __be32 **_bp, - struct afs_file_status *status, - struct afs_vnode *vnode, - const afs_dataversion_t *expected_version, - struct afs_read *read_req) +static time64_t xdr_decode_expiry(struct afs_call *call, u32 expiry) { - int ret; - - if (!vnode) - return xdr_decode_AFSFetchStatus(call, _bp, status, vnode, - expected_version, read_req); - - write_seqlock(&vnode->cb_lock); - ret = xdr_decode_AFSFetchStatus(call, _bp, status, vnode, - expected_version, read_req); - write_sequnlock(&vnode->cb_lock); - return ret; + return ktime_divns(call->issue_time, NSEC_PER_SEC) + expiry; } -/* - * decode an AFSCallBack block - */ -static void xdr_decode_AFSCallBack(struct afs_call *call, - struct afs_vnode *vnode, - const __be32 **_bp) +static void xdr_decode_AFSCallBack(const __be32 **_bp, + struct afs_call *call, + struct afs_status_cb *scb) { - struct afs_cb_interest *old, *cbi = call->cbi; + struct afs_callback *cb = &scb->callback; const __be32 *bp = *_bp; - u32 cb_expiry; - - write_seqlock(&vnode->cb_lock); - - if (!afs_cb_is_broken(call->cb_break, vnode, cbi)) { - vnode->cb_version = ntohl(*bp++); - cb_expiry = ntohl(*bp++); - vnode->cb_type = ntohl(*bp++); - vnode->cb_expires_at = cb_expiry + ktime_get_real_seconds(); - old = vnode->cb_interest; - if (old != call->cbi) { - vnode->cb_interest = cbi; - cbi = old; - } - set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - } else { - bp += 3; - } - - write_sequnlock(&vnode->cb_lock); - call->cbi = cbi; - *_bp = bp; -} -static ktime_t xdr_decode_expiry(struct afs_call *call, u32 expiry) -{ - return ktime_add_ns(call->reply_time, expiry * NSEC_PER_SEC); -} - -static void xdr_decode_AFSCallBack_raw(struct afs_call *call, - const __be32 **_bp, - struct afs_callback *cb) -{ - const __be32 *bp = *_bp; - - cb->version = ntohl(*bp++); + bp++; /* version */ cb->expires_at = xdr_decode_expiry(call, ntohl(*bp++)); - cb->type = ntohl(*bp++); + bp++; /* type */ + scb->have_cb = true; *_bp = bp; } @@ -393,9 +235,10 @@ static void xdr_decode_AFSFetchVolumeStatus(const __be32 **_bp, /* * deliver reply data to an FS.FetchStatus */ -static int afs_deliver_fs_fetch_status_vnode(struct afs_call *call) +static int afs_deliver_fs_fetch_status(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[op->fetch_status.which]; const __be32 *bp; int ret; @@ -403,16 +246,11 @@ static int afs_deliver_fs_fetch_status_vnode(struct afs_call *call) if (ret < 0) return ret; - _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_AFSCallBack(call, vnode, &bp); - xdr_decode_AFSVolSync(&bp, call->reply[1]); + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSCallBack(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -421,54 +259,40 @@ static int afs_deliver_fs_fetch_status_vnode(struct afs_call *call) /* * FS.FetchStatus operation type */ -static const struct afs_call_type afs_RXFSFetchStatus_vnode = { - .name = "FS.FetchStatus(vnode)", +static const struct afs_call_type afs_RXFSFetchStatus = { + .name = "FS.FetchStatus", .op = afs_FS_FetchStatus, - .deliver = afs_deliver_fs_fetch_status_vnode, + .deliver = afs_deliver_fs_fetch_status, .destructor = afs_flat_call_destructor, }; /* * fetch the status information for a file */ -int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsync, - bool new_inode) +void afs_fs_fetch_status(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[op->fetch_status.which]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_fetch_file_status(fc, volsync, new_inode); - _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &afs_RXFSFetchStatus_vnode, + call = afs_alloc_flat_call(op->net, &afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = volsync; - call->expected_version = new_inode ? 1 : vnode->status.data_version; - call->want_reply_time = true; + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSFETCHSTATUS); - bp[1] = htonl(vnode->fid.vid); - bp[2] = htonl(vnode->fid.vnode); - bp[3] = htonl(vnode->fid.unique); - - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + bp[1] = htonl(vp->fid.vid); + bp[2] = htonl(vp->fid.vnode); + bp[3] = htonl(vp->fid.unique); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -476,20 +300,20 @@ int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsy */ static int afs_deliver_fs_fetch_data(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; - struct afs_read *req = call->reply[2]; + struct afs_operation *op = call->op; + struct netfs_io_subrequest *subreq = op->fetch.subreq; + struct afs_vnode_param *vp = &op->file[0]; const __be32 *bp; - unsigned int size; + size_t count_before; int ret; - _enter("{%u,%zu/%llu}", - call->unmarshall, iov_iter_count(&call->iter), req->actual_len); + _enter("{%u,%zu,%zu/%llu}", + call->unmarshall, call->iov_len, iov_iter_count(call->iter), + call->remaining); switch (call->unmarshall) { case 0: - req->actual_len = 0; - req->index = 0; - req->offset = req->pos & (PAGE_SIZE - 1); + call->remaining = 0; call->unmarshall++; if (call->operation_ID == FSFETCHDATA64) { afs_extract_to_tmp64(call); @@ -497,64 +321,52 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) call->tmp_u = htonl(0); afs_extract_to_tmp(call); } + fallthrough; - /* extract the returned data length */ + /* Extract the returned data length into ->remaining. + * This may indicate more or less data than was + * requested will be returned. + */ case 1: _debug("extract data length"); ret = afs_extract_data(call, true); if (ret < 0) return ret; - req->actual_len = be64_to_cpu(call->tmp64); - _debug("DATA length: %llu", req->actual_len); - req->remain = min(req->len, req->actual_len); - if (req->remain == 0) + call->remaining = be64_to_cpu(call->tmp64); + _debug("DATA length: %llu", call->remaining); + + if (call->remaining == 0) goto no_more_data; + call->iter = &subreq->io_iter; + call->iov_len = umin(call->remaining, subreq->len - subreq->transferred); call->unmarshall++; - - begin_page: - ASSERTCMP(req->index, <, req->nr_pages); - if (req->remain > PAGE_SIZE - req->offset) - size = PAGE_SIZE - req->offset; - else - size = req->remain; - call->bvec[0].bv_len = size; - call->bvec[0].bv_offset = req->offset; - call->bvec[0].bv_page = req->pages[req->index]; - iov_iter_bvec(&call->iter, READ, call->bvec, 1, size); - ASSERTCMP(size, <=, PAGE_SIZE); + fallthrough; /* extract the returned data */ case 2: - _debug("extract data %zu/%llu", - iov_iter_count(&call->iter), req->remain); + count_before = call->iov_len; + _debug("extract data %zu/%llu", count_before, call->remaining); ret = afs_extract_data(call, true); + subreq->transferred += count_before - call->iov_len; + call->remaining -= count_before - call->iov_len; if (ret < 0) return ret; - req->remain -= call->bvec[0].bv_len; - req->offset += call->bvec[0].bv_len; - ASSERTCMP(req->offset, <=, PAGE_SIZE); - if (req->offset == PAGE_SIZE) { - req->offset = 0; - if (req->page_done) - req->page_done(call, req); - req->index++; - if (req->remain > 0) - goto begin_page; - } - ASSERTCMP(req->remain, ==, 0); - if (req->actual_len <= req->len) + call->iter = &call->def_iter; + if (call->remaining) goto no_more_data; /* Discard any excess data the server gave us */ - iov_iter_discard(&call->iter, READ, req->actual_len - req->len); + afs_extract_discard(call, call->remaining); call->unmarshall = 3; + fallthrough; + case 3: _debug("extract discard %zu/%llu", - iov_iter_count(&call->iter), req->actual_len - req->len); + iov_iter_count(call->iter), call->remaining); ret = afs_extract_data(call, true); if (ret < 0) @@ -563,6 +375,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) no_more_data: call->unmarshall = 4; afs_extract_to_buf(call, (21 + 3 + 6) * 4); + fallthrough; /* extract the metadata */ case 4: @@ -571,143 +384,111 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) return ret; bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &vnode->status.data_version, req); - if (ret < 0) - return ret; - xdr_decode_AFSCallBack(call, vnode, &bp); - xdr_decode_AFSVolSync(&bp, call->reply[1]); + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSCallBack(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); + + if (subreq->start + subreq->transferred >= vp->scb.status.size) + __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); call->unmarshall++; + fallthrough; case 5: break; } - for (; req->index < req->nr_pages; req->index++) { - if (req->offset < PAGE_SIZE) - zero_user_segment(req->pages[req->index], - req->offset, PAGE_SIZE); - if (req->page_done) - req->page_done(call, req); - req->offset = 0; - } - _leave(" = 0 [done]"); return 0; } -static void afs_fetch_data_destructor(struct afs_call *call) -{ - struct afs_read *req = call->reply[2]; - - afs_put_read(req); - afs_flat_call_destructor(call); -} - /* * FS.FetchData operation type */ static const struct afs_call_type afs_RXFSFetchData = { .name = "FS.FetchData", .op = afs_FS_FetchData, + .async_rx = afs_fetch_data_async_rx, .deliver = afs_deliver_fs_fetch_data, - .destructor = afs_fetch_data_destructor, + .immediate_cancel = afs_fetch_data_immediate_cancel, + .destructor = afs_flat_call_destructor, }; static const struct afs_call_type afs_RXFSFetchData64 = { .name = "FS.FetchData64", .op = afs_FS_FetchData64, + .async_rx = afs_fetch_data_async_rx, .deliver = afs_deliver_fs_fetch_data, - .destructor = afs_fetch_data_destructor, + .immediate_cancel = afs_fetch_data_immediate_cancel, + .destructor = afs_flat_call_destructor, }; /* * fetch data from a very large file */ -static int afs_fs_fetch_data64(struct afs_fs_cursor *fc, struct afs_read *req) +static void afs_fs_fetch_data64(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct netfs_io_subrequest *subreq = op->fetch.subreq; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; _enter(""); - call = afs_alloc_flat_call(net, &afs_RXFSFetchData64, 32, (21 + 3 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSFetchData64, 32, (21 + 3 + 6) * 4); if (!call) - return -ENOMEM; + return afs_op_nomem(op); - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = NULL; /* volsync */ - call->reply[2] = req; - call->expected_version = vnode->status.data_version; - call->want_reply_time = true; + if (op->flags & AFS_OPERATION_ASYNC) + call->async = true; /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSFETCHDATA64); - bp[1] = htonl(vnode->fid.vid); - bp[2] = htonl(vnode->fid.vnode); - bp[3] = htonl(vnode->fid.unique); - bp[4] = htonl(upper_32_bits(req->pos)); - bp[5] = htonl(lower_32_bits(req->pos)); + bp[1] = htonl(vp->fid.vid); + bp[2] = htonl(vp->fid.vnode); + bp[3] = htonl(vp->fid.unique); + bp[4] = htonl(upper_32_bits(subreq->start + subreq->transferred)); + bp[5] = htonl(lower_32_bits(subreq->start + subreq->transferred)); bp[6] = 0; - bp[7] = htonl(lower_32_bits(req->len)); + bp[7] = htonl(lower_32_bits(subreq->len - subreq->transferred)); - refcount_inc(&req->usage); - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * fetch data from a file */ -int afs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) +void afs_fs_fetch_data(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct netfs_io_subrequest *subreq = op->fetch.subreq; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_fetch_data(fc, req); - - if (upper_32_bits(req->pos) || - upper_32_bits(req->len) || - upper_32_bits(req->pos + req->len)) - return afs_fs_fetch_data64(fc, req); + if (test_bit(AFS_SERVER_FL_HAS_FS64, &op->server->flags)) + return afs_fs_fetch_data64(op); _enter(""); - call = afs_alloc_flat_call(net, &afs_RXFSFetchData, 24, (21 + 3 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSFetchData, 24, (21 + 3 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = NULL; /* volsync */ - call->reply[2] = req; - call->expected_version = vnode->status.data_version; - call->want_reply_time = true; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSFETCHDATA); - bp[1] = htonl(vnode->fid.vid); - bp[2] = htonl(vnode->fid.vnode); - bp[3] = htonl(vnode->fid.unique); - bp[4] = htonl(lower_32_bits(req->pos)); - bp[5] = htonl(lower_32_bits(req->len)); - - refcount_inc(&req->usage); - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + bp[1] = htonl(vp->fid.vid); + bp[2] = htonl(vp->fid.vnode); + bp[3] = htonl(vp->fid.unique); + bp[4] = htonl(lower_32_bits(subreq->start + subreq->transferred)); + bp[5] = htonl(lower_32_bits(subreq->len + subreq->transferred)); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -715,28 +496,23 @@ int afs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) */ static int afs_deliver_fs_create_vnode(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; - _enter("{%u}", call->unmarshall); - ret = afs_transfer_reply(call); if (ret < 0) return ret; /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFid(&bp, call->reply[1]); - ret = afs_decode_status(call, &bp, call->reply[2], NULL, NULL, NULL); - if (ret < 0) - return ret; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_AFSCallBack_raw(call, &bp, call->reply[3]); - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + xdr_decode_AFSFid(&bp, &op->file[1].fid); + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_AFSCallBack(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -752,6 +528,53 @@ static const struct afs_call_type afs_RXFSCreateFile = { .destructor = afs_flat_call_destructor, }; +/* + * Create a file. + */ +void afs_fs_create_file(struct afs_operation *op) +{ + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_call *call; + size_t namesz, reqsz, padsz; + __be32 *bp; + + _enter(""); + + namesz = name->len; + padsz = (4 - (namesz & 3)) & 3; + reqsz = (5 * 4) + namesz + padsz + (6 * 4); + + call = afs_alloc_flat_call(op->net, &afs_RXFSCreateFile, + reqsz, (3 + 21 + 21 + 3 + 6) * 4); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSCREATEFILE); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); + *bp++ = htonl(namesz); + memcpy(bp, name->name, namesz); + bp = (void *) bp + namesz; + if (padsz > 0) { + memset(bp, 0, padsz); + bp = (void *) bp + padsz; + } + *bp++ = htonl(AFS_SET_MODE | AFS_SET_MTIME); + *bp++ = htonl(op->mtime.tv_sec); /* mtime */ + *bp++ = 0; /* owner */ + *bp++ = 0; /* group */ + *bp++ = htonl(op->create.mode & S_IALLUGO); /* unix mode */ + *bp++ = 0; /* segment size */ + + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); +} + static const struct afs_call_type afs_RXFSMakeDir = { .name = "FS.MakeDir", .op = afs_FS_MakeDir, @@ -760,169 +583,172 @@ static const struct afs_call_type afs_RXFSMakeDir = { }; /* - * create a file or make a directory + * Create a new directory */ -int afs_fs_create(struct afs_fs_cursor *fc, - const char *name, - umode_t mode, - u64 current_data_version, - struct afs_fid *newfid, - struct afs_file_status *newstatus, - struct afs_callback *newcb) +void afs_fs_make_dir(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); size_t namesz, reqsz, padsz; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)){ - if (S_ISDIR(mode)) - return yfs_fs_make_dir(fc, name, mode, current_data_version, - newfid, newstatus, newcb); - else - return yfs_fs_create_file(fc, name, mode, current_data_version, - newfid, newstatus, newcb); - } - _enter(""); - namesz = strlen(name); + namesz = name->len; padsz = (4 - (namesz & 3)) & 3; reqsz = (5 * 4) + namesz + padsz + (6 * 4); - call = afs_alloc_flat_call( - net, S_ISDIR(mode) ? &afs_RXFSMakeDir : &afs_RXFSCreateFile, - reqsz, (3 + 21 + 21 + 3 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSMakeDir, + reqsz, (3 + 21 + 21 + 3 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = newfid; - call->reply[2] = newstatus; - call->reply[3] = newcb; - call->expected_version = current_data_version + 1; - call->want_reply_time = true; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; - *bp++ = htonl(S_ISDIR(mode) ? FSMAKEDIR : FSCREATEFILE); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(FSMAKEDIR); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); *bp++ = htonl(namesz); - memcpy(bp, name, namesz); + memcpy(bp, name->name, namesz); bp = (void *) bp + namesz; if (padsz > 0) { memset(bp, 0, padsz); bp = (void *) bp + padsz; } *bp++ = htonl(AFS_SET_MODE | AFS_SET_MTIME); - *bp++ = htonl(vnode->vfs_inode.i_mtime.tv_sec); /* mtime */ + *bp++ = htonl(op->mtime.tv_sec); /* mtime */ *bp++ = 0; /* owner */ *bp++ = 0; /* group */ - *bp++ = htonl(mode & S_IALLUGO); /* unix mode */ + *bp++ = htonl(op->create.mode & S_IALLUGO); /* unix mode */ *bp++ = 0; /* segment size */ - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* - * deliver reply data to an FS.RemoveFile or FS.RemoveDir + * Deliver reply data to any operation that returns status and volume sync. */ -static int afs_deliver_fs_remove(struct afs_call *call) +static int afs_deliver_fs_file_status_and_vol(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[0]; const __be32 *bp; int ret; - _enter("{%u}", call->unmarshall); - ret = afs_transfer_reply(call); if (ret < 0) return ret; /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; } /* - * FS.RemoveDir/FS.RemoveFile operation type + * FS.RemoveFile operation type */ static const struct afs_call_type afs_RXFSRemoveFile = { .name = "FS.RemoveFile", .op = afs_FS_RemoveFile, - .deliver = afs_deliver_fs_remove, + .deliver = afs_deliver_fs_file_status_and_vol, .destructor = afs_flat_call_destructor, }; +/* + * Remove a file. + */ +void afs_fs_remove_file(struct afs_operation *op) +{ + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_call *call; + size_t namesz, reqsz, padsz; + __be32 *bp; + + _enter(""); + + namesz = name->len; + padsz = (4 - (namesz & 3)) & 3; + reqsz = (5 * 4) + namesz + padsz; + + call = afs_alloc_flat_call(op->net, &afs_RXFSRemoveFile, + reqsz, (21 + 6) * 4); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSREMOVEFILE); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); + *bp++ = htonl(namesz); + memcpy(bp, name->name, namesz); + bp = (void *) bp + namesz; + if (padsz > 0) { + memset(bp, 0, padsz); + bp = (void *) bp + padsz; + } + + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); +} + static const struct afs_call_type afs_RXFSRemoveDir = { .name = "FS.RemoveDir", .op = afs_FS_RemoveDir, - .deliver = afs_deliver_fs_remove, + .deliver = afs_deliver_fs_file_status_and_vol, .destructor = afs_flat_call_destructor, }; /* - * remove a file or directory + * Remove a directory. */ -int afs_fs_remove(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name, bool isdir, u64 current_data_version) +void afs_fs_remove_dir(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(dvnode); size_t namesz, reqsz, padsz; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_remove(fc, vnode, name, isdir, current_data_version); - _enter(""); - namesz = strlen(name); + namesz = name->len; padsz = (4 - (namesz & 3)) & 3; reqsz = (5 * 4) + namesz + padsz; - call = afs_alloc_flat_call( - net, isdir ? &afs_RXFSRemoveDir : &afs_RXFSRemoveFile, - reqsz, (21 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSRemoveDir, + reqsz, (21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = vnode; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; - *bp++ = htonl(isdir ? FSREMOVEDIR : FSREMOVEFILE); - *bp++ = htonl(dvnode->fid.vid); - *bp++ = htonl(dvnode->fid.vnode); - *bp++ = htonl(dvnode->fid.unique); + *bp++ = htonl(FSREMOVEDIR); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); *bp++ = htonl(namesz); - memcpy(bp, name, namesz); + memcpy(bp, name->name, namesz); bp = (void *) bp + namesz; if (padsz > 0) { memset(bp, 0, padsz); bp = (void *) bp + padsz; } - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -930,7 +756,9 @@ int afs_fs_remove(struct afs_fs_cursor *fc, struct afs_vnode *vnode, */ static int afs_deliver_fs_link(struct afs_call *call) { - struct afs_vnode *dvnode = call->reply[0], *vnode = call->reply[1]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; @@ -942,14 +770,9 @@ static int afs_deliver_fs_link(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, NULL, NULL); - if (ret < 0) - return ret; - ret = afs_decode_status(call, &bp, &dvnode->status, dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -968,53 +791,45 @@ static const struct afs_call_type afs_RXFSLink = { /* * make a hard link */ -int afs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name, u64 current_data_version) +void afs_fs_link(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); size_t namesz, reqsz, padsz; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_link(fc, vnode, name, current_data_version); - _enter(""); - namesz = strlen(name); + namesz = name->len; padsz = (4 - (namesz & 3)) & 3; reqsz = (5 * 4) + namesz + padsz + (3 * 4); - call = afs_alloc_flat_call(net, &afs_RXFSLink, reqsz, (21 + 21 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSLink, reqsz, (21 + 21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = vnode; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSLINK); - *bp++ = htonl(dvnode->fid.vid); - *bp++ = htonl(dvnode->fid.vnode); - *bp++ = htonl(dvnode->fid.unique); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); *bp++ = htonl(namesz); - memcpy(bp, name, namesz); + memcpy(bp, name->name, namesz); bp = (void *) bp + namesz; if (padsz > 0) { memset(bp, 0, padsz); bp = (void *) bp + padsz; } - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call1(call, &vp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1022,7 +837,9 @@ int afs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, */ static int afs_deliver_fs_symlink(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; @@ -1034,15 +851,10 @@ static int afs_deliver_fs_symlink(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFid(&bp, call->reply[1]); - ret = afs_decode_status(call, &bp, call->reply[2], NULL, NULL, NULL); - if (ret < 0) - return ret; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + xdr_decode_AFSFid(&bp, &vp->fid); + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1061,74 +873,59 @@ static const struct afs_call_type afs_RXFSSymlink = { /* * create a symbolic link */ -int afs_fs_symlink(struct afs_fs_cursor *fc, - const char *name, - const char *contents, - u64 current_data_version, - struct afs_fid *newfid, - struct afs_file_status *newstatus) +void afs_fs_symlink(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); size_t namesz, reqsz, padsz, c_namesz, c_padsz; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_symlink(fc, name, contents, current_data_version, - newfid, newstatus); - _enter(""); - namesz = strlen(name); + namesz = name->len; padsz = (4 - (namesz & 3)) & 3; - c_namesz = strlen(contents); + c_namesz = strlen(op->create.symlink); c_padsz = (4 - (c_namesz & 3)) & 3; reqsz = (6 * 4) + namesz + padsz + c_namesz + c_padsz + (6 * 4); - call = afs_alloc_flat_call(net, &afs_RXFSSymlink, reqsz, + call = afs_alloc_flat_call(op->net, &afs_RXFSSymlink, reqsz, (3 + 21 + 21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = newfid; - call->reply[2] = newstatus; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSYMLINK); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); *bp++ = htonl(namesz); - memcpy(bp, name, namesz); + memcpy(bp, name->name, namesz); bp = (void *) bp + namesz; if (padsz > 0) { memset(bp, 0, padsz); bp = (void *) bp + padsz; } *bp++ = htonl(c_namesz); - memcpy(bp, contents, c_namesz); + memcpy(bp, op->create.symlink, c_namesz); bp = (void *) bp + c_namesz; if (c_padsz > 0) { memset(bp, 0, c_padsz); bp = (void *) bp + c_padsz; } *bp++ = htonl(AFS_SET_MODE | AFS_SET_MTIME); - *bp++ = htonl(vnode->vfs_inode.i_mtime.tv_sec); /* mtime */ + *bp++ = htonl(op->mtime.tv_sec); /* mtime */ *bp++ = 0; /* owner */ *bp++ = 0; /* group */ *bp++ = htonl(S_IRWXUGO); /* unix mode */ *bp++ = 0; /* segment size */ - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1136,29 +933,23 @@ int afs_fs_symlink(struct afs_fs_cursor *fc, */ static int afs_deliver_fs_rename(struct afs_call *call) { - struct afs_vnode *orig_dvnode = call->reply[0], *new_dvnode = call->reply[1]; + struct afs_operation *op = call->op; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; const __be32 *bp; int ret; - _enter("{%u}", call->unmarshall); - ret = afs_transfer_reply(call); if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = afs_decode_status(call, &bp, &orig_dvnode->status, orig_dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - if (new_dvnode != orig_dvnode) { - ret = afs_decode_status(call, &bp, &new_dvnode->status, new_dvnode, - &call->expected_version_2, NULL); - if (ret < 0) - return ret; - } - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + /* If the two dirs are the same, we have two copies of the same status + * report, so we just decode it twice. + */ + xdr_decode_AFSFetchStatus(&bp, call, &orig_dvp->scb); + xdr_decode_AFSFetchStatus(&bp, call, &new_dvp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1175,33 +966,24 @@ static const struct afs_call_type afs_RXFSRename = { }; /* - * create a symbolic link + * Rename/move a file or directory. */ -int afs_fs_rename(struct afs_fs_cursor *fc, - const char *orig_name, - struct afs_vnode *new_dvnode, - const char *new_name, - u64 current_orig_data_version, - u64 current_new_data_version) +void afs_fs_rename(struct afs_operation *op) { - struct afs_vnode *orig_dvnode = fc->vnode; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + const struct qstr *orig_name = &op->dentry->d_name; + const struct qstr *new_name = &op->dentry_2->d_name; struct afs_call *call; - struct afs_net *net = afs_v2net(orig_dvnode); size_t reqsz, o_namesz, o_padsz, n_namesz, n_padsz; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_rename(fc, orig_name, - new_dvnode, new_name, - current_orig_data_version, - current_new_data_version); - _enter(""); - o_namesz = strlen(orig_name); + o_namesz = orig_name->len; o_padsz = (4 - (o_namesz & 3)) & 3; - n_namesz = strlen(new_name); + n_namesz = new_name->len; n_padsz = (4 - (n_namesz & 3)) & 3; reqsz = (4 * 4) + @@ -1209,52 +991,47 @@ int afs_fs_rename(struct afs_fs_cursor *fc, (3 * 4) + 4 + n_namesz + n_padsz; - call = afs_alloc_flat_call(net, &afs_RXFSRename, reqsz, (21 + 21 + 6) * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSRename, reqsz, (21 + 21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = orig_dvnode; - call->reply[1] = new_dvnode; - call->expected_version = current_orig_data_version + 1; - call->expected_version_2 = current_new_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSRENAME); - *bp++ = htonl(orig_dvnode->fid.vid); - *bp++ = htonl(orig_dvnode->fid.vnode); - *bp++ = htonl(orig_dvnode->fid.unique); + *bp++ = htonl(orig_dvp->fid.vid); + *bp++ = htonl(orig_dvp->fid.vnode); + *bp++ = htonl(orig_dvp->fid.unique); *bp++ = htonl(o_namesz); - memcpy(bp, orig_name, o_namesz); + memcpy(bp, orig_name->name, o_namesz); bp = (void *) bp + o_namesz; if (o_padsz > 0) { memset(bp, 0, o_padsz); bp = (void *) bp + o_padsz; } - *bp++ = htonl(new_dvnode->fid.vid); - *bp++ = htonl(new_dvnode->fid.vnode); - *bp++ = htonl(new_dvnode->fid.unique); + *bp++ = htonl(new_dvp->fid.vid); + *bp++ = htonl(new_dvp->fid.vnode); + *bp++ = htonl(new_dvp->fid.unique); *bp++ = htonl(n_namesz); - memcpy(bp, new_name, n_namesz); + memcpy(bp, new_name->name, n_namesz); bp = (void *) bp + n_namesz; if (n_padsz > 0) { memset(bp, 0, n_padsz); bp = (void *) bp + n_padsz; } - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &orig_dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = orig_dvp->fid; + trace_afs_make_fs_call2(call, &orig_dvp->fid, orig_name, new_name); + afs_make_op_call(op, call, GFP_NOFS); } /* - * deliver reply data to an FS.StoreData + * Deliver reply data to FS.StoreData or FS.StoreStatus */ static int afs_deliver_fs_store_data(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[0]; const __be32 *bp; int ret; @@ -1266,13 +1043,8 @@ static int afs_deliver_fs_store_data(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ - - afs_pages_written_back(vnode, call); + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1298,162 +1070,98 @@ static const struct afs_call_type afs_RXFSStoreData64 = { /* * store a set of pages to a very large file */ -static int afs_fs_store_data64(struct afs_fs_cursor *fc, - struct address_space *mapping, - pgoff_t first, pgoff_t last, - unsigned offset, unsigned to, - loff_t size, loff_t pos, loff_t i_size) +static void afs_fs_store_data64(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &afs_RXFSStoreData64, + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreData64, (4 + 6 + 3 * 2) * 4, (21 + 6) * 4); if (!call) - return -ENOMEM; + return afs_op_nomem(op); - call->key = fc->key; - call->mapping = mapping; - call->reply[0] = vnode; - call->first = first; - call->last = last; - call->first_offset = offset; - call->last_to = to; - call->send_pages = true; - call->expected_version = vnode->status.data_version + 1; + call->write_iter = op->store.write_iter; /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSTOREDATA64); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); *bp++ = htonl(AFS_SET_MTIME); /* mask */ - *bp++ = htonl(vnode->vfs_inode.i_mtime.tv_sec); /* mtime */ + *bp++ = htonl(op->mtime.tv_sec); /* mtime */ *bp++ = 0; /* owner */ *bp++ = 0; /* group */ *bp++ = 0; /* unix mode */ *bp++ = 0; /* segment size */ - *bp++ = htonl(pos >> 32); - *bp++ = htonl((u32) pos); - *bp++ = htonl(size >> 32); - *bp++ = htonl((u32) size); - *bp++ = htonl(i_size >> 32); - *bp++ = htonl((u32) i_size); + *bp++ = htonl(upper_32_bits(op->store.pos)); + *bp++ = htonl(lower_32_bits(op->store.pos)); + *bp++ = htonl(upper_32_bits(op->store.size)); + *bp++ = htonl(lower_32_bits(op->store.size)); + *bp++ = htonl(upper_32_bits(op->store.i_size)); + *bp++ = htonl(lower_32_bits(op->store.i_size)); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* - * store a set of pages + * Write data to a file on the server. */ -int afs_fs_store_data(struct afs_fs_cursor *fc, struct address_space *mapping, - pgoff_t first, pgoff_t last, - unsigned offset, unsigned to) +void afs_fs_store_data(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - loff_t size, pos, i_size; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_store_data(fc, mapping, first, last, offset, to); - _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); - - size = (loff_t)to - (loff_t)offset; - if (first != last) - size += (loff_t)(last - first) << PAGE_SHIFT; - pos = (loff_t)first << PAGE_SHIFT; - pos += offset; - - i_size = i_size_read(&vnode->vfs_inode); - if (pos + size > i_size) - i_size = size + pos; + key_serial(op->key), vp->fid.vid, vp->fid.vnode); _debug("size %llx, at %llx, i_size %llx", - (unsigned long long) size, (unsigned long long) pos, - (unsigned long long) i_size); + (unsigned long long)op->store.size, + (unsigned long long)op->store.pos, + (unsigned long long)op->store.i_size); - if (pos >> 32 || i_size >> 32 || size >> 32 || (pos + size) >> 32) - return afs_fs_store_data64(fc, mapping, first, last, offset, to, - size, pos, i_size); + if (test_bit(AFS_SERVER_FL_HAS_FS64, &op->server->flags)) + return afs_fs_store_data64(op); - call = afs_alloc_flat_call(net, &afs_RXFSStoreData, + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreData, (4 + 6 + 3) * 4, (21 + 6) * 4); if (!call) - return -ENOMEM; + return afs_op_nomem(op); - call->key = fc->key; - call->mapping = mapping; - call->reply[0] = vnode; - call->first = first; - call->last = last; - call->first_offset = offset; - call->last_to = to; - call->send_pages = true; - call->expected_version = vnode->status.data_version + 1; + call->write_iter = op->store.write_iter; /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSTOREDATA); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); *bp++ = htonl(AFS_SET_MTIME); /* mask */ - *bp++ = htonl(vnode->vfs_inode.i_mtime.tv_sec); /* mtime */ + *bp++ = htonl(op->mtime.tv_sec); /* mtime */ *bp++ = 0; /* owner */ *bp++ = 0; /* group */ *bp++ = 0; /* unix mode */ *bp++ = 0; /* segment size */ - *bp++ = htonl(pos); - *bp++ = htonl(size); - *bp++ = htonl(i_size); - - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); -} - -/* - * deliver reply data to an FS.StoreStatus - */ -static int afs_deliver_fs_store_status(struct afs_call *call) -{ - struct afs_vnode *vnode = call->reply[0]; - const __be32 *bp; - int ret; - - _enter(""); - - ret = afs_transfer_reply(call); - if (ret < 0) - return ret; - - /* unmarshall the reply once we've received all of it */ - bp = call->buffer; - ret = afs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + *bp++ = htonl(lower_32_bits(op->store.pos)); + *bp++ = htonl(lower_32_bits(op->store.size)); + *bp++ = htonl(lower_32_bits(op->store.i_size)); - _leave(" = 0 [done]"); - return 0; + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1462,21 +1170,21 @@ static int afs_deliver_fs_store_status(struct afs_call *call) static const struct afs_call_type afs_RXFSStoreStatus = { .name = "FS.StoreStatus", .op = afs_FS_StoreStatus, - .deliver = afs_deliver_fs_store_status, + .deliver = afs_deliver_fs_store_data, .destructor = afs_flat_call_destructor, }; static const struct afs_call_type afs_RXFSStoreData_as_Status = { .name = "FS.StoreData", .op = afs_FS_StoreData, - .deliver = afs_deliver_fs_store_status, + .deliver = afs_deliver_fs_store_data, .destructor = afs_flat_call_destructor, }; static const struct afs_call_type afs_RXFSStoreData64_as_Status = { .name = "FS.StoreData64", .op = afs_FS_StoreData64, - .deliver = afs_deliver_fs_store_status, + .deliver = afs_deliver_fs_store_data, .destructor = afs_flat_call_destructor, }; @@ -1484,137 +1192,122 @@ static const struct afs_call_type afs_RXFSStoreData64_as_Status = { * set the attributes on a very large file, using FS.StoreData rather than * FS.StoreStatus so as to alter the file size also */ -static int afs_fs_setattr_size64(struct afs_fs_cursor *fc, struct iattr *attr) +static void afs_fs_setattr_size64(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); + struct iattr *attr = op->setattr.attr; __be32 *bp; _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); ASSERT(attr->ia_valid & ATTR_SIZE); - call = afs_alloc_flat_call(net, &afs_RXFSStoreData64_as_Status, + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreData64_as_Status, (4 + 6 + 3 * 2) * 4, (21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->expected_version = vnode->status.data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSTOREDATA64); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); xdr_encode_AFS_StoreStatus(&bp, attr); - *bp++ = 0; /* position of start of write */ + *bp++ = htonl(upper_32_bits(attr->ia_size)); /* position of start of write */ + *bp++ = htonl(lower_32_bits(attr->ia_size)); + *bp++ = 0; /* size of write */ *bp++ = 0; - *bp++ = 0; /* size of write */ - *bp++ = 0; - *bp++ = htonl(attr->ia_size >> 32); /* new file length */ - *bp++ = htonl((u32) attr->ia_size); + *bp++ = htonl(upper_32_bits(attr->ia_size)); /* new file length */ + *bp++ = htonl(lower_32_bits(attr->ia_size)); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * set the attributes on a file, using FS.StoreData rather than FS.StoreStatus * so as to alter the file size also */ -static int afs_fs_setattr_size(struct afs_fs_cursor *fc, struct iattr *attr) +static void afs_fs_setattr_size(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); + struct iattr *attr = op->setattr.attr; __be32 *bp; _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); ASSERT(attr->ia_valid & ATTR_SIZE); - if (attr->ia_size >> 32) - return afs_fs_setattr_size64(fc, attr); + if (test_bit(AFS_SERVER_FL_HAS_FS64, &op->server->flags)) + return afs_fs_setattr_size64(op); - call = afs_alloc_flat_call(net, &afs_RXFSStoreData_as_Status, + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreData_as_Status, (4 + 6 + 3) * 4, (21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->expected_version = vnode->status.data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSTOREDATA); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); xdr_encode_AFS_StoreStatus(&bp, attr); - *bp++ = 0; /* position of start of write */ + *bp++ = htonl(attr->ia_size); /* position of start of write */ *bp++ = 0; /* size of write */ *bp++ = htonl(attr->ia_size); /* new file length */ - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * set the attributes on a file, using FS.StoreData if there's a change in file * size, and FS.StoreStatus otherwise */ -int afs_fs_setattr(struct afs_fs_cursor *fc, struct iattr *attr) +void afs_fs_setattr(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); + struct iattr *attr = op->setattr.attr; __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_setattr(fc, attr); - if (attr->ia_valid & ATTR_SIZE) - return afs_fs_setattr_size(fc, attr); + return afs_fs_setattr_size(op); _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &afs_RXFSStoreStatus, + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreStatus, (4 + 6) * 4, (21 + 6) * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->expected_version = vnode->status.data_version; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSTORESTATUS); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); - xdr_encode_AFS_StoreStatus(&bp, attr); + xdr_encode_AFS_StoreStatus(&bp, op->setattr.attr); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1622,6 +1315,7 @@ int afs_fs_setattr(struct afs_fs_cursor *fc, struct iattr *attr) */ static int afs_deliver_fs_get_volume_status(struct afs_call *call) { + struct afs_operation *op = call->op; const __be32 *bp; char *p; u32 size; @@ -1633,6 +1327,7 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) case 0: call->unmarshall++; afs_extract_to_buf(call, 12 * 4); + fallthrough; /* extract the returned status record */ case 1: @@ -1642,9 +1337,10 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) return ret; bp = call->buffer; - xdr_decode_AFSFetchVolumeStatus(&bp, call->reply[1]); + xdr_decode_AFSFetchVolumeStatus(&bp, &op->volstatus.vs); call->unmarshall++; afs_extract_to_tmp(call); + fallthrough; /* extract the volume name length */ case 2: @@ -1655,11 +1351,11 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("volname length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_volname_len); + return afs_protocol_error(call, afs_eproto_volname_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the volume name */ case 3: @@ -1668,11 +1364,12 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("volname '%s'", p); afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* extract the offline message length */ case 4: @@ -1683,11 +1380,11 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("offline msg length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_offline_msg_len); + return afs_protocol_error(call, afs_eproto_offline_msg_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the offline message */ case 5: @@ -1696,12 +1393,13 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("offline '%s'", p); afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* extract the message of the day length */ case 6: @@ -1712,11 +1410,11 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("motd length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_motd_len); + return afs_protocol_error(call, afs_eproto_motd_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the message of the day */ case 7: @@ -1725,11 +1423,12 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("motd '%s'", p); call->unmarshall++; + fallthrough; case 8: break; @@ -1740,65 +1439,39 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) } /* - * destroy an FS.GetVolumeStatus call - */ -static void afs_get_volume_status_call_destructor(struct afs_call *call) -{ - kfree(call->reply[2]); - call->reply[2] = NULL; - afs_flat_call_destructor(call); -} - -/* * FS.GetVolumeStatus operation type */ static const struct afs_call_type afs_RXFSGetVolumeStatus = { .name = "FS.GetVolumeStatus", .op = afs_FS_GetVolumeStatus, .deliver = afs_deliver_fs_get_volume_status, - .destructor = afs_get_volume_status_call_destructor, + .destructor = afs_flat_call_destructor, }; /* * fetch the status of a volume */ -int afs_fs_get_volume_status(struct afs_fs_cursor *fc, - struct afs_volume_status *vs) +void afs_fs_get_volume_status(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - void *tmpbuf; - - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_get_volume_status(fc, vs); _enter(""); - tmpbuf = kmalloc(AFSOPAQUEMAX, GFP_KERNEL); - if (!tmpbuf) - return -ENOMEM; - - call = afs_alloc_flat_call(net, &afs_RXFSGetVolumeStatus, 2 * 4, 12 * 4); - if (!call) { - kfree(tmpbuf); - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = vs; - call->reply[2] = tmpbuf; + call = afs_alloc_flat_call(op->net, &afs_RXFSGetVolumeStatus, 2 * 4, + max(12 * 4, AFSOPAQUEMAX + 1)); + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSGETVOLUMESTATUS); - bp[1] = htonl(vnode->fid.vid); + bp[1] = htonl(vp->fid.vid); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1806,6 +1479,7 @@ int afs_fs_get_volume_status(struct afs_fs_cursor *fc, */ static int afs_deliver_fs_xxxx_lock(struct afs_call *call) { + struct afs_operation *op = call->op; const __be32 *bp; int ret; @@ -1817,7 +1491,7 @@ static int afs_deliver_fs_xxxx_lock(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ + xdr_decode_AFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1830,6 +1504,7 @@ static const struct afs_call_type afs_RXFSSetLock = { .name = "FS.SetLock", .op = afs_FS_SetLock, .deliver = afs_deliver_fs_xxxx_lock, + .done = afs_lock_op_done, .destructor = afs_flat_call_destructor, }; @@ -1840,6 +1515,7 @@ static const struct afs_call_type afs_RXFSExtendLock = { .name = "FS.ExtendLock", .op = afs_FS_ExtendLock, .deliver = afs_deliver_fs_xxxx_lock, + .done = afs_lock_op_done, .destructor = afs_flat_call_destructor, }; @@ -1856,104 +1532,83 @@ static const struct afs_call_type afs_RXFSReleaseLock = { /* * Set a lock on a file */ -int afs_fs_set_lock(struct afs_fs_cursor *fc, afs_lock_type_t type) +void afs_fs_set_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_set_lock(fc, type); - _enter(""); - call = afs_alloc_flat_call(net, &afs_RXFSSetLock, 5 * 4, 6 * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSSetLock, 5 * 4, 6 * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSSETLOCK); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); - *bp++ = htonl(type); - - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); + *bp++ = htonl(op->lock.type); + + call->fid = vp->fid; + trace_afs_make_fs_calli(call, &vp->fid, op->lock.type); + afs_make_op_call(op, call, GFP_NOFS); } /* * extend a lock on a file */ -int afs_fs_extend_lock(struct afs_fs_cursor *fc) +void afs_fs_extend_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_extend_lock(fc); - _enter(""); - call = afs_alloc_flat_call(net, &afs_RXFSExtendLock, 4 * 4, 6 * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSExtendLock, 4 * 4, 6 * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSEXTENDLOCK); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * release a lock on a file */ -int afs_fs_release_lock(struct afs_fs_cursor *fc) +void afs_fs_release_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_release_lock(fc); - _enter(""); - call = afs_alloc_flat_call(net, &afs_RXFSReleaseLock, 4 * 4, 6 * 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSReleaseLock, 4 * 4, 6 * 4); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSRELEASELOCK); - *bp++ = htonl(vnode->fid.vid); - *bp++ = htonl(vnode->fid.vnode); - *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1977,13 +1632,12 @@ static const struct afs_call_type afs_RXFSGiveUpAllCallBacks = { /* * Flush all the callbacks we have on a server. */ -int afs_fs_give_up_all_callbacks(struct afs_net *net, - struct afs_server *server, - struct afs_addr_cursor *ac, - struct key *key) +int afs_fs_give_up_all_callbacks(struct afs_net *net, struct afs_server *server, + struct afs_address *addr, struct key *key) { struct afs_call *call; __be32 *bp; + int ret; _enter(""); @@ -1991,14 +1645,22 @@ int afs_fs_give_up_all_callbacks(struct afs_net *net, if (!call) return -ENOMEM; - call->key = key; + call->key = key; + call->peer = rxrpc_kernel_get_peer(addr->peer); + call->service_id = server->service_id; /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSGIVEUPALLCALLBACKS); - /* Can't take a ref on server */ - return afs_make_call(ac, call, GFP_NOFS, false); + call->server = afs_use_server(server, false, afs_server_trace_use_give_up_cb); + afs_make_call(call, GFP_NOFS); + afs_wait_for_call_to_complete(call); + ret = call->error; + if (call->responded) + set_bit(AFS_SERVER_FL_RESPONDING, &server->flags); + afs_put_call(call); + return ret; } /* @@ -2009,12 +1671,13 @@ static int afs_deliver_fs_get_capabilities(struct afs_call *call) u32 count; int ret; - _enter("{%u,%zu}", call->unmarshall, iov_iter_count(&call->iter)); + _enter("{%u,%zu}", call->unmarshall, iov_iter_count(call->iter)); switch (call->unmarshall) { case 0: afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* Extract the capabilities word count */ case 1: @@ -2023,19 +1686,33 @@ static int afs_deliver_fs_get_capabilities(struct afs_call *call) return ret; count = ntohl(call->tmp); - call->count = count; call->count2 = count; - iov_iter_discard(&call->iter, READ, count * sizeof(__be32)); + if (count == 0) { + call->unmarshall = 4; + call->tmp = 0; + break; + } + + /* Extract the first word of the capabilities to call->tmp */ + afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; - /* Extract capabilities words */ case 2: ret = afs_extract_data(call, false); if (ret < 0) return ret; - /* TODO: Examine capabilities */ + afs_extract_discard(call, (count - 1) * sizeof(__be32)); + call->unmarshall++; + fallthrough; + + /* Extract remaining capabilities words */ + case 3: + ret = afs_extract_data(call, false); + if (ret < 0) + return ret; call->unmarshall++; break; @@ -2045,11 +1722,9 @@ static int afs_deliver_fs_get_capabilities(struct afs_call *call) return 0; } -static void afs_destroy_fs_get_capabilities(struct afs_call *call) +static void afs_fs_get_capabilities_destructor(struct afs_call *call) { - struct afs_server *server = call->reply[0]; - - afs_put_server(call->net, server); + afs_put_endpoint_state(call->probe, afs_estate_trace_put_getcaps); afs_flat_call_destructor(call); } @@ -2061,19 +1736,19 @@ static const struct afs_call_type afs_RXFSGetCapabilities = { .op = afs_FS_GetCapabilities, .deliver = afs_deliver_fs_get_capabilities, .done = afs_fileserver_probe_result, - .destructor = afs_destroy_fs_get_capabilities, + .immediate_cancel = afs_fileserver_probe_result, + .destructor = afs_fs_get_capabilities_destructor, }; /* - * Probe a fileserver for the capabilities that it supports. This can - * return up to 196 words. - */ -int afs_fs_get_capabilities(struct afs_net *net, - struct afs_server *server, - struct afs_addr_cursor *ac, - struct key *key, - unsigned int server_index, - bool async) + * Probe a fileserver for the capabilities that it supports. This RPC can + * reply with up to 196 words. The operation is asynchronous and if we managed + * to allocate a call, true is returned the result is delivered through the + * ->done() - otherwise we return false to indicate we didn't even try. + */ +bool afs_fs_get_capabilities(struct afs_net *net, struct afs_server *server, + struct afs_endpoint_state *estate, unsigned int addr_index, + struct key *key) { struct afs_call *call; __be32 *bp; @@ -2082,108 +1757,26 @@ int afs_fs_get_capabilities(struct afs_net *net, call = afs_alloc_flat_call(net, &afs_RXFSGetCapabilities, 1 * 4, 16 * 4); if (!call) - return -ENOMEM; - - call->key = key; - call->reply[0] = afs_get_server(server); - call->reply[1] = (void *)(long)server_index; - call->upgrade = true; - call->want_reply_time = true; + return false; + + call->key = key; + call->server = afs_use_server(server, false, afs_server_trace_use_get_caps); + call->peer = rxrpc_kernel_get_peer(estate->addresses->addrs[addr_index].peer); + call->probe = afs_get_endpoint_state(estate, afs_estate_trace_get_getcaps); + call->probe_index = addr_index; + call->service_id = server->service_id; + call->upgrade = true; + call->async = true; + call->max_lifespan = AFS_PROBE_MAX_LIFESPAN; /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSGETCAPABILITIES); - /* Can't take a ref on server */ trace_afs_make_fs_call(call, NULL); - return afs_make_call(ac, call, GFP_NOFS, async); -} - -/* - * Deliver reply data to an FS.FetchStatus with no vnode. - */ -static int afs_deliver_fs_fetch_status(struct afs_call *call) -{ - struct afs_file_status *status = call->reply[1]; - struct afs_callback *callback = call->reply[2]; - struct afs_volsync *volsync = call->reply[3]; - struct afs_fid *fid = call->reply[0]; - const __be32 *bp; - int ret; - - ret = afs_transfer_reply(call); - if (ret < 0) - return ret; - - _enter("{%llx:%llu}", fid->vid, fid->vnode); - - /* unmarshall the reply once we've received all of it */ - bp = call->buffer; - ret = afs_decode_status(call, &bp, status, NULL, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_AFSCallBack_raw(call, &bp, callback); - xdr_decode_AFSVolSync(&bp, volsync); - - _leave(" = 0 [done]"); - return 0; -} - -/* - * FS.FetchStatus operation type - */ -static const struct afs_call_type afs_RXFSFetchStatus = { - .name = "FS.FetchStatus", - .op = afs_FS_FetchStatus, - .deliver = afs_deliver_fs_fetch_status, - .destructor = afs_flat_call_destructor, -}; - -/* - * Fetch the status information for a fid without needing a vnode handle. - */ -int afs_fs_fetch_status(struct afs_fs_cursor *fc, - struct afs_net *net, - struct afs_fid *fid, - struct afs_file_status *status, - struct afs_callback *callback, - struct afs_volsync *volsync) -{ - struct afs_call *call; - __be32 *bp; - - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_fetch_status(fc, net, fid, status, callback, volsync); - - _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), fid->vid, fid->vnode); - - call = afs_alloc_flat_call(net, &afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = fid; - call->reply[1] = status; - call->reply[2] = callback; - call->reply[3] = volsync; - call->expected_version = 1; /* vnode->status.data_version */ - call->want_reply_time = true; - - /* marshall the parameters */ - bp = call->request; - bp[0] = htonl(FSFETCHSTATUS); - bp[1] = htonl(fid->vid); - bp[2] = htonl(fid->vnode); - bp[3] = htonl(fid->unique); - - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + afs_make_call(call, GFP_NOFS); + afs_put_call(call); + return true; } /* @@ -2191,9 +1784,8 @@ int afs_fs_fetch_status(struct afs_fs_cursor *fc, */ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) { - struct afs_file_status *statuses; - struct afs_callback *callbacks; - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_status_cb *scb; const __be32 *bp; u32 tmp; int ret; @@ -2204,6 +1796,7 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) case 0: afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* Extract the file status count and array in two steps */ case 1: @@ -2213,15 +1806,15 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; tmp = ntohl(call->tmp); - _debug("status count: %u/%u", tmp, call->count2); - if (tmp != call->count2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_ibulkst_count); + _debug("status count: %u/%u", tmp, op->nr_files); + if (tmp != op->nr_files) + return afs_protocol_error(call, afs_eproto_ibulkst_count); call->count = 0; call->unmarshall++; more_counts: afs_extract_to_buf(call, 21 * sizeof(__be32)); + fallthrough; case 2: _debug("extract status array %u", call->count); @@ -2229,21 +1822,29 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) if (ret < 0) return ret; + switch (call->count) { + case 0: + scb = &op->file[0].scb; + break; + case 1: + scb = &op->file[1].scb; + break; + default: + scb = &op->more_files[call->count - 2].scb; + break; + } + bp = call->buffer; - statuses = call->reply[1]; - ret = afs_decode_status(call, &bp, &statuses[call->count], - call->count == 0 ? vnode : NULL, - NULL, NULL); - if (ret < 0) - return ret; + xdr_decode_AFSFetchStatus(&bp, call, scb); call->count++; - if (call->count < call->count2) + if (call->count < op->nr_files) goto more_counts; call->count = 0; call->unmarshall++; afs_extract_to_tmp(call); + fallthrough; /* Extract the callback count and array in two steps */ case 3: @@ -2254,13 +1855,13 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) tmp = ntohl(call->tmp); _debug("CB count: %u", tmp); - if (tmp != call->count2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_ibulkst_cb_count); + if (tmp != op->nr_files) + return afs_protocol_error(call, afs_eproto_ibulkst_cb_count); call->count = 0; call->unmarshall++; more_cbs: afs_extract_to_buf(call, 3 * sizeof(__be32)); + fallthrough; case 4: _debug("extract CB array"); @@ -2269,20 +1870,27 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; _debug("unmarshall CB array"); + switch (call->count) { + case 0: + scb = &op->file[0].scb; + break; + case 1: + scb = &op->file[1].scb; + break; + default: + scb = &op->more_files[call->count - 2].scb; + break; + } + bp = call->buffer; - callbacks = call->reply[2]; - callbacks[call->count].version = ntohl(bp[0]); - callbacks[call->count].expires_at = xdr_decode_expiry(call, ntohl(bp[1])); - callbacks[call->count].type = ntohl(bp[2]); - statuses = call->reply[1]; - if (call->count == 0 && vnode && statuses[0].abort_code == 0) - xdr_decode_AFSCallBack(call, vnode, &bp); + xdr_decode_AFSCallBack(&bp, call, scb); call->count++; - if (call->count < call->count2) + if (call->count < op->nr_files) goto more_cbs; afs_extract_to_buf(call, 6 * sizeof(__be32)); call->unmarshall++; + fallthrough; case 5: ret = afs_extract_data(call, false); @@ -2290,9 +1898,13 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; bp = call->buffer; - xdr_decode_AFSVolSync(&bp, call->reply[3]); + /* Unfortunately, prior to OpenAFS-1.6, volsync here is filled + * with rubbish. + */ + xdr_decode_AFSVolSync(&bp, NULL); call->unmarshall++; + fallthrough; case 6: break; @@ -2302,6 +1914,16 @@ static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) return 0; } +static void afs_done_fs_inline_bulk_status(struct afs_call *call) +{ + if (call->error == -ECONNABORTED && + call->abort_code == RX_INVALID_OPERATION) { + set_bit(AFS_SERVER_FL_NO_IBULK, &call->server->flags); + if (call->op) + set_bit(AFS_VOLUME_MAYBE_NO_IBULK, &call->op->volume->flags); + } +} + /* * FS.InlineBulkStatus operation type */ @@ -2309,59 +1931,204 @@ static const struct afs_call_type afs_RXFSInlineBulkStatus = { .name = "FS.InlineBulkStatus", .op = afs_FS_InlineBulkStatus, .deliver = afs_deliver_fs_inline_bulk_status, + .done = afs_done_fs_inline_bulk_status, .destructor = afs_flat_call_destructor, }; /* * Fetch the status information for up to 50 files */ -int afs_fs_inline_bulk_status(struct afs_fs_cursor *fc, - struct afs_net *net, - struct afs_fid *fids, - struct afs_file_status *statuses, - struct afs_callback *callbacks, - unsigned int nr_fids, - struct afs_volsync *volsync) +void afs_fs_inline_bulk_status(struct afs_operation *op) { + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_call *call; __be32 *bp; int i; - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc->cbi->server->flags)) - return yfs_fs_inline_bulk_status(fc, net, fids, statuses, callbacks, - nr_fids, volsync); + if (test_bit(AFS_SERVER_FL_NO_IBULK, &op->server->flags)) { + afs_op_set_error(op, -ENOTSUPP); + return; + } _enter(",%x,{%llx:%llu},%u", - key_serial(fc->key), fids[0].vid, fids[1].vnode, nr_fids); + key_serial(op->key), vp->fid.vid, vp->fid.vnode, op->nr_files); - call = afs_alloc_flat_call(net, &afs_RXFSInlineBulkStatus, - (2 + nr_fids * 3) * 4, + call = afs_alloc_flat_call(op->net, &afs_RXFSInlineBulkStatus, + (2 + op->nr_files * 3) * 4, 21 * 4); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = NULL; /* vnode for fid[0] */ - call->reply[1] = statuses; - call->reply[2] = callbacks; - call->reply[3] = volsync; - call->count2 = nr_fids; - call->want_reply_time = true; + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; *bp++ = htonl(FSINLINEBULKSTATUS); - *bp++ = htonl(nr_fids); - for (i = 0; i < nr_fids; i++) { - *bp++ = htonl(fids[i].vid); - *bp++ = htonl(fids[i].vnode); - *bp++ = htonl(fids[i].unique); + *bp++ = htonl(op->nr_files); + *bp++ = htonl(dvp->fid.vid); + *bp++ = htonl(dvp->fid.vnode); + *bp++ = htonl(dvp->fid.unique); + *bp++ = htonl(vp->fid.vid); + *bp++ = htonl(vp->fid.vnode); + *bp++ = htonl(vp->fid.unique); + for (i = 0; i < op->nr_files - 2; i++) { + *bp++ = htonl(op->more_files[i].fid.vid); + *bp++ = htonl(op->more_files[i].fid.vnode); + *bp++ = htonl(op->more_files[i].fid.unique); + } + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); +} + +/* + * deliver reply data to an FS.FetchACL + */ +static int afs_deliver_fs_fetch_acl(struct afs_call *call) +{ + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[0]; + struct afs_acl *acl; + const __be32 *bp; + unsigned int size; + int ret; + + _enter("{%u}", call->unmarshall); + + switch (call->unmarshall) { + case 0: + afs_extract_to_tmp(call); + call->unmarshall++; + fallthrough; + + /* extract the returned data length */ + case 1: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + size = call->count2 = ntohl(call->tmp); + size = round_up(size, 4); + + acl = kmalloc(struct_size(acl, data, size), GFP_KERNEL); + if (!acl) + return -ENOMEM; + op->acl = acl; + acl->size = call->count2; + afs_extract_begin(call, acl->data, size); + call->unmarshall++; + fallthrough; + + /* extract the returned data */ + case 2: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + afs_extract_to_buf(call, (21 + 6) * 4); + call->unmarshall++; + fallthrough; + + /* extract the metadata */ + case 3: + ret = afs_extract_data(call, false); + if (ret < 0) + return ret; + + bp = call->buffer; + xdr_decode_AFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_AFSVolSync(&bp, &op->volsync); + + call->unmarshall++; + fallthrough; + + case 4: + break; } - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &fids[0]); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.FetchACL operation type + */ +static const struct afs_call_type afs_RXFSFetchACL = { + .name = "FS.FetchACL", + .op = afs_FS_FetchACL, + .deliver = afs_deliver_fs_fetch_acl, +}; + +/* + * Fetch the ACL for a file. + */ +void afs_fs_fetch_acl(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%llx:%llu},,", + key_serial(op->key), vp->fid.vid, vp->fid.vnode); + + call = afs_alloc_flat_call(op->net, &afs_RXFSFetchACL, 16, (21 + 6) * 4); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + bp[0] = htonl(FSFETCHACL); + bp[1] = htonl(vp->fid.vid); + bp[2] = htonl(vp->fid.vnode); + bp[3] = htonl(vp->fid.unique); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_KERNEL); +} + +/* + * FS.StoreACL operation type + */ +static const struct afs_call_type afs_RXFSStoreACL = { + .name = "FS.StoreACL", + .op = afs_FS_StoreACL, + .deliver = afs_deliver_fs_file_status_and_vol, + .destructor = afs_flat_call_destructor, +}; + +/* + * Fetch the ACL for a file. + */ +void afs_fs_store_acl(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_call *call; + const struct afs_acl *acl = op->acl; + size_t size; + __be32 *bp; + + _enter(",%x,{%llx:%llu},,", + key_serial(op->key), vp->fid.vid, vp->fid.vnode); + + size = round_up(acl->size, 4); + call = afs_alloc_flat_call(op->net, &afs_RXFSStoreACL, + 5 * 4 + size, (21 + 6) * 4); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + bp[0] = htonl(FSSTOREACL); + bp[1] = htonl(vp->fid.vid); + bp[2] = htonl(vp->fid.vnode); + bp[3] = htonl(vp->fid.unique); + bp[4] = htonl(acl->size); + memcpy(&bp[5], acl->data, acl->size); + if (acl->size != size) + memset((void *)&bp[5] + acl->size, 0, size - acl->size); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_KERNEL); } diff --git a/fs/afs/inode.c b/fs/afs/inode.c index 1a4ce07fb406..dde1857fcabb 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -23,47 +23,186 @@ #include <linux/namei.h> #include <linux/iversion.h> #include "internal.h" +#include "afs_fs.h" + +void afs_init_new_symlink(struct afs_vnode *vnode, struct afs_operation *op) +{ + size_t size = strlen(op->create.symlink) + 1; + size_t dsize = 0; + char *p; + + if (netfs_alloc_folioq_buffer(NULL, &vnode->directory, &dsize, size, + mapping_gfp_mask(vnode->netfs.inode.i_mapping)) < 0) + return; + + vnode->directory_size = dsize; + p = kmap_local_folio(folioq_folio(vnode->directory, 0), 0); + memcpy(p, op->create.symlink, size); + kunmap_local(p); + set_bit(AFS_VNODE_DIR_READ, &vnode->flags); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); +} + +static void afs_put_link(void *arg) +{ + struct folio *folio = virt_to_folio(arg); + + kunmap_local(arg); + folio_put(folio); +} + +const char *afs_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback) +{ + struct afs_vnode *vnode = AFS_FS_I(inode); + struct folio *folio; + char *content; + ssize_t ret; + + if (!dentry) { + /* RCU pathwalk. */ + if (!test_bit(AFS_VNODE_DIR_READ, &vnode->flags) || !afs_check_validity(vnode)) + return ERR_PTR(-ECHILD); + goto good; + } + + if (test_bit(AFS_VNODE_DIR_READ, &vnode->flags)) + goto fetch; + + ret = afs_validate(vnode, NULL); + if (ret < 0) + return ERR_PTR(ret); + + if (!test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) && + test_bit(AFS_VNODE_DIR_READ, &vnode->flags)) + goto good; + +fetch: + ret = afs_read_single(vnode, NULL); + if (ret < 0) + return ERR_PTR(ret); + set_bit(AFS_VNODE_DIR_READ, &vnode->flags); + +good: + folio = folioq_folio(vnode->directory, 0); + folio_get(folio); + content = kmap_local_folio(folio, 0); + set_delayed_call(callback, afs_put_link, content); + return content; +} + +int afs_readlink(struct dentry *dentry, char __user *buffer, int buflen) +{ + DEFINE_DELAYED_CALL(done); + const char *content; + int len; + + content = afs_get_link(dentry, d_inode(dentry), &done); + if (IS_ERR(content)) { + do_delayed_call(&done); + return PTR_ERR(content); + } + + len = umin(strlen(content), buflen); + if (copy_to_user(buffer, content, len)) + len = -EFAULT; + do_delayed_call(&done); + return len; +} static const struct inode_operations afs_symlink_inode_operations = { - .get_link = page_get_link, - .listxattr = afs_listxattr, + .get_link = afs_get_link, + .readlink = afs_readlink, }; +static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *parent_vnode) +{ + static unsigned long once_only; + + pr_warn("kAFS: AFS vnode with undefined type %u\n", vnode->status.type); + pr_warn("kAFS: A=%d m=%o s=%llx v=%llx\n", + vnode->status.abort_code, + vnode->status.mode, + vnode->status.size, + vnode->status.data_version); + pr_warn("kAFS: vnode %llx:%llx:%x\n", + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique); + if (parent_vnode) + pr_warn("kAFS: dir %llx:%llx:%x\n", + parent_vnode->fid.vid, + parent_vnode->fid.vnode, + parent_vnode->fid.unique); + + if (!test_and_set_bit(0, &once_only)) + dump_stack(); +} + +/* + * Set parameters for the netfs library + */ +static void afs_set_netfs_context(struct afs_vnode *vnode) +{ + netfs_inode_init(&vnode->netfs, &afs_req_ops, true); +} + /* * Initialise an inode from the vnode status. */ -static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key) +static int afs_inode_init_from_status(struct afs_operation *op, + struct afs_vnode_param *vp, + struct afs_vnode *vnode) { + struct afs_file_status *status = &vp->scb.status; struct inode *inode = AFS_VNODE_TO_I(vnode); + struct timespec64 t; + + _enter("{%llx:%llu.%u} %s", + vp->fid.vid, vp->fid.vnode, vp->fid.unique, + op->type ? op->type->name : "???"); _debug("FS: ft=%d lk=%d sz=%llu ver=%Lu mod=%hu", - vnode->status.type, - vnode->status.nlink, - (unsigned long long) vnode->status.size, - vnode->status.data_version, - vnode->status.mode); + status->type, + status->nlink, + (unsigned long long) status->size, + status->data_version, + status->mode); + + write_seqlock(&vnode->cb_lock); - read_seqlock_excl(&vnode->cb_lock); + vnode->cb_v_check = op->cb_v_break; + vnode->status = *status; - afs_update_inode_from_status(vnode, &vnode->status, NULL, - AFS_VNODE_NOT_YET_SET); + t = status->mtime_client; + inode_set_ctime_to_ts(inode, t); + inode_set_mtime_to_ts(inode, t); + inode_set_atime_to_ts(inode, t); + inode->i_flags |= S_NOATIME; + inode->i_uid = make_kuid(&init_user_ns, status->owner); + inode->i_gid = make_kgid(&init_user_ns, status->group); + set_nlink(&vnode->netfs.inode, status->nlink); - switch (vnode->status.type) { + switch (status->type) { case AFS_FTYPE_FILE: - inode->i_mode = S_IFREG | vnode->status.mode; + inode->i_mode = S_IFREG | (status->mode & S_IALLUGO); inode->i_op = &afs_file_inode_operations; inode->i_fop = &afs_file_operations; - inode->i_mapping->a_ops = &afs_fs_aops; + inode->i_mapping->a_ops = &afs_file_aops; + mapping_set_large_folios(inode->i_mapping); break; case AFS_FTYPE_DIR: - inode->i_mode = S_IFDIR | vnode->status.mode; + inode->i_mode = S_IFDIR | (status->mode & S_IALLUGO); inode->i_op = &afs_dir_inode_operations; inode->i_fop = &afs_dir_file_operations; inode->i_mapping->a_ops = &afs_dir_aops; + __set_bit(NETFS_ICTX_SINGLE_NO_UPLOAD, &vnode->netfs.flags); + /* Assume locally cached directory data will be valid. */ + __set_bit(AFS_VNODE_DIR_VALID, &vnode->flags); break; case AFS_FTYPE_SYMLINK: /* Symlinks with a mode of 0644 are actually mountpoints. */ - if ((vnode->status.mode & 0777) == 0644) { + if ((status->mode & 0777) == 0644) { inode->i_flags |= S_AUTOMOUNT; set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags); @@ -71,75 +210,291 @@ static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key) inode->i_mode = S_IFDIR | 0555; inode->i_op = &afs_mntpt_inode_operations; inode->i_fop = &afs_mntpt_file_operations; - inode->i_mapping->a_ops = &afs_fs_aops; } else { - inode->i_mode = S_IFLNK | vnode->status.mode; + inode->i_mode = S_IFLNK | status->mode; inode->i_op = &afs_symlink_inode_operations; - inode->i_mapping->a_ops = &afs_fs_aops; } + inode->i_mapping->a_ops = &afs_dir_aops; inode_nohighmem(inode); + mapping_set_release_always(inode->i_mapping); break; default: - printk("kAFS: AFS vnode with undefined type\n"); - read_sequnlock_excl(&vnode->cb_lock); - return afs_protocol_error(NULL, -EBADMSG, afs_eproto_file_type); + dump_vnode(vnode, op->file[0].vnode != vnode ? op->file[0].vnode : NULL); + write_sequnlock(&vnode->cb_lock); + return afs_protocol_error(NULL, afs_eproto_file_type); } - inode->i_blocks = 0; - vnode->invalid_before = vnode->status.data_version; + afs_set_i_size(vnode, status->size); + afs_set_netfs_context(vnode); - read_sequnlock_excl(&vnode->cb_lock); + vnode->invalid_before = status->data_version; + trace_afs_set_dv(vnode, status->data_version); + inode_set_iversion_raw(&vnode->netfs.inode, status->data_version); + + if (!vp->scb.have_cb) { + /* it's a symlink we just created (the fileserver + * didn't give us a callback) */ + afs_clear_cb_promise(vnode, afs_cb_promise_set_new_symlink); + } else { + vnode->cb_server = op->server; + afs_set_cb_promise(vnode, vp->scb.callback.expires_at, + afs_cb_promise_set_new_inode); + } + + write_sequnlock(&vnode->cb_lock); return 0; } /* - * Fetch file status from the volume. + * Update the core inode struct from a returned status record. */ -int afs_fetch_status(struct afs_vnode *vnode, struct key *key, bool new_inode) +static void afs_apply_status(struct afs_operation *op, + struct afs_vnode_param *vp) { - struct afs_fs_cursor fc; + struct afs_file_status *status = &vp->scb.status; + struct afs_vnode *vnode = vp->vnode; + struct inode *inode = &vnode->netfs.inode; + struct timespec64 t; + umode_t mode; + bool unexpected_jump = false; + bool data_changed = false; + bool change_size = vp->set_size; + + _enter("{%llx:%llu.%u} %s", + vp->fid.vid, vp->fid.vnode, vp->fid.unique, + op->type ? op->type->name : "???"); + + BUG_ON(test_bit(AFS_VNODE_UNSET, &vnode->flags)); + + if (status->type != vnode->status.type) { + pr_warn("Vnode %llx:%llx:%x changed type %u to %u\n", + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + status->type, vnode->status.type); + afs_protocol_error(NULL, afs_eproto_bad_status); + return; + } + + if (status->nlink != vnode->status.nlink) + set_nlink(inode, status->nlink); + + if (status->owner != vnode->status.owner) + inode->i_uid = make_kuid(&init_user_ns, status->owner); + + if (status->group != vnode->status.group) + inode->i_gid = make_kgid(&init_user_ns, status->group); + + if (status->mode != vnode->status.mode) { + mode = inode->i_mode; + mode &= ~S_IALLUGO; + mode |= status->mode & S_IALLUGO; + WRITE_ONCE(inode->i_mode, mode); + } + + t = status->mtime_client; + inode_set_mtime_to_ts(inode, t); + if (vp->update_ctime) + inode_set_ctime_to_ts(inode, op->ctime); + + if (vnode->status.data_version != status->data_version) { + trace_afs_set_dv(vnode, status->data_version); + data_changed = true; + } + + vnode->status = *status; + + if (vp->dv_before + vp->dv_delta != status->data_version) { + trace_afs_dv_mismatch(vnode, vp->dv_before, vp->dv_delta, + status->data_version); + + if (vnode->cb_ro_snapshot == atomic_read(&vnode->volume->cb_ro_snapshot) && + atomic64_read(&vnode->cb_expires_at) != AFS_NO_CB_PROMISE) + pr_warn("kAFS: vnode modified {%llx:%llu} %llx->%llx %s (op=%x)\n", + vnode->fid.vid, vnode->fid.vnode, + (unsigned long long)vp->dv_before + vp->dv_delta, + (unsigned long long)status->data_version, + op->type ? op->type->name : "???", + op->debug_id); + + vnode->invalid_before = status->data_version; + if (vnode->status.type == AFS_FTYPE_DIR) + afs_invalidate_dir(vnode, afs_dir_invalid_dv_mismatch); + else + set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); + change_size = true; + data_changed = true; + unexpected_jump = true; + } else if (vnode->status.type == AFS_FTYPE_DIR) { + /* Expected directory change is handled elsewhere so + * that we can locally edit the directory and save on a + * download. + */ + if (test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + data_changed = false; + change_size = true; + } + + if (data_changed) { + inode_set_iversion_raw(inode, status->data_version); + + /* Only update the size if the data version jumped. If the + * file is being modified locally, then we might have our own + * idea of what the size should be that's not the same as + * what's on the server. + */ + vnode->netfs.remote_i_size = status->size; + if (change_size || status->size > i_size_read(inode)) { + afs_set_i_size(vnode, status->size); + if (unexpected_jump) + vnode->netfs.zero_point = status->size; + inode_set_ctime_to_ts(inode, t); + inode_set_atime_to_ts(inode, t); + } + if (op->ops == &afs_fetch_data_operation) + op->fetch.subreq->rreq->i_size = status->size; + } +} + +/* + * Apply a callback to a vnode. + */ +static void afs_apply_callback(struct afs_operation *op, + struct afs_vnode_param *vp) +{ + struct afs_callback *cb = &vp->scb.callback; + struct afs_vnode *vnode = vp->vnode; + + if (!afs_cb_is_broken(vp->cb_break_before, vnode)) { + if (op->volume->type == AFSVL_RWVOL) + vnode->cb_server = op->server; + afs_set_cb_promise(vnode, cb->expires_at, afs_cb_promise_set_apply_cb); + } +} + +/* + * Apply the received status and callback to an inode all in the same critical + * section to avoid races with afs_validate(). + */ +void afs_vnode_commit_status(struct afs_operation *op, struct afs_vnode_param *vp) +{ + struct afs_vnode *vnode = vp->vnode; + + _enter(""); + + write_seqlock(&vnode->cb_lock); + + if (vp->scb.have_error) { + /* A YFS server will return this from RemoveFile2 and AFS and + * YFS will return this from InlineBulkStatus. + */ + if (vp->scb.status.abort_code == VNOVNODE) { + set_bit(AFS_VNODE_DELETED, &vnode->flags); + clear_nlink(&vnode->netfs.inode); + __afs_break_callback(vnode, afs_cb_break_for_deleted); + op->flags &= ~AFS_OPERATION_DIR_CONFLICT; + } + } else if (vp->scb.have_status) { + if (vp->speculative && + (test_bit(AFS_VNODE_MODIFYING, &vnode->flags) || + vp->dv_before != vnode->status.data_version)) + /* Ignore the result of a speculative bulk status fetch + * if it splits around a modification op, thereby + * appearing to regress the data version. + */ + goto out; + afs_apply_status(op, vp); + if (vp->scb.have_cb) + afs_apply_callback(op, vp); + } else if (vp->op_unlinked && !(op->flags & AFS_OPERATION_DIR_CONFLICT)) { + drop_nlink(&vnode->netfs.inode); + if (vnode->netfs.inode.i_nlink == 0) { + set_bit(AFS_VNODE_DELETED, &vnode->flags); + __afs_break_callback(vnode, afs_cb_break_for_deleted); + } + } + +out: + write_sequnlock(&vnode->cb_lock); + + if (vp->scb.have_status) + afs_cache_permit(vnode, op->key, vp->cb_break_before, &vp->scb); +} + +static void afs_fetch_status_success(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[op->fetch_status.which]; + struct afs_vnode *vnode = vp->vnode; int ret; + if (inode_state_read_once(&vnode->netfs.inode) & I_NEW) { + ret = afs_inode_init_from_status(op, vp, vnode); + afs_op_set_error(op, ret); + if (ret == 0) + afs_cache_permit(vnode, op->key, vp->cb_break_before, &vp->scb); + } else { + afs_vnode_commit_status(op, vp); + } +} + +const struct afs_operation_ops afs_fetch_status_operation = { + .issue_afs_rpc = afs_fs_fetch_status, + .issue_yfs_rpc = yfs_fs_fetch_status, + .success = afs_fetch_status_success, + .aborted = afs_check_for_remote_deletion, +}; + +/* + * Fetch file status from the volume. + */ +int afs_fetch_status(struct afs_vnode *vnode, struct key *key, bool is_new, + afs_access_t *_caller_access) +{ + struct afs_operation *op; + _enter("%s,{%llx:%llu.%u,S=%lx}", vnode->volume->name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, vnode->flags); - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_fetch_file_status(&fc, NULL, new_inode); - } + op = afs_alloc_operation(key, vnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } + afs_op_set_vnode(op, 0, vnode); - _leave(" = %d", ret); - return ret; + op->nr_files = 1; + op->ops = &afs_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + + if (_caller_access) + *_caller_access = op->file[0].scb.status.caller_access; + return afs_put_operation(op); } /* - * iget5() comparator + * ilookup() comparator */ -int afs_iget5_test(struct inode *inode, void *opaque) +int afs_ilookup5_test_by_fid(struct inode *inode, void *opaque) { - struct afs_iget_data *data = opaque; struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_fid *fid = opaque; - return memcmp(&vnode->fid, &data->fid, sizeof(data->fid)) == 0; + return (fid->vnode == vnode->fid.vnode && + fid->vnode_hi == vnode->fid.vnode_hi && + fid->unique == vnode->fid.unique); } /* - * iget5() comparator for inode created by autocell operations - * - * These pseudo inodes don't match anything. + * iget5() comparator */ -static int afs_iget5_pseudo_dir_test(struct inode *inode, void *opaque) +static int afs_iget5_test(struct inode *inode, void *opaque) { - return 0; + struct afs_vnode_param *vp = opaque; + //struct afs_vnode *vnode = AFS_FS_I(inode); + + return afs_ilookup5_test_by_fid(inode, &vp->fid); } /* @@ -147,189 +502,89 @@ static int afs_iget5_pseudo_dir_test(struct inode *inode, void *opaque) */ static int afs_iget5_set(struct inode *inode, void *opaque) { - struct afs_iget_data *data = opaque; + struct afs_vnode_param *vp = opaque; + struct afs_super_info *as = AFS_FS_S(inode->i_sb); struct afs_vnode *vnode = AFS_FS_I(inode); - vnode->fid = data->fid; - vnode->volume = data->volume; + vnode->volume = as->volume; + vnode->fid = vp->fid; /* YFS supports 96-bit vnode IDs, but Linux only supports * 64-bit inode numbers. */ - inode->i_ino = data->fid.vnode; - inode->i_generation = data->fid.unique; + inode->i_ino = vnode->fid.vnode; + inode->i_generation = vnode->fid.unique; return 0; } /* - * Create an inode for a dynamic root directory or an autocell dynamic - * automount dir. - */ -struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root) -{ - struct afs_iget_data data; - struct afs_super_info *as; - struct afs_vnode *vnode; - struct inode *inode; - static atomic_t afs_autocell_ino; - - _enter(""); - - as = sb->s_fs_info; - if (as->volume) { - data.volume = as->volume; - data.fid.vid = as->volume->vid; - } - if (root) { - data.fid.vnode = 1; - data.fid.unique = 1; - } else { - data.fid.vnode = atomic_inc_return(&afs_autocell_ino); - data.fid.unique = 0; - } - - inode = iget5_locked(sb, data.fid.vnode, - afs_iget5_pseudo_dir_test, afs_iget5_set, - &data); - if (!inode) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - _debug("GOT INODE %p { ino=%lu, vl=%llx, vn=%llx, u=%x }", - inode, inode->i_ino, data.fid.vid, data.fid.vnode, - data.fid.unique); - - vnode = AFS_FS_I(inode); - - /* there shouldn't be an existing inode */ - BUG_ON(!(inode->i_state & I_NEW)); - - inode->i_size = 0; - inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; - if (root) { - inode->i_op = &afs_dynroot_inode_operations; - inode->i_fop = &afs_dynroot_file_operations; - } else { - inode->i_op = &afs_autocell_inode_operations; - } - set_nlink(inode, 2); - inode->i_uid = GLOBAL_ROOT_UID; - inode->i_gid = GLOBAL_ROOT_GID; - inode->i_ctime.tv_sec = get_seconds(); - inode->i_ctime.tv_nsec = 0; - inode->i_atime = inode->i_mtime = inode->i_ctime; - inode->i_blocks = 0; - inode_set_iversion_raw(inode, 0); - inode->i_generation = 0; - - set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags); - if (!root) { - set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags); - inode->i_flags |= S_AUTOMOUNT; - } - - inode->i_flags |= S_NOATIME; - unlock_new_inode(inode); - _leave(" = %p", inode); - return inode; -} - -/* * Get a cache cookie for an inode. */ static void afs_get_inode_cache(struct afs_vnode *vnode) { #ifdef CONFIG_AFS_FSCACHE struct { - u32 vnode_id; - u32 unique; - u32 vnode_id_ext[2]; /* Allow for a 96-bit key */ + __be32 vnode_id; + __be32 unique; + __be32 vnode_id_ext[2]; /* Allow for a 96-bit key */ } __packed key; struct afs_vnode_cache_aux aux; - if (vnode->status.type == AFS_FTYPE_DIR) { - vnode->cache = NULL; + if (vnode->status.type != AFS_FTYPE_FILE && + vnode->status.type != AFS_FTYPE_DIR && + vnode->status.type != AFS_FTYPE_SYMLINK) { + vnode->netfs.cache = NULL; return; } - key.vnode_id = vnode->fid.vnode; - key.unique = vnode->fid.unique; - key.vnode_id_ext[0] = vnode->fid.vnode >> 32; - key.vnode_id_ext[1] = vnode->fid.vnode_hi; - aux.data_version = vnode->status.data_version; - - vnode->cache = fscache_acquire_cookie(vnode->volume->cache, - &afs_vnode_cache_index_def, - &key, sizeof(key), - &aux, sizeof(aux), - vnode, vnode->status.size, true); + key.vnode_id = htonl(vnode->fid.vnode); + key.unique = htonl(vnode->fid.unique); + key.vnode_id_ext[0] = htonl(vnode->fid.vnode >> 32); + key.vnode_id_ext[1] = htonl(vnode->fid.vnode_hi); + afs_set_cache_aux(vnode, &aux); + + afs_vnode_set_cache(vnode, + fscache_acquire_cookie( + vnode->volume->cache, + vnode->status.type == AFS_FTYPE_FILE ? + 0 : FSCACHE_ADV_SINGLE_CHUNK, + &key, sizeof(key), + &aux, sizeof(aux), + i_size_read(&vnode->netfs.inode))); #endif } /* * inode retrieval */ -struct inode *afs_iget(struct super_block *sb, struct key *key, - struct afs_fid *fid, struct afs_file_status *status, - struct afs_callback *cb, struct afs_cb_interest *cbi) +struct inode *afs_iget(struct afs_operation *op, struct afs_vnode_param *vp) { - struct afs_iget_data data = { .fid = *fid }; - struct afs_super_info *as; + struct afs_vnode_param *dvp = &op->file[0]; + struct super_block *sb = dvp->vnode->netfs.inode.i_sb; struct afs_vnode *vnode; struct inode *inode; int ret; - _enter(",{%llx:%llu.%u},,", fid->vid, fid->vnode, fid->unique); - - as = sb->s_fs_info; - data.volume = as->volume; + _enter(",{%llx:%llu.%u},,", vp->fid.vid, vp->fid.vnode, vp->fid.unique); - inode = iget5_locked(sb, fid->vnode, afs_iget5_test, afs_iget5_set, - &data); + inode = iget5_locked(sb, vp->fid.vnode, afs_iget5_test, afs_iget5_set, vp); if (!inode) { _leave(" = -ENOMEM"); return ERR_PTR(-ENOMEM); } - _debug("GOT INODE %p { vl=%llx vn=%llx, u=%x }", - inode, fid->vid, fid->vnode, fid->unique); - vnode = AFS_FS_I(inode); + _debug("GOT INODE %p { vl=%llx vn=%llx, u=%x }", + inode, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); + /* deal with an existing inode */ - if (!(inode->i_state & I_NEW)) { + if (!(inode_state_read_once(inode) & I_NEW)) { _leave(" = %p", inode); return inode; } - if (!status) { - /* it's a remotely extant inode */ - ret = afs_fetch_status(vnode, key, true); - if (ret < 0) - goto bad_inode; - } else { - /* it's an inode we just created */ - memcpy(&vnode->status, status, sizeof(vnode->status)); - - if (!cb) { - /* it's a symlink we just created (the fileserver - * didn't give us a callback) */ - vnode->cb_version = 0; - vnode->cb_type = 0; - vnode->cb_expires_at = ktime_get(); - } else { - vnode->cb_version = cb->version; - vnode->cb_type = cb->type; - vnode->cb_expires_at = cb->expires_at; - vnode->cb_interest = afs_get_cb_interest(cbi); - set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - } - - vnode->cb_expires_at += ktime_get_real_seconds(); - } - - ret = afs_inode_init_from_status(vnode, key); + ret = afs_inode_init_from_status(op, vp, vnode); if (ret < 0) goto bad_inode; @@ -337,9 +592,8 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, /* success */ clear_bit(AFS_VNODE_UNSET, &vnode->flags); - inode->i_flags |= S_NOATIME; unlock_new_inode(inode); - _leave(" = %p [CB { v=%u t=%u }]", inode, vnode->cb_version, vnode->cb_type); + _leave(" = %p", inode); return inode; /* failure */ @@ -349,141 +603,115 @@ bad_inode: return ERR_PTR(ret); } -/* - * mark the data attached to an inode as obsolete due to a write on the server - * - might also want to ditch all the outstanding writes and dirty pages - */ -void afs_zap_data(struct afs_vnode *vnode) +static int afs_iget5_set_root(struct inode *inode, void *opaque) { - _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); - -#ifdef CONFIG_AFS_FSCACHE - fscache_invalidate(vnode->cache); -#endif + struct afs_super_info *as = AFS_FS_S(inode->i_sb); + struct afs_vnode *vnode = AFS_FS_I(inode); - /* nuke all the non-dirty pages that aren't locked, mapped or being - * written back in a regular file and completely discard the pages in a - * directory or symlink */ - if (S_ISREG(vnode->vfs_inode.i_mode)) - invalidate_remote_inode(&vnode->vfs_inode); - else - invalidate_inode_pages2(vnode->vfs_inode.i_mapping); + vnode->volume = as->volume; + vnode->fid.vid = as->volume->vid; + vnode->fid.vnode = 1; + vnode->fid.unique = 1; + inode->i_ino = 1; + inode->i_generation = 1; + return 0; } /* - * validate a vnode/inode - * - there are several things we need to check - * - parent dir data changes (rm, rmdir, rename, mkdir, create, link, - * symlink) - * - parent dir metadata changed (security changes) - * - dentry data changed (write, truncate) - * - dentry metadata changed (security changes) + * Set up the root inode for a volume. This is always vnode 1, unique 1 within + * the volume. */ -int afs_validate(struct afs_vnode *vnode, struct key *key) +struct inode *afs_root_iget(struct super_block *sb, struct key *key) { - time64_t now = ktime_get_real_seconds(); - bool valid; + struct afs_super_info *as = AFS_FS_S(sb); + struct afs_operation *op; + struct afs_vnode *vnode; + struct inode *inode; int ret; - _enter("{v={%llx:%llu} fl=%lx},%x", - vnode->fid.vid, vnode->fid.vnode, vnode->flags, - key_serial(key)); + _enter(",{%llx},,", as->volume->vid); - /* Quickly check the callback state. Ideally, we'd use read_seqbegin - * here, but we have no way to pass the net namespace to the RCU - * cleanup for the server record. - */ - read_seqlock_excl(&vnode->cb_lock); - - if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { - if (vnode->cb_s_break != vnode->cb_interest->server->cb_s_break || - vnode->cb_v_break != vnode->volume->cb_v_break) { - vnode->cb_s_break = vnode->cb_interest->server->cb_s_break; - vnode->cb_v_break = vnode->volume->cb_v_break; - valid = false; - } else if (vnode->status.type == AFS_FTYPE_DIR && - (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags) || - vnode->cb_expires_at - 10 <= now)) { - valid = false; - } else if (test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) || - vnode->cb_expires_at - 10 <= now) { - valid = false; - } else { - valid = true; - } - } else if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { - valid = true; - } else { - vnode->cb_v_break = vnode->volume->cb_v_break; - valid = false; + inode = iget5_locked(sb, 1, NULL, afs_iget5_set_root, NULL); + if (!inode) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); } - read_sequnlock_excl(&vnode->cb_lock); + _debug("GOT ROOT INODE %p { vl=%llx }", inode, as->volume->vid); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - clear_nlink(&vnode->vfs_inode); + BUG_ON(!(inode_state_read_once(inode) & I_NEW)); - if (valid) - goto valid; - - down_write(&vnode->validate_lock); + vnode = AFS_FS_I(inode); + vnode->cb_v_check = atomic_read(&as->volume->cb_v_break); + afs_set_netfs_context(vnode); - /* if the promise has expired, we need to check the server again to get - * a new promise - note that if the (parent) directory's metadata was - * changed then the security may be different and we may no longer have - * access */ - if (!test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { - _debug("not promised"); - ret = afs_fetch_status(vnode, key, false); - if (ret < 0) { - if (ret == -ENOENT) { - set_bit(AFS_VNODE_DELETED, &vnode->flags); - ret = -ESTALE; - } - goto error_unlock; - } - _debug("new promise [fl=%lx]", vnode->flags); + op = afs_alloc_operation(key, as->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); + goto error; } - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { - _debug("file already deleted"); - ret = -ESTALE; - goto error_unlock; - } + afs_op_set_vnode(op, 0, vnode); - /* if the vnode's data version number changed then its contents are - * different */ - if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) - afs_zap_data(vnode); - up_write(&vnode->validate_lock); -valid: - _leave(" = 0"); - return 0; + op->nr_files = 1; + op->ops = &afs_fetch_status_operation; + ret = afs_do_sync_operation(op); + if (ret < 0) + goto error; -error_unlock: - up_write(&vnode->validate_lock); - _leave(" = %d", ret); - return ret; + afs_get_inode_cache(vnode); + + clear_bit(AFS_VNODE_UNSET, &vnode->flags); + unlock_new_inode(inode); + _leave(" = %p", inode); + return inode; + +error: + iget_failed(inode); + _leave(" = %d [bad]", ret); + return ERR_PTR(ret); } /* * read the attributes of an inode */ -int afs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int afs_getattr(struct mnt_idmap *idmap, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct afs_vnode *vnode = AFS_FS_I(inode); - int seq = 0; + struct key *key; + int ret, seq; _enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation); + if (vnode->volume && + !(query_flags & AT_STATX_DONT_SYNC) && + atomic64_read(&vnode->cb_expires_at) == AFS_NO_CB_PROMISE) { + key = afs_request_key(vnode->volume->cell); + if (IS_ERR(key)) + return PTR_ERR(key); + ret = afs_validate(vnode, key); + key_put(key); + if (ret < 0) + return ret; + } + do { - read_seqbegin_or_lock(&vnode->cb_lock, &seq); - generic_fillattr(inode, stat); - } while (need_seqretry(&vnode->cb_lock, seq)); + seq = read_seqbegin(&vnode->cb_lock); + generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat); + if (test_bit(AFS_VNODE_SILLY_DELETED, &vnode->flags) && + stat->nlink > 0) + stat->nlink -= 1; + + /* Lie about the size of directories. We maintain a locally + * edited copy and may make different allocation decisions on + * it, but we need to give userspace the server's size. + */ + if (S_ISDIR(inode->i_mode)) + stat->size = vnode->netfs.remote_i_size; + } while (read_seqretry(&vnode->cb_lock, seq)); - done_seqretry(&vnode->cb_lock, seq); return 0; } @@ -495,9 +723,9 @@ int afs_drop_inode(struct inode *inode) _enter(""); if (test_bit(AFS_VNODE_PSEUDODIR, &AFS_FS_I(inode)->flags)) - return generic_delete_inode(inode); + return inode_just_drop(inode); else - return generic_drop_inode(inode); + return inode_generic_drop(inode); } /* @@ -505,9 +733,9 @@ int afs_drop_inode(struct inode *inode) */ void afs_evict_inode(struct inode *inode) { - struct afs_vnode *vnode; - - vnode = AFS_FS_I(inode); + struct afs_vnode_cache_aux aux; + struct afs_super_info *sbi = AFS_FS_S(inode->i_sb); + struct afs_vnode *vnode = AFS_FS_I(inode); _enter("{%llx:%llu.%d}", vnode->fid.vid, @@ -518,13 +746,26 @@ void afs_evict_inode(struct inode *inode) ASSERTCMP(inode->i_ino, ==, vnode->fid.vnode); + if ((S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) && + (inode_state_read_once(inode) & I_DIRTY) && + !sbi->dyn_root) { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .for_sync = true, + .range_end = LLONG_MAX, + }; + + afs_single_writepages(inode->i_mapping, &wbc); + } + + netfs_wait_for_outstanding_io(inode); truncate_inode_pages_final(&inode->i_data); - clear_inode(inode); + netfs_free_folioq_buffer(vnode->directory); - if (vnode->cb_interest) { - afs_put_cb_interest(afs_i2net(inode), vnode->cb_interest); - vnode->cb_interest = NULL; - } + afs_set_cache_aux(vnode, &aux); + netfs_clear_inode_writeback(inode, &aux); + clear_inode(inode); while (!list_empty(&vnode->wb_keys)) { struct afs_wb_key *wbk = list_entry(vnode->wb_keys.next, @@ -533,73 +774,156 @@ void afs_evict_inode(struct inode *inode) afs_put_wb_key(wbk); } -#ifdef CONFIG_AFS_FSCACHE - { - struct afs_vnode_cache_aux aux; - - aux.data_version = vnode->status.data_version; - fscache_relinquish_cookie(vnode->cache, &aux, - test_bit(AFS_VNODE_DELETED, &vnode->flags)); - vnode->cache = NULL; - } -#endif + fscache_relinquish_cookie(afs_vnode_cache(vnode), + test_bit(AFS_VNODE_DELETED, &vnode->flags)); + afs_prune_wb_keys(vnode); afs_put_permits(rcu_access_pointer(vnode->permit_cache)); + key_put(vnode->silly_key); + vnode->silly_key = NULL; key_put(vnode->lock_key); vnode->lock_key = NULL; _leave(""); } +static void afs_setattr_success(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct inode *inode = &vp->vnode->netfs.inode; + loff_t old_i_size = i_size_read(inode); + + op->setattr.old_i_size = old_i_size; + afs_vnode_commit_status(op, vp); + /* inode->i_size has now been changed. */ + + if (op->setattr.attr->ia_valid & ATTR_SIZE) { + loff_t size = op->setattr.attr->ia_size; + if (size > old_i_size) + pagecache_isize_extended(inode, old_i_size, size); + } +} + +static void afs_setattr_edit_file(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_vnode *vnode = vp->vnode; + struct inode *inode = &vnode->netfs.inode; + + if (op->setattr.attr->ia_valid & ATTR_SIZE) { + loff_t size = op->setattr.attr->ia_size; + loff_t old = op->setattr.old_i_size; + + /* Note: inode->i_size was updated by afs_apply_status() inside + * the I/O and callback locks. + */ + + if (size != old) { + truncate_pagecache(inode, size); + netfs_resize_file(&vnode->netfs, size, true); + fscache_resize_cookie(afs_vnode_cache(vnode), size); + } + } +} + +static const struct afs_operation_ops afs_setattr_operation = { + .issue_afs_rpc = afs_fs_setattr, + .issue_yfs_rpc = yfs_fs_setattr, + .success = afs_setattr_success, + .edit_dir = afs_setattr_edit_file, +}; + /* * set the attributes of an inode */ -int afs_setattr(struct dentry *dentry, struct iattr *attr) +int afs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, + struct iattr *attr) { - struct afs_fs_cursor fc; + const unsigned int supported = + ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID | + ATTR_MTIME | ATTR_MTIME_SET | ATTR_TIMES_SET | ATTR_TOUCH; + struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - struct key *key; + struct inode *inode = &vnode->netfs.inode; + loff_t i_size; int ret; _enter("{%llx:%llu},{n=%pd},%x", vnode->fid.vid, vnode->fid.vnode, dentry, attr->ia_valid); - if (!(attr->ia_valid & (ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID | - ATTR_MTIME))) { + if (!(attr->ia_valid & supported)) { _leave(" = 0 [unsupported]"); return 0; } - /* flush any dirty data outstanding on a regular file */ - if (S_ISREG(vnode->vfs_inode.i_mode)) - filemap_write_and_wait(vnode->vfs_inode.i_mapping); + i_size = i_size_read(inode); + if (attr->ia_valid & ATTR_SIZE) { + if (!S_ISREG(inode->i_mode)) + return -EISDIR; - if (attr->ia_valid & ATTR_FILE) { - key = afs_file_key(attr->ia_file); - } else { - key = afs_request_key(vnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } + ret = inode_newsize_ok(inode, attr->ia_size); + if (ret) + return ret; + + if (attr->ia_size == i_size) + attr->ia_valid &= ~ATTR_SIZE; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_setattr(&fc, attr); + fscache_use_cookie(afs_vnode_cache(vnode), true); + + /* Prevent any new writebacks from starting whilst we do this. */ + down_write(&vnode->validate_lock); + + if ((attr->ia_valid & ATTR_SIZE) && S_ISREG(inode->i_mode)) { + loff_t size = attr->ia_size; + + /* Wait for any outstanding writes to the server to complete */ + loff_t from = min(size, i_size); + loff_t to = max(size, i_size); + ret = filemap_fdatawait_range(inode->i_mapping, from, to); + if (ret < 0) + goto out_unlock; + + /* Don't talk to the server if we're just shortening in-memory + * writes that haven't gone to the server yet. + */ + if (!(attr->ia_valid & (supported & ~ATTR_SIZE & ~ATTR_MTIME)) && + attr->ia_size < i_size && + attr->ia_size > vnode->netfs.remote_i_size) { + truncate_setsize(inode, attr->ia_size); + netfs_resize_file(&vnode->netfs, size, false); + fscache_resize_cookie(afs_vnode_cache(vnode), + attr->ia_size); + ret = 0; + goto out_unlock; } + } - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); + op = afs_alloc_operation(((attr->ia_valid & ATTR_FILE) ? + afs_file_key(attr->ia_file) : NULL), + vnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); + goto out_unlock; } - if (!(attr->ia_valid & ATTR_FILE)) - key_put(key); + afs_op_set_vnode(op, 0, vnode); + op->setattr.attr = attr; -error: + if (attr->ia_valid & ATTR_SIZE) { + op->file[0].dv_delta = 1; + op->file[0].set_size = true; + } + op->ctime = attr->ia_ctime; + op->file[0].update_ctime = 1; + op->file[0].modification = true; + + op->ops = &afs_setattr_operation; + ret = afs_do_sync_operation(op); + +out_unlock: + up_write(&vnode->validate_lock); + fscache_unuse_cookie(afs_vnode_cache(vnode), NULL, NULL); _leave(" = %d", ret); return ret; } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 8871b9e8645f..009064b8d661 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1,18 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* internal AFS stuff * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/compiler.h> #include <linux/kernel.h> #include <linux/ktime.h> #include <linux/fs.h> +#include <linux/filelock.h> #include <linux/pagemap.h> #include <linux/rxrpc.h> #include <linux/key.h> @@ -23,6 +20,7 @@ #include <linux/uuid.h> #include <linux/mm_types.h> #include <linux/dns_resolver.h> +#include <crypto/krb5.h> #include <net/net_namespace.h> #include <net/netns/generic.h> #include <net/sock.h> @@ -35,27 +33,36 @@ struct pagevec; struct afs_call; +struct afs_vnode; +struct afs_server_probe; + +/* + * Partial file-locking emulation mode. (The problem being that AFS3 only + * allows whole-file locks and no upgrading/downgrading). + */ +enum afs_flock_mode { + afs_flock_mode_unset, + afs_flock_mode_local, /* Local locking only */ + afs_flock_mode_openafs, /* Don't get server lock for a partial lock */ + afs_flock_mode_strict, /* Always get a server lock for a partial lock */ + afs_flock_mode_write, /* Get an exclusive server lock for a partial lock */ +}; -struct afs_mount_params { - bool rwpath; /* T if the parent should be considered R/W */ +struct afs_fs_context { bool force; /* T to force cell type */ bool autocell; /* T if set auto mount operation */ bool dyn_root; /* T if dynamic root */ + bool no_cell; /* T if the source is "none" (for dynroot) */ + enum afs_flock_mode flock_mode; /* Partial file-locking emulation mode */ afs_voltype_t type; /* type of volume requested */ - int volnamesz; /* size of volume name */ + unsigned int volnamesz; /* size of volume name */ const char *volname; /* name of volume to mount */ - struct net *net_ns; /* Network namespace in effect */ struct afs_net *net; /* the AFS net namespace stuff */ struct afs_cell *cell; /* cell in which to find volume */ struct afs_volume *volume; /* volume record */ struct key *key; /* key to use for secure mounting */ }; -struct afs_iget_data { - struct afs_fid fid; - struct afs_volume *volume; /* volume on which resides */ -}; - enum afs_call_state { AFS_CALL_CL_REQUESTING, /* Client: Request is being sent */ AFS_CALL_CL_AWAIT_REPLY, /* Client: Awaiting reply */ @@ -68,22 +75,51 @@ enum afs_call_state { }; /* + * Address preferences. + */ +struct afs_addr_preference { + union { + struct in_addr ipv4_addr; /* AF_INET address to compare against */ + struct in6_addr ipv6_addr; /* AF_INET6 address to compare against */ + }; + sa_family_t family; /* Which address to use */ + u16 prio; /* Priority */ + u8 subnet_mask; /* How many bits to compare */ +}; + +struct afs_addr_preference_list { + struct rcu_head rcu; + u16 version; /* Incremented when prefs list changes */ + u8 ipv6_off; /* Offset of IPv6 addresses */ + u8 nr; /* Number of addresses in total */ + u8 max_prefs; /* Number of prefs allocated */ + struct afs_addr_preference prefs[] __counted_by(max_prefs); +}; + +struct afs_address { + struct rxrpc_peer *peer; + short last_error; /* Last error from this address */ + u16 prio; /* Address priority */ +}; + +/* * List of server addresses. */ struct afs_addr_list { - struct rcu_head rcu; /* Must be first */ + struct rcu_head rcu; refcount_t usage; u32 version; /* Version */ + unsigned int debug_id; + unsigned int addr_pref_version; /* Version of address preference list */ unsigned char max_addrs; unsigned char nr_addrs; unsigned char preferred; /* Preferred address */ unsigned char nr_ipv4; /* Number of IPv4 addresses */ enum dns_record_source source:8; enum dns_lookup_status status:8; - unsigned long probed; /* Mask of servers that have been probed */ - unsigned long failed; /* Mask of addrs that failed locally/ICMP */ + unsigned long probe_failed; /* Mask of addrs that failed locally/ICMP */ unsigned long responded; /* Mask of addrs that responded */ - struct sockaddr_rxrpc addrs[]; + struct afs_address addrs[] __counted_by(max_addrs); #define AFS_MAX_ADDRESSES ((unsigned int)(sizeof(unsigned long) * 8)) }; @@ -92,52 +128,59 @@ struct afs_addr_list { */ struct afs_call { const struct afs_call_type *type; /* type of call */ - struct afs_addr_list *alist; /* Address is alist[addr_ix] */ wait_queue_head_t waitq; /* processes awaiting completion */ struct work_struct async_work; /* async I/O processor */ struct work_struct work; /* actual work processor */ + struct work_struct free_work; /* Deferred free processor */ struct rxrpc_call *rxcall; /* RxRPC call handle */ + struct rxrpc_peer *peer; /* Remote endpoint */ struct key *key; /* security for this call */ struct afs_net *net; /* The network namespace */ - struct afs_server *cm_server; /* Server affected by incoming CM call */ - struct afs_cb_interest *cbi; /* Callback interest for server used */ + struct afs_server *server; /* The fileserver record if fs op (pins ref) */ + struct afs_vlserver *vlserver; /* The vlserver record if vl op */ void *request; /* request data (first part) */ - struct address_space *mapping; /* Pages being written from */ - struct iov_iter iter; /* Buffer iterator */ - struct iov_iter *_iter; /* Iterator currently in use */ - union { /* Convenience for ->iter */ + size_t iov_len; /* Size of *iter to be used */ + struct iov_iter def_iter; /* Default buffer/data iterator */ + struct iov_iter *write_iter; /* Iterator defining write to be made */ + struct iov_iter *iter; /* Iterator currently in use */ + union { /* Convenience for ->def_iter */ struct kvec kvec[1]; struct bio_vec bvec[1]; }; void *buffer; /* reply receive buffer */ - void *reply[4]; /* Where to put the reply */ - pgoff_t first; /* first page in mapping to deal with */ - pgoff_t last; /* last page in mapping to deal with */ - atomic_t usage; + union { + struct afs_endpoint_state *probe; + struct afs_addr_list *vl_probe; + struct afs_addr_list *ret_alist; + struct afs_vldb_entry *ret_vldb; + char *ret_str; + }; + struct afs_fid fid; /* Primary vnode ID (or all zeroes) */ + unsigned char probe_index; /* Address in ->probe_alist */ + struct afs_operation *op; + unsigned int server_index; + refcount_t ref; enum afs_call_state state; spinlock_t state_lock; int error; /* error code */ u32 abort_code; /* Remote abort ID or 0 */ - u32 epoch; + unsigned long long remaining; /* How much is left to receive */ + unsigned int max_lifespan; /* Maximum lifespan in secs to set if not 0 */ unsigned request_size; /* size of request data */ unsigned reply_max; /* maximum size of reply */ - unsigned first_offset; /* offset into mapping[first] */ - unsigned int cb_break; /* cb_break + cb_s_break before the call */ - union { - unsigned last_to; /* amount of mapping[last] */ - unsigned count2; /* count used in unmarshalling */ - }; + unsigned count2; /* count used in unmarshalling */ unsigned char unmarshall; /* unmarshalling phase */ - unsigned char addr_ix; /* Address in ->alist */ - bool incoming; /* T if incoming call */ - bool send_pages; /* T if data from mapping should be sent */ + bool drop_ref; /* T if need to drop ref for incoming call */ bool need_attention; /* T if RxRPC poked us */ bool async; /* T if asynchronous */ - bool ret_reply0; /* T if should return reply[0] on success */ bool upgrade; /* T to request service upgrade */ - bool want_reply_time; /* T if want reply_time */ + bool intr; /* T if interruptible */ + bool unmarshalling_error; /* T if an unmarshalling error occurred */ + bool responded; /* Got a response from the call (may be abort) */ + u8 security_ix; /* Security class */ u16 service_id; /* Actual service ID (after upgrade) */ unsigned int debug_id; /* Trace ID */ + u32 enctype; /* Security encoding type */ u32 operation_ID; /* operation ID for an incoming call */ u32 count; /* count for use in unmarshalling */ union { /* place to extract temporary data */ @@ -147,9 +190,7 @@ struct afs_call { } __attribute__((packed)); __be64 tmp64; }; - afs_dataversion_t expected_version; /* Updated version expected from store */ - afs_dataversion_t expected_version_2; /* 2nd updated version expected from store */ - ktime_t reply_time; /* Time of first reply packet */ + ktime_t issue_time; /* Time of issue of operation */ }; struct afs_call_type { @@ -164,11 +205,17 @@ struct afs_call_type { /* clean up a call */ void (*destructor)(struct afs_call *call); + /* Async receive processing function */ + void (*async_rx)(struct work_struct *work); + /* Work function */ void (*work)(struct work_struct *work); /* Call done function (gets called immediately on success or failure) */ void (*done)(struct afs_call *call); + + /* Handle a call being immediately cancelled. */ + void (*immediate_cancel)(struct afs_call *call); }; /* @@ -196,25 +243,6 @@ static inline struct key *afs_file_key(struct file *file) } /* - * Record of an outstanding read operation on a vnode. - */ -struct afs_read { - loff_t pos; /* Where to start reading */ - loff_t len; /* How much we're asking for */ - loff_t actual_len; /* How much we're actually getting */ - loff_t remain; /* Amount remaining */ - loff_t file_size; /* File size returned by server */ - afs_dataversion_t data_version; /* Version number returned by server */ - refcount_t usage; - unsigned int index; /* Which page we're reading into */ - unsigned int nr_pages; - unsigned int offset; /* offset into current page */ - void (*page_done)(struct afs_call *, struct afs_read *); - struct page **pages; - struct page *array[]; -}; - -/* * AFS superblock private data * - there's one superblock per volume */ @@ -222,6 +250,7 @@ struct afs_super_info { struct net *net_ns; /* Network namespace */ struct afs_cell *cell; /* The cell in which the volume resides */ struct afs_volume *volume; /* volume record */ + enum afs_flock_mode flock_mode:8; /* File locking emulation mode */ bool dyn_root; /* True if dynamic root */ }; @@ -255,17 +284,18 @@ struct afs_net { struct socket *socket; struct afs_call *spare_incoming_call; struct work_struct charge_preallocation_work; + struct work_struct rx_oob_work; struct mutex socket_mutex; atomic_t nr_outstanding_calls; atomic_t nr_superblocks; /* Cell database */ struct rb_root cells; + struct idr cells_dyn_ino; /* cell->dynroot_ino mapping */ struct afs_cell __rcu *ws_cell; - struct work_struct cells_manager; - struct timer_list cells_timer; atomic_t cells_outstanding; - seqlock_t cells_lock; + struct rw_semaphore cells_lock; + struct mutex cells_alias_lock; struct mutex proc_cells_lock; struct hlist_head proc_cells; @@ -274,17 +304,14 @@ struct afs_net { * cell, but in practice, people create aliases and subsets and there's * no easy way to distinguish them. */ - seqlock_t fs_lock; /* For fs_servers */ - struct rb_root fs_servers; /* afs_server (by server UUID or address) */ - struct list_head fs_updates; /* afs_server (by update_at) */ + seqlock_t fs_lock; /* For fs_probe_*, fs_proc */ + struct list_head fs_probe_fast; /* List of afs_server to probe at 30s intervals */ + struct list_head fs_probe_slow; /* List of afs_server to probe at 5m intervals */ struct hlist_head fs_proc; /* procfs servers list */ - struct hlist_head fs_addresses4; /* afs_server (by lowest IPv4 addr) */ - struct hlist_head fs_addresses6; /* afs_server (by lowest IPv6 addr) */ - seqlock_t fs_addr_lock; /* For fs_addresses[46] */ - - struct work_struct fs_manager; - struct timer_list fs_timer; + struct key *fs_cm_token_key; /* Key for creating CM tokens */ + struct work_struct fs_prober; + struct timer_list fs_probe_timer; atomic_t servers_outstanding; /* File locking renewal management */ @@ -295,12 +322,14 @@ struct afs_net { struct proc_dir_entry *proc_afs; /* /proc/net/afs directory */ struct afs_sysnames *sysnames; rwlock_t sysnames_lock; + struct afs_addr_preference_list __rcu *address_prefs; + u16 address_pref_version; /* Statistics counters */ atomic_t n_lookup; /* Number of lookups done */ atomic_t n_reval; /* Number of dentries needing revalidation */ atomic_t n_inval; /* Number of invalidations by the server */ - atomic_t n_relpg; /* Number of invalidations by releasepage */ + atomic_t n_relpg; /* Number of invalidations by release_folio */ atomic_t n_read_dir; /* Number of directory pages read */ atomic_t n_dir_cr; /* Number of directory entry creation edits */ atomic_t n_dir_rm; /* Number of directory entry removal edits */ @@ -313,12 +342,11 @@ struct afs_net { extern const char afs_init_sysname[]; enum afs_cell_state { - AFS_CELL_UNSET, - AFS_CELL_ACTIVATING, + AFS_CELL_SETTING_UP, + AFS_CELL_UNLOOKED, AFS_CELL_ACTIVE, - AFS_CELL_DEACTIVATING, - AFS_CELL_INACTIVE, - AFS_CELL_FAILED, + AFS_CELL_REMOVING, + AFS_CELL_DEAD, }; /* @@ -335,8 +363,10 @@ enum afs_cell_state { * for authentication and encryption. The cell name is not typically used in * the protocol. * - * There is no easy way to determine if two cells are aliases or one is a - * subset of another. + * Two cells are determined to be aliases if they have an explicit alias (YFS + * only), share any VL servers in common or have at least one volume in common. + * "In common" means that the address list of the VL servers or the fileservers + * share at least one endpoint. */ struct afs_cell { union { @@ -344,34 +374,46 @@ struct afs_cell { struct rb_node net_node; /* Node in net->cells */ }; struct afs_net *net; + struct afs_cell *alias_of; /* The cell this is an alias of */ + struct afs_volume *root_volume; /* The root.cell volume if there is one */ struct key *anonymous_key; /* anonymous user key for this cell */ + struct work_struct destroyer; /* Destroyer for cell */ struct work_struct manager; /* Manager for init/deinit/dns */ + struct timer_list management_timer; /* General management timer */ struct hlist_node proc_link; /* /proc cell list link */ -#ifdef CONFIG_AFS_FSCACHE - struct fscache_cookie *cache; /* caching cookie */ -#endif time64_t dns_expiry; /* Time AFSDB/SRV record expires */ time64_t last_inactive; /* Time of last drop of usage count */ - atomic_t usage; + refcount_t ref; /* Struct refcount */ + atomic_t active; /* Active usage counter */ unsigned long flags; -#define AFS_CELL_FL_NOT_READY 0 /* The cell record is not ready for use */ -#define AFS_CELL_FL_NO_GC 1 /* The cell was added manually, don't auto-gc */ -#define AFS_CELL_FL_NOT_FOUND 2 /* Permanent DNS error */ -#define AFS_CELL_FL_DNS_FAIL 3 /* Failed to access DNS */ -#define AFS_CELL_FL_NO_LOOKUP_YET 4 /* Not completed first DNS lookup yet */ +#define AFS_CELL_FL_NO_GC 0 /* The cell was added manually, don't auto-gc */ +#define AFS_CELL_FL_DO_LOOKUP 1 /* DNS lookup requested */ +#define AFS_CELL_FL_CHECK_ALIAS 2 /* Need to check for aliases */ enum afs_cell_state state; short error; + enum dns_record_source dns_source:8; /* Latest source of data from lookup */ + enum dns_lookup_status dns_status:8; /* Latest status of data from lookup */ + unsigned int dns_lookup_count; /* Counter of DNS lookups */ + unsigned int debug_id; + unsigned int dynroot_ino; /* Inode numbers for dynroot (a pair) */ + + /* The volumes belonging to this cell */ + struct rw_semaphore vs_lock; /* Lock for server->volumes */ + struct rb_root volumes; /* Tree of volumes on this server */ + struct hlist_head proc_volumes; /* procfs volume list */ + seqlock_t volume_lock; /* For volumes */ /* Active fileserver interaction state. */ - struct list_head proc_volumes; /* procfs volume list */ - rwlock_t proc_lock; + struct rb_root fs_servers; /* afs_server (by server UUID) */ + struct rw_semaphore fs_lock; /* For fs_servers */ /* VL server list. */ rwlock_t vl_servers_lock; /* Lock on vl_servers */ struct afs_vlserver_list __rcu *vl_servers; u8 name_len; /* Length of name */ - char name[64 + 1]; /* Cell name, case-flattened and NUL-padded */ + char *name; /* Cell name, case-flattened and NUL-padded */ + char *key_desc; /* Authentication key description */ }; /* @@ -384,24 +426,28 @@ struct afs_vlserver { #define AFS_VLSERVER_FL_PROBED 0 /* The VL server has been probed */ #define AFS_VLSERVER_FL_PROBING 1 /* VL server is being probed */ #define AFS_VLSERVER_FL_IS_YFS 2 /* Server is YFS not AFS */ +#define AFS_VLSERVER_FL_RESPONDING 3 /* VL server is responding */ rwlock_t lock; /* Lock on addresses */ - atomic_t usage; + refcount_t ref; + unsigned int rtt; /* Server's current RTT in uS */ + unsigned int debug_id; /* Probe state */ wait_queue_head_t probe_wq; atomic_t probe_outstanding; spinlock_t probe_lock; struct { - unsigned int rtt; /* RTT as ktime/64 */ + unsigned int rtt; /* Best RTT in uS (or UINT_MAX) */ u32 abort_code; short error; - bool have_result; - bool responded:1; - bool is_yfs:1; - bool not_yfs:1; - bool local_failure:1; + unsigned short flags; +#define AFS_VLSERVER_PROBE_RESPONDED 0x01 /* At least once response (may be abort) */ +#define AFS_VLSERVER_PROBE_IS_YFS 0x02 /* The peer appears to be YFS */ +#define AFS_VLSERVER_PROBE_NOT_YFS 0x04 /* The peer appears not to be YFS */ +#define AFS_VLSERVER_PROBE_LOCAL_FAILURE 0x08 /* A local failure prevented a probe */ } probe; + u16 service_id; /* Service ID we're using */ u16 port; u16 name_len; /* Length of name */ char name[]; /* Server name, case-flattened */ @@ -420,7 +466,7 @@ struct afs_vlserver_entry { struct afs_vlserver_list { struct rcu_head rcu; - atomic_t usage; + refcount_t ref; u8 nr_servers; u8 index; /* Server currently in use */ u8 preferred; /* Preferred server */ @@ -446,10 +492,12 @@ struct afs_vldb_entry { #define AFS_VLDB_QUERY_ERROR 4 /* - VL server returned error */ uuid_t fs_server[AFS_NMAXNSERVERS]; + u32 addr_version[AFS_NMAXNSERVERS]; /* Registration change counters */ u8 fs_mask[AFS_NMAXNSERVERS]; #define AFS_VOL_VTM_RW 0x01 /* R/W version of the volume is available (on this server) */ #define AFS_VOL_VTM_RO 0x02 /* R/O version of the volume is available (on this server) */ #define AFS_VOL_VTM_BAK 0x04 /* backup version of the volume is available (on this server) */ + u8 vlsf_flags[AFS_NMAXNSERVERS]; short error; u8 nr_servers; /* Number of server records */ u8 name_len; @@ -457,6 +505,32 @@ struct afs_vldb_entry { }; /* + * Fileserver endpoint state. The records the addresses of a fileserver's + * endpoints and the state and result of a round of probing on them. This + * allows the rotation algorithm to access those results without them being + * erased by a subsequent round of probing. + */ +struct afs_endpoint_state { + struct rcu_head rcu; + struct afs_addr_list *addresses; /* The addresses being probed */ + unsigned long responsive_set; /* Bitset of responsive endpoints */ + unsigned long failed_set; /* Bitset of endpoints we failed to probe */ + refcount_t ref; + unsigned int server_id; /* Debug ID of server */ + unsigned int probe_seq; /* Probe sequence (from server::probe_counter) */ + atomic_t nr_probing; /* Number of outstanding probes */ + unsigned int rtt; /* Best RTT in uS (or UINT_MAX) */ + s32 abort_code; + short error; + unsigned long flags; +#define AFS_ESTATE_RESPONDED 0 /* Set if the server responded */ +#define AFS_ESTATE_SUPERSEDED 1 /* Set if this record has been superseded */ +#define AFS_ESTATE_IS_YFS 2 /* Set if probe upgraded to YFS */ +#define AFS_ESTATE_NOT_YFS 3 /* Set if probe didn't upgrade to YFS */ +#define AFS_ESTATE_LOCAL_FAILURE 4 /* Set if there was a local failure (eg. ENOMEM) */ +}; + +/* * Record of fileserver with which we're actively communicating. */ struct afs_server { @@ -466,94 +540,76 @@ struct afs_server { struct afs_uuid _uuid; }; - struct afs_addr_list __rcu *addresses; - struct rb_node uuid_rb; /* Link in net->servers */ - struct hlist_node addr4_link; /* Link in net->fs_addresses4 */ - struct hlist_node addr6_link; /* Link in net->fs_addresses6 */ + struct afs_cell *cell; /* Cell to which belongs (pins ref) */ + struct rb_node uuid_rb; /* Link in cell->fs_servers */ + struct list_head probe_link; /* Link in net->fs_probe_* */ struct hlist_node proc_link; /* Link in net->fs_proc */ - struct afs_server *gc_next; /* Next server in manager's list */ - time64_t put_time; /* Time at which last put */ - time64_t update_at; /* Time at which to next update the record */ + struct list_head volumes; /* RCU list of afs_server_entry objects */ + struct work_struct destroyer; /* Work item to try and destroy a server */ + struct timer_list timer; /* Management timer */ + struct mutex cm_token_lock; /* Lock governing creation of appdata */ + struct krb5_buffer cm_rxgk_appdata; /* Appdata to be included in RESPONSE packet */ + time64_t unuse_time; /* Time at which last unused */ unsigned long flags; -#define AFS_SERVER_FL_NEW 0 /* New server, don't inc cb_s_break */ -#define AFS_SERVER_FL_NOT_READY 1 /* The record is not ready for use */ -#define AFS_SERVER_FL_NOT_FOUND 2 /* VL server says no such server */ -#define AFS_SERVER_FL_VL_FAIL 3 /* Failed to access VL server */ -#define AFS_SERVER_FL_UPDATING 4 -#define AFS_SERVER_FL_PROBED 5 /* The fileserver has been probed */ -#define AFS_SERVER_FL_PROBING 6 /* Fileserver is being probed */ -#define AFS_SERVER_FL_NO_IBULK 7 /* Fileserver doesn't support FS.InlineBulkStatus */ +#define AFS_SERVER_FL_RESPONDING 0 /* The server is responding */ +#define AFS_SERVER_FL_UPDATING 1 +#define AFS_SERVER_FL_NEEDS_UPDATE 2 /* Fileserver address list is out of date */ +#define AFS_SERVER_FL_UNCREATED 3 /* The record needs creating */ +#define AFS_SERVER_FL_CREATING 4 /* The record is being created */ +#define AFS_SERVER_FL_EXPIRED 5 /* The record has expired */ +#define AFS_SERVER_FL_NOT_FOUND 6 /* VL server says no such server */ +#define AFS_SERVER_FL_VL_FAIL 7 /* Failed to access VL server */ #define AFS_SERVER_FL_MAY_HAVE_CB 8 /* May have callbacks on this fileserver */ -#define AFS_SERVER_FL_IS_YFS 9 /* Server is YFS not AFS */ -#define AFS_SERVER_FL_NO_RM2 10 /* Fileserver doesn't support YFS.RemoveFile2 */ -#define AFS_SERVER_FL_HAVE_EPOCH 11 /* ->epoch is valid */ - atomic_t usage; +#define AFS_SERVER_FL_IS_YFS 16 /* Server is YFS not AFS */ +#define AFS_SERVER_FL_NO_IBULK 17 /* Fileserver doesn't support FS.InlineBulkStatus */ +#define AFS_SERVER_FL_NO_RM2 18 /* Fileserver doesn't support YFS.RemoveFile2 */ +#define AFS_SERVER_FL_HAS_FS64 19 /* Fileserver supports FS.{Fetch,Store}Data64 */ +#define AFS_SERVER_FL_NO_RENAME2 20 /* YFS Fileserver doesn't support enhanced rename */ + refcount_t ref; /* Object refcount */ + atomic_t active; /* Active user count */ u32 addr_version; /* Address list version */ - u32 cm_epoch; /* Server RxRPC epoch */ + u16 service_id; /* Service ID we're using. */ + short create_error; /* Creation error */ + unsigned int rtt; /* Server's current RTT in uS */ + unsigned int debug_id; /* Debugging ID for traces */ /* file service access */ rwlock_t fs_lock; /* access lock */ - /* callback promise management */ - struct hlist_head cb_volumes; /* List of volume interests on this server */ - unsigned cb_s_break; /* Break-everything counter. */ - rwlock_t cb_break_lock; /* Volume finding lock */ - /* Probe state */ + struct afs_endpoint_state __rcu *endpoint_state; /* Latest endpoint/probe state */ + unsigned long probed_at; /* Time last probe was dispatched (jiffies) */ wait_queue_head_t probe_wq; - atomic_t probe_outstanding; + unsigned int probe_counter; /* Number of probes issued */ spinlock_t probe_lock; - struct { - unsigned int rtt; /* RTT as ktime/64 */ - u32 abort_code; - u32 cm_epoch; - short error; - bool have_result; - bool responded:1; - bool is_yfs:1; - bool not_yfs:1; - bool local_failure:1; - bool no_epoch:1; - bool cm_probed:1; - bool said_rebooted:1; - bool said_inconsistent:1; - } probe; -}; - -/* - * Volume collation in the server's callback interest list. - */ -struct afs_vol_interest { - struct hlist_node srv_link; /* Link in server->cb_volumes */ - struct hlist_head cb_interests; /* List of callback interests on the server */ - afs_volid_t vid; /* Volume ID to match */ - unsigned int usage; }; -/* - * Interest by a superblock on a server. - */ -struct afs_cb_interest { - struct hlist_node cb_vlink; /* Link in vol_interest->cb_interests */ - struct afs_vol_interest *vol_interest; - struct afs_server *server; /* Server on which this interest resides */ - struct super_block *sb; /* Superblock on which inodes reside */ - afs_volid_t vid; /* Volume ID to match */ - refcount_t usage; -}; +enum afs_ro_replicating { + AFS_RO_NOT_REPLICATING, /* Not doing replication */ + AFS_RO_REPLICATING_USE_OLD, /* Replicating; use old version */ + AFS_RO_REPLICATING_USE_NEW, /* Replicating; switch to new version */ +} __mode(byte); /* - * Replaceable server list. + * Replaceable volume server list. */ struct afs_server_entry { struct afs_server *server; - struct afs_cb_interest *cb_interest; + struct afs_volume *volume; + struct list_head slink; /* Link in server->volumes */ + time64_t cb_expires_at; /* Time at which volume-level callback expires */ + unsigned long flags; +#define AFS_SE_EXCLUDED 0 /* Set if server is to be excluded in rotation */ +#define AFS_SE_VOLUME_OFFLINE 1 /* Set if volume offline notice given */ +#define AFS_SE_VOLUME_BUSY 2 /* Set if volume busy notice given */ }; struct afs_server_list { + struct rcu_head rcu; refcount_t usage; + bool attached; /* T if attached to servers */ + enum afs_ro_replicating ro_replicating; /* RW->RO update (probably) in progress */ unsigned char nr_servers; - unsigned char preferred; /* Preferred server */ unsigned short vnovol_mask; /* Servers to be skipped due to VNOVOL */ unsigned int seq; /* Set to ->servers_seq when installed */ rwlock_t lock; @@ -564,30 +620,48 @@ struct afs_server_list { * Live AFS volume management. */ struct afs_volume { - afs_volid_t vid; /* volume ID */ - atomic_t usage; + struct rcu_head rcu; + afs_volid_t vid; /* The volume ID of this volume */ + afs_volid_t vids[AFS_MAXTYPES]; /* All associated volume IDs */ + refcount_t ref; + unsigned int debug_id; /* Debugging ID for traces */ time64_t update_at; /* Time at which to next update */ struct afs_cell *cell; /* Cell to which belongs (pins ref) */ - struct list_head proc_link; /* Link in cell->vl_proc */ + struct rb_node cell_node; /* Link in cell->volumes */ + struct hlist_node proc_link; /* Link in cell->proc_volumes */ + struct super_block __rcu *sb; /* Superblock on which inodes reside */ + struct work_struct destructor; /* Deferred destructor */ unsigned long flags; #define AFS_VOLUME_NEEDS_UPDATE 0 /* - T if an update needs performing */ #define AFS_VOLUME_UPDATING 1 /* - T if an update is in progress */ #define AFS_VOLUME_WAIT 2 /* - T if users must wait for update */ #define AFS_VOLUME_DELETED 3 /* - T if volume appears deleted */ -#define AFS_VOLUME_OFFLINE 4 /* - T if volume offline notice given */ -#define AFS_VOLUME_BUSY 5 /* - T if volume busy notice given */ +#define AFS_VOLUME_MAYBE_NO_IBULK 4 /* - T if some servers don't have InlineBulkStatus */ +#define AFS_VOLUME_RM_TREE 5 /* - Set if volume removed from cell->volumes */ #ifdef CONFIG_AFS_FSCACHE - struct fscache_cookie *cache; /* caching cookie */ + struct fscache_volume *cache; /* Caching cookie */ #endif - struct afs_server_list *servers; /* List of servers on which volume resides */ + struct afs_server_list __rcu *servers; /* List of servers on which volume resides */ rwlock_t servers_lock; /* Lock for ->servers */ unsigned int servers_seq; /* Incremented each time ->servers changes */ - unsigned cb_v_break; /* Break-everything counter. */ - rwlock_t cb_break_lock; + /* RO release tracking */ + struct mutex volsync_lock; /* Time/state evaluation lock */ + time64_t creation_time; /* Volume creation time (or TIME64_MIN) */ + time64_t update_time; /* Volume update time (or TIME64_MIN) */ + + /* Callback management */ + struct mutex cb_check_lock; /* Lock to control race to check after v_break */ + time64_t cb_expires_at; /* Earliest volume callback expiry time */ + atomic_t cb_ro_snapshot; /* RO volume update-from-snapshot counter */ + atomic_t cb_v_break; /* Volume-break event counter. */ + atomic_t cb_v_check; /* Volume-break has-been-checked counter. */ + atomic_t cb_scrub; /* Scrub-all-data event counter. */ + rwlock_t cb_v_break_lock; + struct rw_semaphore open_mmaps_lock; + struct list_head open_mmaps; /* List of vnodes that are mmapped */ afs_voltype_t type; /* type of volume */ - short error; char type_force; /* force volume type (suppress R/O -> R/W) */ u8 name_len; u8 name[AFS_MAXVOLNAME + 1]; /* NUL-padded volume name */ @@ -601,6 +675,7 @@ enum afs_lock_state { AFS_VNODE_LOCK_EXTENDING, /* We're extending a lock on the server */ AFS_VNODE_LOCK_NEED_UNLOCK, /* We need to unlock on the server */ AFS_VNODE_LOCK_UNLOCKING, /* We're telling the server to unlock */ + AFS_VNODE_LOCK_DELETED, /* The vnode has been deleted whilst we have a lock */ }; /* @@ -610,60 +685,76 @@ enum afs_lock_state { * leak from one inode to another. */ struct afs_vnode { - struct inode vfs_inode; /* the VFS's inode record */ - + struct netfs_inode netfs; /* Netfslib context and vfs inode */ struct afs_volume *volume; /* volume on which vnode resides */ struct afs_fid fid; /* the file identifier for this inode */ struct afs_file_status status; /* AFS status info for this file */ afs_dataversion_t invalid_before; /* Child dentries are invalid before this */ -#ifdef CONFIG_AFS_FSCACHE - struct fscache_cookie *cache; /* caching cookie */ -#endif struct afs_permits __rcu *permit_cache; /* cache of permits so far obtained */ - struct mutex io_lock; /* Lock for serialising I/O on this mutex */ + struct list_head io_lock_waiters; /* Threads waiting for the I/O lock */ struct rw_semaphore validate_lock; /* lock for validating this vnode */ + struct rw_semaphore rmdir_lock; /* Lock for rmdir vs sillyrename */ + struct key *silly_key; /* Silly rename key */ spinlock_t wb_lock; /* lock for wb_keys */ spinlock_t lock; /* waitqueue/flags lock */ unsigned long flags; -#define AFS_VNODE_CB_PROMISED 0 /* Set if vnode has a callback promise */ +#define AFS_VNODE_IO_LOCK 0 /* Set if the I/O serialisation lock is held */ #define AFS_VNODE_UNSET 1 /* set if vnode attributes not yet set */ #define AFS_VNODE_DIR_VALID 2 /* Set if dir contents are valid */ #define AFS_VNODE_ZAP_DATA 3 /* set if vnode's data should be invalidated */ #define AFS_VNODE_DELETED 4 /* set if vnode deleted on server */ #define AFS_VNODE_MOUNTPOINT 5 /* set if vnode is a mountpoint symlink */ -#define AFS_VNODE_AUTOCELL 6 /* set if Vnode is an auto mount point */ #define AFS_VNODE_PSEUDODIR 7 /* set if Vnode is a pseudo directory */ #define AFS_VNODE_NEW_CONTENT 8 /* Set if file has new content (create/trunc-0) */ +#define AFS_VNODE_SILLY_DELETED 9 /* Set if file has been silly-deleted */ +#define AFS_VNODE_MODIFYING 10 /* Set if we're performing a modification op */ +#define AFS_VNODE_DIR_READ 11 /* Set if we've read a dir's contents */ + struct folio_queue *directory; /* Directory contents */ struct list_head wb_keys; /* List of keys available for writeback */ struct list_head pending_locks; /* locks waiting to be granted */ struct list_head granted_locks; /* locks granted on this file */ struct delayed_work lock_work; /* work to be done in locking */ struct key *lock_key; /* Key to be used in lock ops */ + ktime_t locked_at; /* Time at which lock obtained */ enum afs_lock_state lock_state : 8; afs_lock_type_t lock_type : 8; + unsigned int directory_size; /* Amount of space in ->directory */ /* outstanding callback notification on this file */ - struct afs_cb_interest *cb_interest; /* Server on which this resides */ - unsigned int cb_s_break; /* Mass break counter on ->server */ - unsigned int cb_v_break; /* Mass break counter on ->volume */ + struct work_struct cb_work; /* Work for mmap'd files */ + struct list_head cb_mmap_link; /* Link in cell->fs_open_mmaps */ + void *cb_server; /* Server with callback/filelock */ + atomic_t cb_nr_mmap; /* Number of mmaps */ + unsigned int cb_ro_snapshot; /* RO volume release counter on ->volume */ + unsigned int cb_scrub; /* Scrub counter on ->volume */ unsigned int cb_break; /* Break counter on vnode */ - seqlock_t cb_lock; /* Lock for ->cb_interest, ->status, ->cb_*break */ + unsigned int cb_v_check; /* Break check counter on ->volume */ + seqlock_t cb_lock; /* Lock for ->cb_server, ->status, ->cb_*break */ - time64_t cb_expires_at; /* time at which callback expires */ - unsigned cb_version; /* callback version */ - afs_callback_type_t cb_type; /* type of callback */ + atomic64_t cb_expires_at; /* time at which callback expires */ +#define AFS_NO_CB_PROMISE TIME64_MIN }; static inline struct fscache_cookie *afs_vnode_cache(struct afs_vnode *vnode) { #ifdef CONFIG_AFS_FSCACHE - return vnode->cache; + return netfs_i_cookie(&vnode->netfs); #else return NULL; #endif } +static inline void afs_vnode_set_cache(struct afs_vnode *vnode, + struct fscache_cookie *cookie) +{ +#ifdef CONFIG_AFS_FSCACHE + vnode->netfs.cache = cookie; + if (cookie) + mapping_set_release_always(vnode->netfs.inode.i_mapping); +#endif +} + /* * cached security record for one user's attempt to access a vnode */ @@ -683,174 +774,303 @@ struct afs_permits { refcount_t usage; unsigned short nr_permits; /* Number of records */ bool invalidated; /* Invalidated due to key change */ - struct afs_permit permits[]; /* List of permits sorted by key pointer */ -}; - -/* - * record of one of a system's set of network interfaces - */ -struct afs_interface { - struct in_addr address; /* IPv4 address bound to interface */ - struct in_addr netmask; /* netmask applied to address */ - unsigned mtu; /* MTU of interface */ + struct afs_permit permits[] __counted_by(nr_permits); /* List of permits sorted by key pointer */ }; /* * Error prioritisation and accumulation. */ struct afs_error { - short error; /* Accumulated error */ + s32 abort_code; /* Cumulative abort code */ + short error; /* Cumulative error */ bool responded; /* T if server responded */ -}; - -/* - * Cursor for iterating over a server's address list. - */ -struct afs_addr_cursor { - struct afs_addr_list *alist; /* Current address list (pins ref) */ - unsigned long tried; /* Tried addresses */ - signed char index; /* Current address */ - bool responded; /* T if the current address responded */ - unsigned short nr_iterations; /* Number of address iterations */ - short error; - u32 abort_code; + bool aborted; /* T if ->error is from an abort */ }; /* * Cursor for iterating over a set of volume location servers. */ struct afs_vl_cursor { - struct afs_addr_cursor ac; struct afs_cell *cell; /* The cell we're querying */ struct afs_vlserver_list *server_list; /* Current server list (pins ref) */ struct afs_vlserver *server; /* Server on which this resides */ + struct afs_addr_list *alist; /* Current address list (pins ref) */ struct key *key; /* Key for the server */ - unsigned long untried; /* Bitmask of untried servers */ - short index; /* Current server */ - short error; + unsigned long untried_servers; /* Bitmask of untried servers */ + unsigned long addr_tried; /* Tried addresses */ + struct afs_error cumul_error; /* Cumulative error */ + unsigned int debug_id; + s32 call_abort_code; + short call_error; /* Error from single call */ + short server_index; /* Current server */ + signed char addr_index; /* Current address */ unsigned short flags; #define AFS_VL_CURSOR_STOP 0x0001 /* Set to cease iteration */ #define AFS_VL_CURSOR_RETRY 0x0002 /* Set to do a retry */ #define AFS_VL_CURSOR_RETRIED 0x0004 /* Set if started a retry */ - unsigned short nr_iterations; /* Number of server iterations */ + short nr_iterations; /* Number of server iterations */ + bool call_responded; /* T if the current address responded */ +}; + +/* + * Fileserver state tracking for an operation. An array of these is kept, + * indexed by server index. + */ +struct afs_server_state { + /* Tracking of fileserver probe state. Other operations may interfere + * by probing a fileserver when accessing other volumes. + */ + unsigned int probe_seq; + unsigned long untried_addrs; /* Addresses we haven't tried yet */ + struct wait_queue_entry probe_waiter; + struct afs_endpoint_state *endpoint_state; /* Endpoint state being monitored */ }; /* - * Cursor for iterating over a set of fileservers. + * Fileserver operation methods. */ -struct afs_fs_cursor { - struct afs_addr_cursor ac; +struct afs_operation_ops { + void (*issue_afs_rpc)(struct afs_operation *op); + void (*issue_yfs_rpc)(struct afs_operation *op); + void (*success)(struct afs_operation *op); + void (*aborted)(struct afs_operation *op); + void (*failed)(struct afs_operation *op); + void (*edit_dir)(struct afs_operation *op); + void (*put)(struct afs_operation *op); +}; + +struct afs_vnode_param { struct afs_vnode *vnode; + struct afs_fid fid; /* Fid to access */ + struct afs_status_cb scb; /* Returned status and callback promise */ + afs_dataversion_t dv_before; /* Data version before the call */ + unsigned int cb_break_before; /* cb_break before the call */ + u8 dv_delta; /* Expected change in data version */ + bool put_vnode:1; /* T if we have a ref on the vnode */ + bool need_io_lock:1; /* T if we need the I/O lock on this */ + bool update_ctime:1; /* Need to update the ctime */ + bool set_size:1; /* Must update i_size */ + bool op_unlinked:1; /* True if file was unlinked by op */ + bool speculative:1; /* T if speculative status fetch (no vnode lock) */ + bool modification:1; /* Set if the content gets modified */ +}; + +/* + * Fileserver operation wrapper, handling server and address rotation + * asynchronously. May make simultaneous calls to multiple servers. + */ +struct afs_operation { + struct afs_net *net; /* Network namespace */ + struct key *key; /* Key for the cell */ + const struct afs_call_type *type; /* Type of call done */ + const struct afs_operation_ops *ops; + + /* Parameters/results for the operation */ + struct afs_volume *volume; /* Volume being accessed */ + struct afs_vnode_param file[2]; + struct afs_vnode_param *more_files; + struct afs_volsync pre_volsync; /* Volsync before op */ + struct afs_volsync volsync; /* Volsync returned by op */ + struct dentry *dentry; /* Dentry to be altered */ + struct dentry *dentry_2; /* Second dentry to be altered */ + struct timespec64 mtime; /* Modification time to record */ + struct timespec64 ctime; /* Change time to set */ + struct afs_error cumul_error; /* Cumulative error */ + short nr_files; /* Number of entries in file[], more_files */ + unsigned int debug_id; + + unsigned int cb_v_break; /* Volume break counter before op */ + + union { + struct { + int which; /* Which ->file[] to fetch for */ + } fetch_status; + struct { + int reason; /* enum afs_edit_dir_reason */ + mode_t mode; + const char *symlink; + } create; + struct { + bool need_rehash; + } unlink; + struct { + struct dentry *rehash; + struct dentry *tmp; + unsigned int rename_flags; + bool new_negative; + } rename; + struct { + struct netfs_io_subrequest *subreq; + } fetch; + struct { + afs_lock_type_t type; + } lock; + struct { + struct iov_iter *write_iter; + loff_t pos; + loff_t size; + loff_t i_size; + } store; + struct { + struct iattr *attr; + loff_t old_i_size; + } setattr; + struct afs_acl *acl; + struct yfs_acl *yacl; + struct { + struct afs_volume_status vs; + struct kstatfs *buf; + } volstatus; + }; + + /* Fileserver iteration state */ struct afs_server_list *server_list; /* Current server list (pins ref) */ - struct afs_cb_interest *cbi; /* Server on which this resides (pins ref) */ - struct key *key; /* Key for the server */ - unsigned long untried; /* Bitmask of untried servers */ - unsigned int cb_break; /* cb_break + cb_s_break before the call */ - unsigned int cb_break_2; /* cb_break + cb_s_break (2nd vnode) */ - short index; /* Current server */ - short error; - unsigned short flags; -#define AFS_FS_CURSOR_STOP 0x0001 /* Set to cease iteration */ -#define AFS_FS_CURSOR_VBUSY 0x0002 /* Set if seen VBUSY */ -#define AFS_FS_CURSOR_VMOVED 0x0004 /* Set if seen VMOVED */ -#define AFS_FS_CURSOR_VNOVOL 0x0008 /* Set if seen VNOVOL */ -#define AFS_FS_CURSOR_CUR_ONLY 0x0010 /* Set if current server only (file lock held) */ -#define AFS_FS_CURSOR_NO_VSLEEP 0x0020 /* Set to prevent sleep on VBUSY, VOFFLINE, ... */ - unsigned short nr_iterations; /* Number of server iterations */ + struct afs_server *server; /* Server we're using (ref pinned by server_list) */ + struct afs_endpoint_state *estate; /* Current endpoint state (doesn't pin ref) */ + struct afs_server_state *server_states; /* States of the servers involved */ + struct afs_call *call; + unsigned long untried_servers; /* Bitmask of untried servers */ + unsigned long addr_tried; /* Tried addresses */ + s32 call_abort_code; /* Abort code from single call */ + short call_error; /* Error from single call */ + short server_index; /* Current server */ + short nr_iterations; /* Number of server iterations */ + signed char addr_index; /* Current address */ + bool call_responded; /* T if the current address responded */ + + unsigned int flags; +#define AFS_OPERATION_STOP 0x0001 /* Set to cease iteration */ +#define AFS_OPERATION_VBUSY 0x0002 /* Set if seen VBUSY */ +#define AFS_OPERATION_VMOVED 0x0004 /* Set if seen VMOVED */ +#define AFS_OPERATION_VNOVOL 0x0008 /* Set if seen VNOVOL */ +#define AFS_OPERATION_CUR_ONLY 0x0010 /* Set if current server only (file lock held) */ +#define AFS_OPERATION_NO_VSLEEP 0x0020 /* Set to prevent sleep on VBUSY, VOFFLINE, ... */ +#define AFS_OPERATION_UNINTR 0x0040 /* Set if op is uninterruptible */ +#define AFS_OPERATION_DOWNGRADE 0x0080 /* Set to retry with downgraded opcode */ +#define AFS_OPERATION_LOCK_0 0x0100 /* Set if have io_lock on file[0] */ +#define AFS_OPERATION_LOCK_1 0x0200 /* Set if have io_lock on file[1] */ +#define AFS_OPERATION_TRIED_ALL 0x0400 /* Set if we've tried all the fileservers */ +#define AFS_OPERATION_RETRY_SERVER 0x0800 /* Set if we should retry the current server */ +#define AFS_OPERATION_DIR_CONFLICT 0x1000 /* Set if we detected a 3rd-party dir change */ +#define AFS_OPERATION_ASYNC 0x2000 /* Set if should run asynchronously */ }; /* * Cache auxiliary data. */ struct afs_vnode_cache_aux { - u64 data_version; + __be64 data_version; } __packed; +static inline void afs_set_cache_aux(struct afs_vnode *vnode, + struct afs_vnode_cache_aux *aux) +{ + aux->data_version = cpu_to_be64(vnode->status.data_version); +} + +static inline void afs_invalidate_cache(struct afs_vnode *vnode, unsigned int flags) +{ + struct afs_vnode_cache_aux aux; + + afs_set_cache_aux(vnode, &aux); + fscache_invalidate(afs_vnode_cache(vnode), &aux, + i_size_read(&vnode->netfs.inode), flags); +} + +/* + * Directory iteration management. + */ +struct afs_dir_iter { + struct afs_vnode *dvnode; + union afs_xdr_dir_block *block; + struct folio_queue *fq; + unsigned int fpos; + int fq_slot; + unsigned int loop_check; + u8 nr_slots; + u8 bucket; + unsigned int prev_entry; +}; + #include <trace/events/afs.h> /*****************************************************************************/ /* * addr_list.c */ -static inline struct afs_addr_list *afs_get_addrlist(struct afs_addr_list *alist) -{ - if (alist) - refcount_inc(&alist->usage); - return alist; -} -extern struct afs_addr_list *afs_alloc_addrlist(unsigned int, - unsigned short, - unsigned short); -extern void afs_put_addrlist(struct afs_addr_list *); +struct afs_addr_list *afs_get_addrlist(struct afs_addr_list *alist, enum afs_alist_trace reason); +extern struct afs_addr_list *afs_alloc_addrlist(unsigned int nr); +extern void afs_put_addrlist(struct afs_addr_list *alist, enum afs_alist_trace reason); extern struct afs_vlserver_list *afs_parse_text_addrs(struct afs_net *, const char *, size_t, char, unsigned short, unsigned short); +bool afs_addr_list_same(const struct afs_addr_list *a, + const struct afs_addr_list *b); extern struct afs_vlserver_list *afs_dns_query(struct afs_cell *, time64_t *); -extern bool afs_iterate_addresses(struct afs_addr_cursor *); -extern int afs_end_cursor(struct afs_addr_cursor *); -extern void afs_merge_fs_addr4(struct afs_addr_list *, __be32, u16); -extern void afs_merge_fs_addr6(struct afs_addr_list *, __be32 *, u16); +extern int afs_merge_fs_addr4(struct afs_net *net, struct afs_addr_list *addr, + __be32 xdr, u16 port); +extern int afs_merge_fs_addr6(struct afs_net *net, struct afs_addr_list *addr, + __be32 *xdr, u16 port); +void afs_set_peer_appdata(struct afs_server *server, + struct afs_addr_list *old_alist, + struct afs_addr_list *new_alist); /* - * cache.c + * addr_prefs.c */ -#ifdef CONFIG_AFS_FSCACHE -extern struct fscache_netfs afs_cache_netfs; -extern struct fscache_cookie_def afs_cell_cache_index_def; -extern struct fscache_cookie_def afs_volume_cache_index_def; -extern struct fscache_cookie_def afs_vnode_cache_index_def; -#else -#define afs_cell_cache_index_def (*(struct fscache_cookie_def *) NULL) -#define afs_volume_cache_index_def (*(struct fscache_cookie_def *) NULL) -#define afs_vnode_cache_index_def (*(struct fscache_cookie_def *) NULL) -#endif +int afs_proc_addr_prefs_write(struct file *file, char *buf, size_t size); +void afs_get_address_preferences_rcu(struct afs_net *net, struct afs_addr_list *alist); +void afs_get_address_preferences(struct afs_net *net, struct afs_addr_list *alist); /* * callback.c */ +extern void afs_invalidate_mmap_work(struct work_struct *); extern void afs_init_callback_state(struct afs_server *); -extern void __afs_break_callback(struct afs_vnode *); -extern void afs_break_callback(struct afs_vnode *); -extern void afs_break_callbacks(struct afs_server *, size_t, struct afs_callback_break*); - -extern int afs_register_server_cb_interest(struct afs_vnode *, - struct afs_server_list *, unsigned int); -extern void afs_put_cb_interest(struct afs_net *, struct afs_cb_interest *); -extern void afs_clear_callback_interests(struct afs_net *, struct afs_server_list *); - -static inline struct afs_cb_interest *afs_get_cb_interest(struct afs_cb_interest *cbi) -{ - if (cbi) - refcount_inc(&cbi->usage); - return cbi; -} +extern void __afs_break_callback(struct afs_vnode *, enum afs_cb_break_reason); +extern void afs_break_callback(struct afs_vnode *, enum afs_cb_break_reason); +extern void afs_break_callbacks(struct afs_server *, size_t, struct afs_callback_break *); static inline unsigned int afs_calc_vnode_cb_break(struct afs_vnode *vnode) { - return vnode->cb_break + vnode->cb_s_break + vnode->cb_v_break; + return vnode->cb_break + vnode->cb_ro_snapshot + vnode->cb_scrub; } static inline bool afs_cb_is_broken(unsigned int cb_break, - const struct afs_vnode *vnode, - const struct afs_cb_interest *cbi) + const struct afs_vnode *vnode) { - return !cbi || cb_break != (vnode->cb_break + - cbi->server->cb_s_break + - vnode->volume->cb_v_break); + return cb_break != (vnode->cb_break + + atomic_read(&vnode->volume->cb_ro_snapshot) + + atomic_read(&vnode->volume->cb_scrub)); } /* * cell.c */ extern int afs_cell_init(struct afs_net *, const char *); -extern struct afs_cell *afs_lookup_cell_rcu(struct afs_net *, const char *, unsigned); -extern struct afs_cell *afs_lookup_cell(struct afs_net *, const char *, unsigned, - const char *, bool); -extern struct afs_cell *afs_get_cell(struct afs_cell *); -extern void afs_put_cell(struct afs_net *, struct afs_cell *); -extern void afs_manage_cells(struct work_struct *); -extern void afs_cells_timer(struct timer_list *); +extern struct afs_cell *afs_find_cell(struct afs_net *, const char *, unsigned, + enum afs_cell_trace); +enum afs_lookup_cell_for { + AFS_LOOKUP_CELL_DYNROOT, + AFS_LOOKUP_CELL_MOUNTPOINT, + AFS_LOOKUP_CELL_DIRECT_MOUNT, + AFS_LOOKUP_CELL_PRELOAD, + AFS_LOOKUP_CELL_ROOTCELL, + AFS_LOOKUP_CELL_ALIAS_CHECK, +}; +struct afs_cell *afs_lookup_cell(struct afs_net *net, + const char *name, unsigned int namesz, + const char *vllist, + enum afs_lookup_cell_for reason, + enum afs_cell_trace trace); +extern struct afs_cell *afs_use_cell(struct afs_cell *, enum afs_cell_trace); +void afs_unuse_cell(struct afs_cell *cell, enum afs_cell_trace reason); +extern struct afs_cell *afs_get_cell(struct afs_cell *, enum afs_cell_trace); +extern void afs_see_cell(struct afs_cell *, enum afs_cell_trace); +extern void afs_put_cell(struct afs_cell *, enum afs_cell_trace); +extern void afs_queue_cell(struct afs_cell *, enum afs_cell_trace); +void afs_set_cell_timer(struct afs_cell *cell, unsigned int delay_secs); extern void __net_exit afs_cell_purge(struct afs_net *); /* @@ -859,6 +1079,19 @@ extern void __net_exit afs_cell_purge(struct afs_net *); extern bool afs_cm_incoming_call(struct afs_call *); /* + * cm_security.c + */ +void afs_process_oob_queue(struct work_struct *work); +#ifdef CONFIG_RXGK +int afs_create_token_key(struct afs_net *net, struct socket *socket); +#else +static inline int afs_create_token_key(struct afs_net *net, struct socket *socket) +{ + return 0; +} +#endif + +/* * dir.c */ extern const struct file_operations afs_dir_file_operations; @@ -866,48 +1099,72 @@ extern const struct inode_operations afs_dir_inode_operations; extern const struct address_space_operations afs_dir_aops; extern const struct dentry_operations afs_fs_dentry_operations; +ssize_t afs_read_single(struct afs_vnode *dvnode, struct file *file); +ssize_t afs_read_dir(struct afs_vnode *dvnode, struct file *file) + __acquires(&dvnode->validate_lock); extern void afs_d_release(struct dentry *); +extern void afs_check_for_remote_deletion(struct afs_operation *); +int afs_single_writepages(struct address_space *mapping, + struct writeback_control *wbc); /* * dir_edit.c */ -extern void afs_edit_dir_add(struct afs_vnode *, struct qstr *, struct afs_fid *, +extern void afs_edit_dir_add(struct afs_vnode *, const struct qstr *, struct afs_fid *, enum afs_edit_dir_reason); -extern void afs_edit_dir_remove(struct afs_vnode *, struct qstr *, enum afs_edit_dir_reason); +extern void afs_edit_dir_remove(struct afs_vnode *, const struct qstr *, enum afs_edit_dir_reason); +void afs_edit_dir_update(struct afs_vnode *vnode, const struct qstr *name, + struct afs_vnode *new_dvnode, enum afs_edit_dir_reason why); +void afs_mkdir_init_dir(struct afs_vnode *dvnode, struct afs_vnode *parent_vnode); + +/* + * dir_search.c + */ +unsigned int afs_dir_hash_name(const struct qstr *name); +bool afs_dir_init_iter(struct afs_dir_iter *iter, const struct qstr *name); +union afs_xdr_dir_block *afs_dir_find_block(struct afs_dir_iter *iter, size_t block); +int afs_dir_search_bucket(struct afs_dir_iter *iter, const struct qstr *name, + struct afs_fid *_fid); +int afs_dir_search(struct afs_vnode *dvnode, const struct qstr *name, + struct afs_fid *_fid, afs_dataversion_t *_dir_version); + +/* + * dir_silly.c + */ +extern int afs_sillyrename(struct afs_vnode *, struct afs_vnode *, + struct dentry *, struct key *); +extern int afs_silly_iput(struct dentry *, struct inode *); /* * dynroot.c */ -extern const struct file_operations afs_dynroot_file_operations; extern const struct inode_operations afs_dynroot_inode_operations; extern const struct dentry_operations afs_dynroot_dentry_operations; -extern struct inode *afs_try_auto_mntpt(struct dentry *, struct inode *); -extern int afs_dynroot_mkdir(struct afs_net *, struct afs_cell *); -extern void afs_dynroot_rmdir(struct afs_net *, struct afs_cell *); -extern int afs_dynroot_populate(struct super_block *); -extern void afs_dynroot_depopulate(struct super_block *); +struct inode *afs_dynroot_iget_root(struct super_block *sb); /* * file.c */ -extern const struct address_space_operations afs_fs_aops; +extern const struct address_space_operations afs_file_aops; extern const struct inode_operations afs_file_inode_operations; extern const struct file_operations afs_file_operations; +extern const struct afs_operation_ops afs_fetch_data_operation; +extern const struct netfs_request_ops afs_req_ops; extern int afs_cache_wb_key(struct afs_vnode *, struct afs_file *); extern void afs_put_wb_key(struct afs_wb_key *); extern int afs_open(struct inode *, struct file *); extern int afs_release(struct inode *, struct file *); -extern int afs_fetch_data(struct afs_vnode *, struct key *, struct afs_read *); -extern int afs_page_filler(void *, struct page *); -extern void afs_put_read(struct afs_read *); +void afs_fetch_data_async_rx(struct work_struct *work); +void afs_fetch_data_immediate_cancel(struct afs_call *call); /* * flock.c */ extern struct workqueue_struct *afs_lock_manager; +extern void afs_lock_op_done(struct afs_call *); extern void afs_lock_work(struct work_struct *); extern void afs_lock_may_be_available(struct afs_vnode *); extern int afs_lock(struct file *, int, struct file_lock *); @@ -916,63 +1173,92 @@ extern int afs_flock(struct file *, int, struct file_lock *); /* * fsclient.c */ -#define AFS_VNODE_NOT_YET_SET 0x01 -#define AFS_VNODE_META_CHANGED 0x02 -#define AFS_VNODE_DATA_CHANGED 0x04 -extern void afs_update_inode_from_status(struct afs_vnode *, struct afs_file_status *, - const afs_dataversion_t *, u8); - -extern int afs_fs_fetch_file_status(struct afs_fs_cursor *, struct afs_volsync *, bool); -extern int afs_fs_give_up_callbacks(struct afs_net *, struct afs_server *); -extern int afs_fs_fetch_data(struct afs_fs_cursor *, struct afs_read *); -extern int afs_fs_create(struct afs_fs_cursor *, const char *, umode_t, u64, - struct afs_fid *, struct afs_file_status *, struct afs_callback *); -extern int afs_fs_remove(struct afs_fs_cursor *, struct afs_vnode *, const char *, bool, u64); -extern int afs_fs_link(struct afs_fs_cursor *, struct afs_vnode *, const char *, u64); -extern int afs_fs_symlink(struct afs_fs_cursor *, const char *, const char *, u64, - struct afs_fid *, struct afs_file_status *); -extern int afs_fs_rename(struct afs_fs_cursor *, const char *, - struct afs_vnode *, const char *, u64, u64); -extern int afs_fs_store_data(struct afs_fs_cursor *, struct address_space *, - pgoff_t, pgoff_t, unsigned, unsigned); -extern int afs_fs_setattr(struct afs_fs_cursor *, struct iattr *); -extern int afs_fs_get_volume_status(struct afs_fs_cursor *, struct afs_volume_status *); -extern int afs_fs_set_lock(struct afs_fs_cursor *, afs_lock_type_t); -extern int afs_fs_extend_lock(struct afs_fs_cursor *); -extern int afs_fs_release_lock(struct afs_fs_cursor *); -extern int afs_fs_give_up_all_callbacks(struct afs_net *, struct afs_server *, - struct afs_addr_cursor *, struct key *); -extern int afs_fs_get_capabilities(struct afs_net *, struct afs_server *, - struct afs_addr_cursor *, struct key *, unsigned int, bool); -extern int afs_fs_inline_bulk_status(struct afs_fs_cursor *, struct afs_net *, - struct afs_fid *, struct afs_file_status *, - struct afs_callback *, unsigned int, - struct afs_volsync *); -extern int afs_fs_fetch_status(struct afs_fs_cursor *, struct afs_net *, - struct afs_fid *, struct afs_file_status *, - struct afs_callback *, struct afs_volsync *); +extern void afs_fs_fetch_status(struct afs_operation *); +extern void afs_fs_fetch_data(struct afs_operation *); +extern void afs_fs_create_file(struct afs_operation *); +extern void afs_fs_make_dir(struct afs_operation *); +extern void afs_fs_remove_file(struct afs_operation *); +extern void afs_fs_remove_dir(struct afs_operation *); +extern void afs_fs_link(struct afs_operation *); +extern void afs_fs_symlink(struct afs_operation *); +extern void afs_fs_rename(struct afs_operation *); +extern void afs_fs_store_data(struct afs_operation *); +extern void afs_fs_setattr(struct afs_operation *); +extern void afs_fs_get_volume_status(struct afs_operation *); +extern void afs_fs_set_lock(struct afs_operation *); +extern void afs_fs_extend_lock(struct afs_operation *); +extern void afs_fs_release_lock(struct afs_operation *); +int afs_fs_give_up_all_callbacks(struct afs_net *net, struct afs_server *server, + struct afs_address *addr, struct key *key); +bool afs_fs_get_capabilities(struct afs_net *net, struct afs_server *server, + struct afs_endpoint_state *estate, unsigned int addr_index, + struct key *key); +extern void afs_fs_inline_bulk_status(struct afs_operation *); + +struct afs_acl { + u32 size; + u8 data[] __counted_by(size); +}; + +extern void afs_fs_fetch_acl(struct afs_operation *); +extern void afs_fs_store_acl(struct afs_operation *); + +/* + * fs_operation.c + */ +extern struct afs_operation *afs_alloc_operation(struct key *, struct afs_volume *); +extern int afs_put_operation(struct afs_operation *); +extern bool afs_begin_vnode_operation(struct afs_operation *); +extern void afs_end_vnode_operation(struct afs_operation *op); +extern void afs_wait_for_operation(struct afs_operation *); +extern int afs_do_sync_operation(struct afs_operation *); + +static inline void afs_op_set_vnode(struct afs_operation *op, unsigned int n, + struct afs_vnode *vnode) +{ + op->file[n].vnode = vnode; + op->file[n].need_io_lock = true; +} + +static inline void afs_op_set_fid(struct afs_operation *op, unsigned int n, + const struct afs_fid *fid) +{ + op->file[n].fid = *fid; +} /* * fs_probe.c */ +struct afs_endpoint_state *afs_get_endpoint_state(struct afs_endpoint_state *estate, + enum afs_estate_trace where); +void afs_put_endpoint_state(struct afs_endpoint_state *estate, enum afs_estate_trace where); extern void afs_fileserver_probe_result(struct afs_call *); -extern int afs_probe_fileservers(struct afs_net *, struct key *, struct afs_server_list *); -extern int afs_wait_for_fs_probes(struct afs_server_list *, unsigned long); +int afs_fs_probe_fileserver(struct afs_net *net, struct afs_server *server, + struct afs_addr_list *new_alist, struct key *key); +int afs_wait_for_fs_probes(struct afs_operation *op, struct afs_server_state *states, bool intr); +extern void afs_probe_fileserver(struct afs_net *, struct afs_server *); +extern void afs_fs_probe_dispatcher(struct work_struct *); +int afs_wait_for_one_fs_probe(struct afs_server *server, struct afs_endpoint_state *estate, + unsigned long exclude, bool is_intr); +extern void afs_fs_probe_cleanup(struct afs_net *); /* * inode.c */ -extern int afs_fetch_status(struct afs_vnode *, struct key *, bool); -extern int afs_iget5_test(struct inode *, void *); -extern struct inode *afs_iget_pseudo_dir(struct super_block *, bool); -extern struct inode *afs_iget(struct super_block *, struct key *, - struct afs_fid *, struct afs_file_status *, - struct afs_callback *, - struct afs_cb_interest *); -extern void afs_zap_data(struct afs_vnode *); -extern int afs_validate(struct afs_vnode *, struct key *); -extern int afs_getattr(const struct path *, struct kstat *, u32, unsigned int); -extern int afs_setattr(struct dentry *, struct iattr *); +extern const struct afs_operation_ops afs_fetch_status_operation; + +void afs_init_new_symlink(struct afs_vnode *vnode, struct afs_operation *op); +const char *afs_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback); +int afs_readlink(struct dentry *dentry, char __user *buffer, int buflen); +extern void afs_vnode_commit_status(struct afs_operation *, struct afs_vnode_param *); +extern int afs_fetch_status(struct afs_vnode *, struct key *, bool, afs_access_t *); +extern int afs_ilookup5_test_by_fid(struct inode *, void *); +extern struct inode *afs_iget(struct afs_operation *, struct afs_vnode_param *); +extern struct inode *afs_root_iget(struct super_block *, struct key *); +extern int afs_getattr(struct mnt_idmap *idmap, const struct path *, + struct kstat *, u32, unsigned int); +extern int afs_setattr(struct mnt_idmap *idmap, struct dentry *, struct iattr *); extern void afs_evict_inode(struct inode *); extern int afs_drop_inode(struct inode *); @@ -1004,7 +1290,7 @@ static inline struct afs_net *afs_i2net(struct inode *inode) static inline struct afs_net *afs_v2net(struct afs_vnode *vnode) { - return afs_i2net(&vnode->vfs_inode); + return afs_i2net(&vnode->netfs.inode); } static inline struct afs_net *afs_sock2net(struct sock *sk) @@ -1025,6 +1311,31 @@ static inline void __afs_stat(atomic_t *s) extern int afs_abort_to_error(u32); extern void afs_prioritise_error(struct afs_error *, int, u32); +static inline void afs_op_nomem(struct afs_operation *op) +{ + op->cumul_error.error = -ENOMEM; +} + +static inline int afs_op_error(const struct afs_operation *op) +{ + return op->cumul_error.error; +} + +static inline s32 afs_op_abort_code(const struct afs_operation *op) +{ + return op->cumul_error.abort_code; +} + +static inline int afs_op_set_error(struct afs_operation *op, int error) +{ + return op->cumul_error.error = error; +} + +static inline void afs_op_accumulate_error(struct afs_operation *op, int error, s32 abort_code) +{ + afs_prioritise_error(&op->cumul_error, error, abort_code); +} + /* * mntpt.c */ @@ -1036,12 +1347,6 @@ extern struct vfsmount *afs_d_automount(struct path *); extern void afs_mntpt_kill_timer(void); /* - * netdevices.c - */ -extern int afs_get_ipv4_interfaces(struct afs_net *, struct afs_interface *, - size_t, bool); - -/* * proc.c */ #ifdef CONFIG_PROC_FS @@ -1061,11 +1366,9 @@ static inline void afs_put_sysnames(struct afs_sysnames *sysnames) {} /* * rotate.c */ -extern bool afs_begin_vnode_operation(struct afs_fs_cursor *, struct afs_vnode *, - struct key *); -extern bool afs_select_fileserver(struct afs_fs_cursor *); -extern bool afs_select_current_fileserver(struct afs_fs_cursor *); -extern int afs_end_vnode_operation(struct afs_fs_cursor *); +void afs_clear_server_states(struct afs_operation *op); +extern bool afs_select_fileserver(struct afs_operation *); +extern void afs_dump_edestaddrreq(const struct afs_operation *); /* * rxrpc.c @@ -1076,7 +1379,10 @@ extern int __net_init afs_open_socket(struct afs_net *); extern void __net_exit afs_close_socket(struct afs_net *); extern void afs_charge_preallocation(struct work_struct *); extern void afs_put_call(struct afs_call *); -extern long afs_make_call(struct afs_addr_cursor *, struct afs_call *, gfp_t, bool); +void afs_deferred_put_call(struct afs_call *call); +void afs_make_call(struct afs_call *call, gfp_t gfp); +void afs_deliver_to_call(struct afs_call *call); +void afs_wait_for_call_to_complete(struct afs_call *call); extern struct afs_call *afs_alloc_flat_call(struct afs_net *, const struct afs_call_type *, size_t, size_t); @@ -1084,32 +1390,74 @@ extern void afs_flat_call_destructor(struct afs_call *); extern void afs_send_empty_reply(struct afs_call *); extern void afs_send_simple_reply(struct afs_call *, const void *, size_t); extern int afs_extract_data(struct afs_call *, bool); -extern int afs_protocol_error(struct afs_call *, int, enum afs_eproto_cause); +extern int afs_protocol_error(struct afs_call *, enum afs_eproto_cause); + +static inline struct afs_call *afs_get_call(struct afs_call *call, + enum afs_call_trace why) +{ + int r; + + __refcount_inc(&call->ref, &r); + + trace_afs_call(call->debug_id, why, r + 1, + atomic_read(&call->net->nr_outstanding_calls), + __builtin_return_address(0)); + return call; +} + +static inline void afs_see_call(struct afs_call *call, enum afs_call_trace why) +{ + int r = refcount_read(&call->ref); + + trace_afs_call(call->debug_id, why, r, + atomic_read(&call->net->nr_outstanding_calls), + __builtin_return_address(0)); +} + +static inline void afs_make_op_call(struct afs_operation *op, struct afs_call *call, + gfp_t gfp) +{ + struct afs_addr_list *alist = op->estate->addresses; + + op->call = call; + op->type = call->type; + call->op = op; + call->key = op->key; + call->intr = !(op->flags & AFS_OPERATION_UNINTR); + call->peer = rxrpc_kernel_get_peer(alist->addrs[op->addr_index].peer); + call->service_id = op->server->service_id; + afs_make_call(call, gfp); +} static inline void afs_extract_begin(struct afs_call *call, void *buf, size_t size) { + call->iov_len = size; call->kvec[0].iov_base = buf; call->kvec[0].iov_len = size; - iov_iter_kvec(&call->iter, READ, call->kvec, 1, size); + iov_iter_kvec(&call->def_iter, ITER_DEST, call->kvec, 1, size); } static inline void afs_extract_to_tmp(struct afs_call *call) { + call->iov_len = sizeof(call->tmp); afs_extract_begin(call, &call->tmp, sizeof(call->tmp)); } static inline void afs_extract_to_tmp64(struct afs_call *call) { + call->iov_len = sizeof(call->tmp64); afs_extract_begin(call, &call->tmp64, sizeof(call->tmp64)); } static inline void afs_extract_discard(struct afs_call *call, size_t size) { - iov_iter_discard(&call->iter, READ, size); + call->iov_len = size; + iov_iter_discard(&call->def_iter, ITER_DEST, size); } static inline void afs_extract_to_buf(struct afs_call *call, size_t size) { + call->iov_len = size; afs_extract_begin(call, call->buffer, size); } @@ -1157,8 +1505,16 @@ static inline void afs_set_call_complete(struct afs_call *call, ok = true; } spin_unlock_bh(&call->state_lock); - if (ok) + if (ok) { trace_afs_call_done(call); + + /* Asynchronous calls have two refs to release - one from the alloc and + * one queued with the work item - and we can't just deallocate the + * call because the work item may be queued again. + */ + if (call->drop_ref) + afs_put_call(call); + } } /* @@ -1166,11 +1522,12 @@ static inline void afs_set_call_complete(struct afs_call *call, */ extern void afs_put_permits(struct afs_permits *); extern void afs_clear_permits(struct afs_vnode *); -extern void afs_cache_permit(struct afs_vnode *, struct key *, unsigned int); -extern void afs_zap_permits(struct rcu_head *); +extern void afs_cache_permit(struct afs_vnode *, struct key *, unsigned int, + struct afs_status_cb *); extern struct key *afs_request_key(struct afs_cell *); +extern struct key *afs_request_key_rcu(struct afs_cell *); extern int afs_check_permit(struct afs_vnode *, struct key *, afs_access_t *); -extern int afs_permission(struct inode *, int); +extern int afs_permission(struct mnt_idmap *, struct inode *, int); extern void __exit afs_clean_up_permit_cache(void); /* @@ -1178,21 +1535,45 @@ extern void __exit afs_clean_up_permit_cache(void); */ extern spinlock_t afs_server_peer_lock; -static inline struct afs_server *afs_get_server(struct afs_server *server) +struct afs_server *afs_find_server(const struct rxrpc_peer *peer); +extern struct afs_server *afs_lookup_server(struct afs_cell *, struct key *, const uuid_t *, u32); +extern struct afs_server *afs_get_server(struct afs_server *, enum afs_server_trace); +struct afs_server *afs_use_server(struct afs_server *server, bool activate, + enum afs_server_trace reason); +void afs_unuse_server(struct afs_net *net, struct afs_server *server, + enum afs_server_trace reason); +void afs_unuse_server_notime(struct afs_net *net, struct afs_server *server, + enum afs_server_trace reason); +extern void afs_put_server(struct afs_net *, struct afs_server *, enum afs_server_trace); +void afs_purge_servers(struct afs_cell *cell); +extern void afs_fs_probe_timer(struct timer_list *); +void __net_exit afs_wait_for_servers(struct afs_net *net); +bool afs_check_server_record(struct afs_operation *op, struct afs_server *server, struct key *key); + +static inline void afs_see_server(struct afs_server *server, enum afs_server_trace trace) { - atomic_inc(&server->usage); - return server; + int r = refcount_read(&server->ref); + int a = atomic_read(&server->active); + + trace_afs_server(server->debug_id, r, a, trace); + } -extern struct afs_server *afs_find_server(struct afs_net *, - const struct sockaddr_rxrpc *); -extern struct afs_server *afs_find_server_by_uuid(struct afs_net *, const uuid_t *); -extern struct afs_server *afs_lookup_server(struct afs_cell *, struct key *, const uuid_t *); -extern void afs_put_server(struct afs_net *, struct afs_server *); -extern void afs_manage_servers(struct work_struct *); -extern void afs_servers_timer(struct timer_list *); -extern void __net_exit afs_purge_servers(struct afs_net *); -extern bool afs_check_server_record(struct afs_fs_cursor *, struct afs_server *); +static inline void afs_inc_servers_outstanding(struct afs_net *net) +{ + atomic_inc(&net->servers_outstanding); +} + +static inline void afs_dec_servers_outstanding(struct afs_net *net) +{ + if (atomic_dec_and_test(&net->servers_outstanding)) + wake_up_var(&net->servers_outstanding); +} + +static inline bool afs_is_probing_server(struct afs_server *server) +{ + return list_empty(&server->probe_link); +} /* * server_list.c @@ -1204,10 +1585,14 @@ static inline struct afs_server_list *afs_get_serverlist(struct afs_server_list } extern void afs_put_serverlist(struct afs_net *, struct afs_server_list *); -extern struct afs_server_list *afs_alloc_server_list(struct afs_cell *, struct key *, - struct afs_vldb_entry *, - u8); +struct afs_server_list *afs_alloc_server_list(struct afs_volume *volume, + struct key *key, + struct afs_vldb_entry *vldb); extern bool afs_annotate_server_list(struct afs_server_list *, struct afs_server_list *); +void afs_attach_volume_to_servers(struct afs_volume *volume, struct afs_server_list *slist); +void afs_reattach_volume_to_servers(struct afs_volume *volume, struct afs_server_list *slist, + struct afs_server_list *old); +void afs_detach_volume_from_servers(struct afs_volume *volume, struct afs_server_list *slist); /* * super.c @@ -1216,14 +1601,31 @@ extern int __init afs_fs_init(void); extern void afs_fs_exit(void); /* + * validation.c + */ +bool afs_check_validity(const struct afs_vnode *vnode); +int afs_update_volume_state(struct afs_operation *op); +int afs_validate(struct afs_vnode *vnode, struct key *key); + +/* * vlclient.c */ extern struct afs_vldb_entry *afs_vl_get_entry_by_name_u(struct afs_vl_cursor *, const char *, int); extern struct afs_addr_list *afs_vl_get_addrs_u(struct afs_vl_cursor *, const uuid_t *); -extern int afs_vl_get_capabilities(struct afs_net *, struct afs_addr_cursor *, struct key *, - struct afs_vlserver *, unsigned int, bool); +struct afs_call *afs_vl_get_capabilities(struct afs_net *net, + struct afs_addr_list *alist, + unsigned int addr_index, + struct key *key, + struct afs_vlserver *server, + unsigned int server_index); extern struct afs_addr_list *afs_yfsvl_get_endpoints(struct afs_vl_cursor *, const uuid_t *); +extern char *afs_yfsvl_get_cell_name(struct afs_vl_cursor *); + +/* + * vl_alias.c + */ +extern int afs_cell_detect_alias(struct afs_cell *, struct key *); /* * vl_probe.c @@ -1246,14 +1648,14 @@ extern int afs_end_vlserver_operation(struct afs_vl_cursor *); */ static inline struct afs_vlserver *afs_get_vlserver(struct afs_vlserver *vlserver) { - atomic_inc(&vlserver->usage); + refcount_inc(&vlserver->ref); return vlserver; } static inline struct afs_vlserver_list *afs_get_vlserverlist(struct afs_vlserver_list *vllist) { if (vllist) - atomic_inc(&vllist->usage); + refcount_inc(&vllist->ref); return vllist; } @@ -1267,103 +1669,116 @@ extern struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *, /* * volume.c */ -static inline struct afs_volume *__afs_get_volume(struct afs_volume *volume) -{ - if (volume) - atomic_inc(&volume->usage); - return volume; -} - -extern struct afs_volume *afs_create_volume(struct afs_mount_params *); -extern void afs_activate_volume(struct afs_volume *); +extern struct afs_volume *afs_create_volume(struct afs_fs_context *); +extern int afs_activate_volume(struct afs_volume *); extern void afs_deactivate_volume(struct afs_volume *); -extern void afs_put_volume(struct afs_cell *, struct afs_volume *); -extern int afs_check_volume_status(struct afs_volume *, struct key *); +bool afs_try_get_volume(struct afs_volume *volume, enum afs_volume_trace reason); +extern struct afs_volume *afs_get_volume(struct afs_volume *, enum afs_volume_trace); +void afs_put_volume(struct afs_volume *volume, enum afs_volume_trace reason); +extern int afs_check_volume_status(struct afs_volume *, struct afs_operation *); /* * write.c */ -extern int afs_set_page_dirty(struct page *); -extern int afs_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata); -extern int afs_write_end(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata); -extern int afs_writepage(struct page *, struct writeback_control *); +void afs_prepare_write(struct netfs_io_subrequest *subreq); +void afs_issue_write(struct netfs_io_subrequest *subreq); +void afs_begin_writeback(struct netfs_io_request *wreq); +void afs_retry_request(struct netfs_io_request *wreq, struct netfs_io_stream *stream); extern int afs_writepages(struct address_space *, struct writeback_control *); -extern void afs_pages_written_back(struct afs_vnode *, struct afs_call *); -extern ssize_t afs_file_write(struct kiocb *, struct iov_iter *); extern int afs_fsync(struct file *, loff_t, loff_t, int); extern vm_fault_t afs_page_mkwrite(struct vm_fault *vmf); extern void afs_prune_wb_keys(struct afs_vnode *); -extern int afs_launder_page(struct page *); /* * xattr.c */ -extern const struct xattr_handler *afs_xattr_handlers[]; -extern ssize_t afs_listxattr(struct dentry *, char *, size_t); +extern const struct xattr_handler * const afs_xattr_handlers[]; /* * yfsclient.c */ -extern int yfs_fs_fetch_file_status(struct afs_fs_cursor *, struct afs_volsync *, bool); -extern int yfs_fs_fetch_data(struct afs_fs_cursor *, struct afs_read *); -extern int yfs_fs_create_file(struct afs_fs_cursor *, const char *, umode_t, u64, - struct afs_fid *, struct afs_file_status *, struct afs_callback *); -extern int yfs_fs_make_dir(struct afs_fs_cursor *, const char *, umode_t, u64, - struct afs_fid *, struct afs_file_status *, struct afs_callback *); -extern int yfs_fs_remove_file2(struct afs_fs_cursor *, struct afs_vnode *, const char *, u64); -extern int yfs_fs_remove(struct afs_fs_cursor *, struct afs_vnode *, const char *, bool, u64); -extern int yfs_fs_link(struct afs_fs_cursor *, struct afs_vnode *, const char *, u64); -extern int yfs_fs_symlink(struct afs_fs_cursor *, const char *, const char *, u64, - struct afs_fid *, struct afs_file_status *); -extern int yfs_fs_rename(struct afs_fs_cursor *, const char *, - struct afs_vnode *, const char *, u64, u64); -extern int yfs_fs_store_data(struct afs_fs_cursor *, struct address_space *, - pgoff_t, pgoff_t, unsigned, unsigned); -extern int yfs_fs_setattr(struct afs_fs_cursor *, struct iattr *); -extern int yfs_fs_get_volume_status(struct afs_fs_cursor *, struct afs_volume_status *); -extern int yfs_fs_set_lock(struct afs_fs_cursor *, afs_lock_type_t); -extern int yfs_fs_extend_lock(struct afs_fs_cursor *); -extern int yfs_fs_release_lock(struct afs_fs_cursor *); -extern int yfs_fs_fetch_status(struct afs_fs_cursor *, struct afs_net *, - struct afs_fid *, struct afs_file_status *, - struct afs_callback *, struct afs_volsync *); -extern int yfs_fs_inline_bulk_status(struct afs_fs_cursor *, struct afs_net *, - struct afs_fid *, struct afs_file_status *, - struct afs_callback *, unsigned int, - struct afs_volsync *); +extern void yfs_fs_fetch_data(struct afs_operation *); +extern void yfs_fs_create_file(struct afs_operation *); +extern void yfs_fs_make_dir(struct afs_operation *); +extern void yfs_fs_remove_file2(struct afs_operation *); +extern void yfs_fs_remove_file(struct afs_operation *); +extern void yfs_fs_remove_dir(struct afs_operation *); +extern void yfs_fs_link(struct afs_operation *); +extern void yfs_fs_symlink(struct afs_operation *); +extern void yfs_fs_rename(struct afs_operation *); +void yfs_fs_rename_replace(struct afs_operation *op); +void yfs_fs_rename_noreplace(struct afs_operation *op); +void yfs_fs_rename_exchange(struct afs_operation *op); +extern void yfs_fs_store_data(struct afs_operation *); +extern void yfs_fs_setattr(struct afs_operation *); +extern void yfs_fs_get_volume_status(struct afs_operation *); +extern void yfs_fs_set_lock(struct afs_operation *); +extern void yfs_fs_extend_lock(struct afs_operation *); +extern void yfs_fs_release_lock(struct afs_operation *); +extern void yfs_fs_fetch_status(struct afs_operation *); +extern void yfs_fs_inline_bulk_status(struct afs_operation *); + +struct yfs_acl { + struct afs_acl *acl; /* Dir/file/symlink ACL */ + struct afs_acl *vol_acl; /* Whole volume ACL */ + u32 inherit_flag; /* True if ACL is inherited from parent dir */ + u32 num_cleaned; /* Number of ACEs removed due to subject removal */ + unsigned int flags; +#define YFS_ACL_WANT_ACL 0x01 /* Set if caller wants ->acl */ +#define YFS_ACL_WANT_VOL_ACL 0x02 /* Set if caller wants ->vol_acl */ +}; + +extern void yfs_free_opaque_acl(struct yfs_acl *); +extern void yfs_fs_fetch_opaque_acl(struct afs_operation *); +extern void yfs_fs_store_opaque_acl2(struct afs_operation *); /* * Miscellaneous inline functions. */ static inline struct afs_vnode *AFS_FS_I(struct inode *inode) { - return container_of(inode, struct afs_vnode, vfs_inode); + return container_of(inode, struct afs_vnode, netfs.inode); } static inline struct inode *AFS_VNODE_TO_I(struct afs_vnode *vnode) { - return &vnode->vfs_inode; + return &vnode->netfs.inode; } -static inline void afs_vnode_commit_status(struct afs_fs_cursor *fc, - struct afs_vnode *vnode, - unsigned int cb_break) +/* + * Note that a dentry got changed. We need to set d_fsdata to the data version + * number derived from the result of the operation. It doesn't matter if + * d_fsdata goes backwards as we'll just revalidate. + */ +static inline void afs_update_dentry_version(struct afs_operation *op, + struct afs_vnode_param *dir_vp, + struct dentry *dentry) { - if (fc->ac.error == 0) - afs_cache_permit(vnode, fc->key, cb_break); + if (!op->cumul_error.error) + dentry->d_fsdata = + (void *)(unsigned long)dir_vp->scb.status.data_version; } -static inline void afs_check_for_remote_deletion(struct afs_fs_cursor *fc, - struct afs_vnode *vnode) +/* + * Set the file size and block count. Estimate the number of 512 bytes blocks + * used, rounded up to nearest 1K for consistency with other AFS clients. + */ +static inline void afs_set_i_size(struct afs_vnode *vnode, u64 size) { - if (fc->ac.error == -ENOENT) { - set_bit(AFS_VNODE_DELETED, &vnode->flags); - afs_break_callback(vnode); - } + i_size_write(&vnode->netfs.inode, size); + vnode->netfs.inode.i_blocks = ((size + 1023) >> 10) << 1; +} + +/* + * Check for a conflicting operation on a directory that we just unlinked from. + * If someone managed to sneak a link or an unlink in on the file we just + * unlinked, we won't be able to trust nlink on an AFS file (but not YFS). + */ +static inline void afs_check_dir_conflict(struct afs_operation *op, + struct afs_vnode_param *dvp) +{ + if (dvp->dv_before + dvp->dv_delta != dvp->scb.status.data_version) + op->flags |= AFS_OPERATION_DIR_CONFLICT; } static inline int afs_io_error(struct afs_call *call, enum afs_io_error where) @@ -1378,6 +1793,38 @@ static inline int afs_bad(struct afs_vnode *vnode, enum afs_file_error where) return -EIO; } +/* + * Set the callback promise on a vnode. + */ +static inline void afs_set_cb_promise(struct afs_vnode *vnode, time64_t expires_at, + enum afs_cb_promise_trace trace) +{ + atomic64_set(&vnode->cb_expires_at, expires_at); + trace_afs_cb_promise(vnode, trace); +} + +/* + * Clear the callback promise on a vnode, returning true if it was promised. + */ +static inline bool afs_clear_cb_promise(struct afs_vnode *vnode, + enum afs_cb_promise_trace trace) +{ + trace_afs_cb_promise(vnode, trace); + return atomic64_xchg(&vnode->cb_expires_at, AFS_NO_CB_PROMISE) != AFS_NO_CB_PROMISE; +} + +/* + * Mark a directory as being invalid. + */ +static inline void afs_invalidate_dir(struct afs_vnode *dvnode, + enum afs_dir_invalid_trace trace) +{ + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { + trace_afs_dir_invalid(dvnode, trace); + afs_stat_v(dvnode, n_inval); + } +} + /*****************************************************************************/ /* * debug tracing diff --git a/fs/afs/main.c b/fs/afs/main.c index 107427688edd..e6bb8237db98 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS client file system * * Copyright (C) 2002,5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/module.h> @@ -45,8 +41,6 @@ const char afs_init_sysname[] = "arm_linux26"; const char afs_init_sysname[] = "aarch64_linux26"; #elif defined(CONFIG_X86_32) const char afs_init_sysname[] = "i386_linux26"; -#elif defined(CONFIG_IA64) -const char afs_init_sysname[] = "ia64_linux26"; #elif defined(CONFIG_PPC64) const char afs_init_sysname[] = "ppc64_linux26"; #elif defined(CONFIG_PPC32) @@ -79,27 +73,24 @@ static int __net_init afs_net_init(struct net *net_ns) generate_random_uuid((unsigned char *)&net->uuid); INIT_WORK(&net->charge_preallocation_work, afs_charge_preallocation); + INIT_WORK(&net->rx_oob_work, afs_process_oob_queue); mutex_init(&net->socket_mutex); net->cells = RB_ROOT; - seqlock_init(&net->cells_lock); - INIT_WORK(&net->cells_manager, afs_manage_cells); - timer_setup(&net->cells_timer, afs_cells_timer, 0); - + idr_init(&net->cells_dyn_ino); + init_rwsem(&net->cells_lock); + mutex_init(&net->cells_alias_lock); mutex_init(&net->proc_cells_lock); INIT_HLIST_HEAD(&net->proc_cells); seqlock_init(&net->fs_lock); - net->fs_servers = RB_ROOT; - INIT_LIST_HEAD(&net->fs_updates); + INIT_LIST_HEAD(&net->fs_probe_fast); + INIT_LIST_HEAD(&net->fs_probe_slow); INIT_HLIST_HEAD(&net->fs_proc); - INIT_HLIST_HEAD(&net->fs_addresses4); - INIT_HLIST_HEAD(&net->fs_addresses6); - seqlock_init(&net->fs_addr_lock); - - INIT_WORK(&net->fs_manager, afs_manage_servers); - timer_setup(&net->fs_timer, afs_servers_timer, 0); + INIT_WORK(&net->fs_prober, afs_fs_probe_dispatcher); + timer_setup(&net->fs_probe_timer, afs_fs_probe_timer, 0); + atomic_set(&net->servers_outstanding, 1); ret = -ENOMEM; sysnames = kzalloc(sizeof(*sysnames), GFP_KERNEL); @@ -130,14 +121,16 @@ static int __net_init afs_net_init(struct net *net_ns) error_open_socket: net->live = false; + afs_fs_probe_cleanup(net); afs_cell_purge(net); - afs_purge_servers(net); + afs_wait_for_servers(net); error_cell_init: net->live = false; afs_proc_cleanup(net); error_proc: afs_put_sysnames(net->sysnames); error_sysnames: + idr_destroy(&net->cells_dyn_ino); net->live = false; return ret; } @@ -150,11 +143,14 @@ static void __net_exit afs_net_exit(struct net *net_ns) struct afs_net *net = afs_net(net_ns); net->live = false; + afs_fs_probe_cleanup(net); afs_cell_purge(net); - afs_purge_servers(net); + afs_wait_for_servers(net); afs_close_socket(net); afs_proc_cleanup(net); afs_put_sysnames(net->sysnames); + idr_destroy(&net->cells_dyn_ino); + kfree_rcu(rcu_access_pointer(net->address_prefs), rcu); } static struct pernet_operations afs_net_ops = { @@ -173,24 +169,17 @@ static int __init afs_init(void) printk(KERN_INFO "kAFS: Red Hat AFS client v0.1 registering.\n"); - afs_wq = alloc_workqueue("afs", 0, 0); + afs_wq = alloc_workqueue("afs", WQ_PERCPU, 0); if (!afs_wq) goto error_afs_wq; - afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM, 0); + afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM | WQ_UNBOUND, 0); if (!afs_async_calls) goto error_async; - afs_lock_manager = alloc_workqueue("kafs_lockd", WQ_MEM_RECLAIM, 0); + afs_lock_manager = alloc_workqueue("kafs_lockd", WQ_MEM_RECLAIM | WQ_PERCPU, 0); if (!afs_lock_manager) goto error_lockmgr; -#ifdef CONFIG_AFS_FSCACHE - /* we want to be able to cache */ - ret = fscache_register_netfs(&afs_cache_netfs); - if (ret < 0) - goto error_cache; -#endif - - ret = register_pernet_subsys(&afs_net_ops); + ret = register_pernet_device(&afs_net_ops); if (ret < 0) goto error_net; @@ -200,8 +189,8 @@ static int __init afs_init(void) goto error_fs; afs_proc_symlink = proc_symlink("fs/afs", NULL, "../self/net/afs"); - if (IS_ERR(afs_proc_symlink)) { - ret = PTR_ERR(afs_proc_symlink); + if (!afs_proc_symlink) { + ret = -ENOMEM; goto error_proc; } @@ -210,12 +199,8 @@ static int __init afs_init(void) error_proc: afs_fs_exit(); error_fs: - unregister_pernet_subsys(&afs_net_ops); + unregister_pernet_device(&afs_net_ops); error_net: -#ifdef CONFIG_AFS_FSCACHE - fscache_unregister_netfs(&afs_cache_netfs); -error_cache: -#endif destroy_workqueue(afs_lock_manager); error_lockmgr: destroy_workqueue(afs_async_calls); @@ -241,10 +226,7 @@ static void __exit afs_exit(void) proc_remove(afs_proc_symlink); afs_fs_exit(); - unregister_pernet_subsys(&afs_net_ops); -#ifdef CONFIG_AFS_FSCACHE - fscache_unregister_netfs(&afs_cache_netfs); -#endif + unregister_pernet_device(&afs_net_ops); destroy_workqueue(afs_lock_manager); destroy_workqueue(afs_async_calls); destroy_workqueue(afs_wq); diff --git a/fs/afs/misc.c b/fs/afs/misc.c index bbb1fd51b019..c8a7f266080d 100644 --- a/fs/afs/misc.c +++ b/fs/afs/misc.c @@ -1,19 +1,17 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* miscellaneous bits * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> #include <linux/module.h> #include <linux/errno.h> +#include <crypto/krb5.h> #include "internal.h" #include "afs_fs.h" +#include "protocol_uae.h" /* * convert an AFS abort code to a Linux error number @@ -69,34 +67,27 @@ int afs_abort_to_error(u32 abort_code) case AFSVL_PERM: return -EACCES; case AFSVL_NOMEM: return -EREMOTEIO; - /* Unified AFS error table; ET "uae" == 0x2f6df00 */ - case 0x2f6df00: return -EPERM; - case 0x2f6df01: return -ENOENT; - case 0x2f6df04: return -EIO; - case 0x2f6df0a: return -EAGAIN; - case 0x2f6df0b: return -ENOMEM; - case 0x2f6df0c: return -EACCES; - case 0x2f6df0f: return -EBUSY; - case 0x2f6df10: return -EEXIST; - case 0x2f6df11: return -EXDEV; - case 0x2f6df12: return -ENODEV; - case 0x2f6df13: return -ENOTDIR; - case 0x2f6df14: return -EISDIR; - case 0x2f6df15: return -EINVAL; - case 0x2f6df1a: return -EFBIG; - case 0x2f6df1b: return -ENOSPC; - case 0x2f6df1d: return -EROFS; - case 0x2f6df1e: return -EMLINK; - case 0x2f6df20: return -EDOM; - case 0x2f6df21: return -ERANGE; - case 0x2f6df22: return -EDEADLK; - case 0x2f6df23: return -ENAMETOOLONG; - case 0x2f6df24: return -ENOLCK; - case 0x2f6df26: return -ENOTEMPTY; - case 0x2f6df28: return -EWOULDBLOCK; - case 0x2f6df69: return -ENOTCONN; - case 0x2f6df6c: return -ETIMEDOUT; - case 0x2f6df78: return -EDQUOT; + /* Unified AFS error table */ + case UAEPERM: return -EPERM; + case UAENOENT: return -ENOENT; + case UAEAGAIN: return -EAGAIN; + case UAEACCES: return -EACCES; + case UAEBUSY: return -EBUSY; + case UAEEXIST: return -EEXIST; + case UAENOTDIR: return -ENOTDIR; + case UAEISDIR: return -EISDIR; + case UAEFBIG: return -EFBIG; + case UAENOSPC: return -ENOSPC; + case UAEROFS: return -EROFS; + case UAEMLINK: return -EMLINK; + case UAEDEADLK: return -EDEADLK; + case UAENAMETOOLONG: return -ENAMETOOLONG; + case UAENOLCK: return -ENOLCK; + case UAENOTEMPTY: return -ENOTEMPTY; + case UAELOOP: return -ELOOP; + case UAEOVERFLOW: return -EOVERFLOW; + case UAENOMEDIUM: return -ENOMEDIUM; + case UAEDQUOT: return -EDQUOT; /* RXKAD abort codes; from include/rxrpc/packet.h. ET "RXK" == 0x1260B00 */ case RXKADINCONSISTENCY: return -EPROTO; @@ -113,7 +104,34 @@ int afs_abort_to_error(u32 abort_code) case RXKADDATALEN: return -EKEYREJECTED; case RXKADILLEGALLEVEL: return -EKEYREJECTED; + case RXGK_INCONSISTENCY: return -EPROTO; + case RXGK_PACKETSHORT: return -EPROTO; + case RXGK_BADCHALLENGE: return -EPROTO; + case RXGK_SEALEDINCON: return -EKEYREJECTED; + case RXGK_NOTAUTH: return -EKEYREJECTED; + case RXGK_EXPIRED: return -EKEYEXPIRED; + case RXGK_BADLEVEL: return -EKEYREJECTED; + case RXGK_BADKEYNO: return -EKEYREJECTED; + case RXGK_NOTRXGK: return -EKEYREJECTED; + case RXGK_UNSUPPORTED: return -EKEYREJECTED; + case RXGK_GSSERROR: return -EKEYREJECTED; +#ifdef RXGK_BADETYPE + case RXGK_BADETYPE: return -ENOPKG; +#endif +#ifdef RXGK_BADTOKEN + case RXGK_BADTOKEN: return -EKEYREJECTED; +#endif +#ifdef RXGK_BADETYPE + case RXGK_DATALEN: return -EPROTO; +#endif +#ifdef RXGK_BADQOP + case RXGK_BADQOP: return -EKEYREJECTED; +#endif + + case KRB5_PROG_KEYTYPE_NOSUPP: return -ENOPKG; + case RXGEN_OPCODE: return -ENOTSUPP; + case RX_INVALID_OPERATION: return -ENOTSUPP; default: return -EREMOTEIO; } @@ -126,47 +144,65 @@ void afs_prioritise_error(struct afs_error *e, int error, u32 abort_code) { switch (error) { case 0: + e->aborted = false; + e->error = 0; return; default: if (e->error == -ETIMEDOUT || e->error == -ETIME) return; + fallthrough; case -ETIMEDOUT: case -ETIME: if (e->error == -ENOMEM || e->error == -ENONET) return; + fallthrough; case -ENOMEM: case -ENONET: if (e->error == -ERFKILL) return; + fallthrough; case -ERFKILL: if (e->error == -EADDRNOTAVAIL) return; + fallthrough; case -EADDRNOTAVAIL: if (e->error == -ENETUNREACH) return; + fallthrough; case -ENETUNREACH: if (e->error == -EHOSTUNREACH) return; + fallthrough; case -EHOSTUNREACH: if (e->error == -EHOSTDOWN) return; + fallthrough; case -EHOSTDOWN: if (e->error == -ECONNREFUSED) return; + fallthrough; case -ECONNREFUSED: if (e->error == -ECONNRESET) return; + fallthrough; case -ECONNRESET: /* Responded, but call expired. */ if (e->responded) return; e->error = error; + e->aborted = false; return; case -ECONNABORTED: - e->responded = true; e->error = afs_abort_to_error(abort_code); + e->aborted = true; + e->responded = true; + return; + case -ENETRESET: /* Responded, but we seem to have changed address */ + e->aborted = false; + e->responded = true; + e->error = error; return; } } diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 2e51c6994148..57c204a3c04e 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* mountpoint management * * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -17,6 +13,7 @@ #include <linux/mount.h> #include <linux/namei.h> #include <linux/gfp.h> +#include <linux/fs_context.h> #include "internal.h" @@ -33,9 +30,8 @@ const struct file_operations afs_mntpt_file_operations = { const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, - .readlink = page_readlink, + .readlink = afs_readlink, .getattr = afs_getattr, - .listxattr = afs_listxattr, }; const struct inode_operations afs_autocell_inode_operations = { @@ -47,6 +43,8 @@ static DECLARE_DELAYED_WORK(afs_mntpt_expiry_timer, afs_mntpt_expiry_timed_out); static unsigned long afs_mntpt_expiry_timeout = 10 * 60; +static const char afs_root_volume[] = "root.cell"; + /* * no valid lookup procedure on this sort of dir */ @@ -68,108 +66,116 @@ static int afs_mntpt_open(struct inode *inode, struct file *file) } /* - * create a vfsmount to be automounted + * Set the parameters for the proposed superblock. */ -static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt) +static int afs_mntpt_set_params(struct fs_context *fc, struct dentry *mntpt) { - struct afs_super_info *as; - struct vfsmount *mnt; - struct afs_vnode *vnode; - struct page *page; - char *devname, *options; - bool rwpath = false; + struct afs_fs_context *ctx = fc->fs_private; + struct afs_super_info *src_as = AFS_FS_S(mntpt->d_sb); + struct afs_vnode *vnode = AFS_FS_I(d_inode(mntpt)); + struct afs_cell *cell; + const char *p; int ret; - _enter("{%pd}", mntpt); - - BUG_ON(!d_inode(mntpt)); - - ret = -ENOMEM; - devname = (char *) get_zeroed_page(GFP_KERNEL); - if (!devname) - goto error_no_devname; - - options = (char *) get_zeroed_page(GFP_KERNEL); - if (!options) - goto error_no_options; + if (fc->net_ns != src_as->net_ns) { + put_net(fc->net_ns); + fc->net_ns = get_net(src_as->net_ns); + } - vnode = AFS_FS_I(d_inode(mntpt)); + if (src_as->volume && src_as->volume->type == AFSVL_RWVOL) { + ctx->type = AFSVL_RWVOL; + ctx->force = true; + } + if (ctx->cell) { + afs_unuse_cell(ctx->cell, afs_cell_trace_unuse_mntpt); + ctx->cell = NULL; + } if (test_bit(AFS_VNODE_PSEUDODIR, &vnode->flags)) { /* if the directory is a pseudo directory, use the d_name */ - static const char afs_root_cell[] = ":root.cell."; unsigned size = mntpt->d_name.len; - ret = -ENOENT; - if (size < 2 || size > AFS_MAXCELLNAME) - goto error_no_page; + if (size < 2) + return -ENOENT; + p = mntpt->d_name.name; if (mntpt->d_name.name[0] == '.') { - devname[0] = '%'; - memcpy(devname + 1, mntpt->d_name.name + 1, size - 1); - memcpy(devname + size, afs_root_cell, - sizeof(afs_root_cell)); - rwpath = true; - } else { - devname[0] = '#'; - memcpy(devname + 1, mntpt->d_name.name, size); - memcpy(devname + size + 1, afs_root_cell, - sizeof(afs_root_cell)); + size--; + p++; + ctx->type = AFSVL_RWVOL; + ctx->force = true; } + if (size > AFS_MAXCELLNAME) + return -ENAMETOOLONG; + + cell = afs_lookup_cell(ctx->net, p, size, NULL, + AFS_LOOKUP_CELL_MOUNTPOINT, + afs_cell_trace_use_lookup_mntpt); + if (IS_ERR(cell)) { + pr_err("kAFS: unable to lookup cell '%pd'\n", mntpt); + return PTR_ERR(cell); + } + ctx->cell = cell; + + ctx->volname = afs_root_volume; + ctx->volnamesz = sizeof(afs_root_volume) - 1; } else { /* read the contents of the AFS special symlink */ + DEFINE_DELAYED_CALL(cleanup); + const char *content; loff_t size = i_size_read(d_inode(mntpt)); - char *buf; - ret = -EINVAL; - if (size > PAGE_SIZE - 1) - goto error_no_page; + if (src_as->cell) + ctx->cell = afs_use_cell(src_as->cell, afs_cell_trace_use_mntpt); - page = read_mapping_page(d_inode(mntpt)->i_mapping, 0, NULL); - if (IS_ERR(page)) { - ret = PTR_ERR(page); - goto error_no_page; - } + if (size < 2 || size > PAGE_SIZE - 1) + return -EINVAL; - if (PageError(page)) { - ret = afs_bad(AFS_FS_I(d_inode(mntpt)), afs_file_error_mntpt); - goto error; + content = afs_get_link(mntpt, d_inode(mntpt), &cleanup); + if (IS_ERR(content)) { + do_delayed_call(&cleanup); + return PTR_ERR(content); } - buf = kmap_atomic(page); - memcpy(devname, buf, size); - kunmap_atomic(buf); - put_page(page); - page = NULL; + ret = -EINVAL; + if (content[size - 1] == '.') + ret = vfs_parse_fs_qstr(fc, "source", + &QSTR_LEN(content, size - 1)); + do_delayed_call(&cleanup); + if (ret < 0) + return ret; + + /* Don't cross a backup volume mountpoint from a backup volume */ + if (src_as->volume && src_as->volume->type == AFSVL_BACKVOL && + ctx->type == AFSVL_BACKVOL) + return -ENODEV; } - /* work out what options we want */ - as = AFS_FS_S(mntpt->d_sb); - if (as->cell) { - memcpy(options, "cell=", 5); - strcpy(options + 5, as->cell->name); - if ((as->volume && as->volume->type == AFSVL_RWVOL) || rwpath) - strcat(options, ",rwpath"); - } + return 0; +} - /* try and do the mount */ - _debug("--- attempting mount %s -o %s ---", devname, options); - mnt = vfs_submount(mntpt, &afs_fs_type, devname, options); - _debug("--- mount result %p ---", mnt); +/* + * create a vfsmount to be automounted + */ +static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt) +{ + struct fs_context *fc; + struct vfsmount *mnt; + int ret; - free_page((unsigned long) devname); - free_page((unsigned long) options); - _leave(" = %p", mnt); - return mnt; + BUG_ON(!d_inode(mntpt)); -error: - put_page(page); -error_no_page: - free_page((unsigned long) options); -error_no_options: - free_page((unsigned long) devname); -error_no_devname: - _leave(" = %d", ret); - return ERR_PTR(ret); + fc = fs_context_for_submount(&afs_fs_type, mntpt); + if (IS_ERR(fc)) + return ERR_CAST(fc); + + ret = afs_mntpt_set_params(fc, mntpt); + if (!ret) + mnt = fc_mount(fc); + else + mnt = ERR_PTR(ret); + + put_fs_context(fc); + return mnt; } /* @@ -185,7 +191,6 @@ struct vfsmount *afs_d_automount(struct path *path) if (IS_ERR(newmnt)) return newmnt; - mntget(newmnt); /* prevent immediate expiration */ mnt_set_expiry(newmnt, &afs_vfsmounts); queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, afs_mntpt_expiry_timeout * HZ); diff --git a/fs/afs/netdevices.c b/fs/afs/netdevices.c deleted file mode 100644 index 2a009d1939d7..000000000000 --- a/fs/afs/netdevices.c +++ /dev/null @@ -1,48 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* AFS network device helpers - * - * Copyright (c) 2007 Patrick McHardy <kaber@trash.net> - */ - -#include <linux/string.h> -#include <linux/rtnetlink.h> -#include <linux/inetdevice.h> -#include <linux/netdevice.h> -#include <linux/if_arp.h> -#include <net/net_namespace.h> -#include "internal.h" - -/* - * get a list of this system's interface IPv4 addresses, netmasks and MTUs - * - maxbufs must be at least 1 - * - returns the number of interface records in the buffer - */ -int afs_get_ipv4_interfaces(struct afs_net *net, struct afs_interface *bufs, - size_t maxbufs, bool wantloopback) -{ - struct net_device *dev; - struct in_device *idev; - int n = 0; - - ASSERT(maxbufs > 0); - - rtnl_lock(); - for_each_netdev(net->net, dev) { - if (dev->type == ARPHRD_LOOPBACK && !wantloopback) - continue; - idev = __in_dev_get_rtnl(dev); - if (!idev) - continue; - for_primary_ifa(idev) { - bufs[n].address.s_addr = ifa->ifa_address; - bufs[n].netmask.s_addr = ifa->ifa_mask; - bufs[n].mtu = dev->mtu; - n++; - if (n >= maxbufs) - goto out; - } endfor_ifa(idev); - } -out: - rtnl_unlock(); - return n; -} diff --git a/fs/afs/proc.c b/fs/afs/proc.c index be2ee3bbd0a9..44520549b509 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* /proc interface for AFS * * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/slab.h> @@ -42,7 +38,7 @@ static int afs_proc_cells_show(struct seq_file *m, void *v) if (v == SEQ_START_TOKEN) { /* display header on line 1 */ - seq_puts(m, "USE TTL SV NAME\n"); + seq_puts(m, "USE ACT TTL SV ST NAME\n"); return 0; } @@ -50,10 +46,12 @@ static int afs_proc_cells_show(struct seq_file *m, void *v) vllist = rcu_dereference(cell->vl_servers); /* display one cell per line on subsequent lines */ - seq_printf(m, "%3u %6lld %2u %s\n", - atomic_read(&cell->usage), + seq_printf(m, "%3u %3u %6lld %2u %2u %s\n", + refcount_read(&cell->ref), + atomic_read(&cell->active), cell->dns_expiry - ktime_get_real_seconds(), vllist ? vllist->nr_servers : 0, + cell->state, cell->name); return 0; } @@ -124,14 +122,16 @@ static int afs_proc_cells_write(struct file *file, char *buf, size_t size) if (strcmp(buf, "add") == 0) { struct afs_cell *cell; - cell = afs_lookup_cell(net, name, strlen(name), args, true); + cell = afs_lookup_cell(net, name, strlen(name), args, + AFS_LOOKUP_CELL_PRELOAD, + afs_cell_trace_use_lookup_add); if (IS_ERR(cell)) { ret = PTR_ERR(cell); goto done; } if (test_and_set_bit(AFS_CELL_FL_NO_GC, &cell->flags)) - afs_put_cell(net, cell); + afs_unuse_cell(cell, afs_cell_trace_unuse_no_pin); } else { goto inval; } @@ -149,6 +149,56 @@ inval: } /* + * Display the list of addr_prefs known to the namespace. + */ +static int afs_proc_addr_prefs_show(struct seq_file *m, void *v) +{ + struct afs_addr_preference_list *preflist; + struct afs_addr_preference *pref; + struct afs_net *net = afs_seq2net_single(m); + union { + struct sockaddr_in sin; + struct sockaddr_in6 sin6; + } addr; + unsigned int i; + char buf[44]; /* Maximum ipv6 + max subnet is 43 */ + + rcu_read_lock(); + preflist = rcu_dereference(net->address_prefs); + + if (!preflist) { + seq_puts(m, "NO PREFS\n"); + goto out; + } + + seq_printf(m, "PROT SUBNET PRIOR (v=%u n=%u/%u/%u)\n", + preflist->version, preflist->ipv6_off, preflist->nr, preflist->max_prefs); + + memset(&addr, 0, sizeof(addr)); + + for (i = 0; i < preflist->nr; i++) { + pref = &preflist->prefs[i]; + + addr.sin.sin_family = pref->family; + if (pref->family == AF_INET) { + memcpy(&addr.sin.sin_addr, &pref->ipv4_addr, + sizeof(addr.sin.sin_addr)); + snprintf(buf, sizeof(buf), "%pISc/%u", &addr.sin, pref->subnet_mask); + seq_printf(m, "UDP %-43.43s %5u\n", buf, pref->prio); + } else { + memcpy(&addr.sin6.sin6_addr, &pref->ipv6_addr, + sizeof(addr.sin6.sin6_addr)); + snprintf(buf, sizeof(buf), "%pISc/%u", &addr.sin6, pref->subnet_mask); + seq_printf(m, "UDP %-43.43s %5u\n", buf, pref->prio); + } + } + +out: + rcu_read_unlock(); + return 0; +} + +/* * Display the name of the current workstation cell. */ static int afs_proc_rootcell_show(struct seq_file *m, void *v) @@ -157,13 +207,11 @@ static int afs_proc_rootcell_show(struct seq_file *m, void *v) struct afs_net *net; net = afs_seq2net_single(m); - if (rcu_access_pointer(net->ws_cell)) { - rcu_read_lock(); - cell = rcu_dereference(net->ws_cell); - if (cell) - seq_printf(m, "%s\n", cell->name); - rcu_read_unlock(); - } + down_read(&net->cells_lock); + cell = rcu_dereference_protected(net->ws_cell, lockdep_is_held(&net->cells_lock)); + if (cell) + seq_printf(m, "%s\n", cell->name); + up_read(&net->cells_lock); return 0; } @@ -194,7 +242,13 @@ static int afs_proc_rootcell_write(struct file *file, char *buf, size_t size) /* determine command to perform */ _debug("rootcell=%s", buf); - ret = afs_cell_init(net, buf); + ret = -EEXIST; + inode_lock(file_inode(file)); + if (!rcu_access_pointer(net->ws_cell)) + ret = afs_cell_init(net, buf); + else + printk("busy\n"); + inode_unlock(file_inode(file)); out: _leave(" = %d", ret); @@ -212,18 +266,18 @@ static const char afs_vol_types[3][3] = { */ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); - struct afs_volume *vol = list_entry(v, struct afs_volume, proc_link); + struct afs_volume *vol = hlist_entry(v, struct afs_volume, proc_link); /* Display header on line 1 */ - if (v == &cell->proc_volumes) { - seq_puts(m, "USE VID TY\n"); + if (v == SEQ_START_TOKEN) { + seq_puts(m, "USE VID TY NAME\n"); return 0; } - seq_printf(m, "%3d %08llx %s\n", - atomic_read(&vol->usage), vol->vid, - afs_vol_types[vol->type]); + seq_printf(m, "%3d %08llx %s %s\n", + refcount_read(&vol->ref), vol->vid, + afs_vol_types[vol->type], + vol->name); return 0; } @@ -231,26 +285,24 @@ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) __acquires(cell->proc_lock) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); - read_lock(&cell->proc_lock); - return seq_list_start_head(&cell->proc_volumes, *_pos); + rcu_read_lock(); + return seq_hlist_start_head_rcu(&cell->proc_volumes, *_pos); } static void *afs_proc_cell_volumes_next(struct seq_file *m, void *v, loff_t *_pos) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); - return seq_list_next(v, &cell->proc_volumes, _pos); + return seq_hlist_next_rcu(v, &cell->proc_volumes, _pos); } static void afs_proc_cell_volumes_stop(struct seq_file *m, void *v) __releases(cell->proc_lock) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); - - read_unlock(&cell->proc_lock); + rcu_read_unlock(); } static const struct seq_operations afs_proc_cell_volumes_ops = { @@ -296,8 +348,8 @@ static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v) if (v == SEQ_START_TOKEN) { seq_printf(m, "# source %s, status %s\n", - dns_record_sources[vllist->source], - dns_lookup_statuses[vllist->status]); + dns_record_sources[vllist ? vllist->source : 0], + dns_lookup_statuses[vllist ? vllist->status : 0]); return 0; } @@ -313,8 +365,13 @@ static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v) for (i = 0; i < alist->nr_addrs; i++) seq_printf(m, " %c %pISpc\n", alist->preferred == i ? '>' : '-', - &alist->addrs[i].transport); + rxrpc_kernel_remote_addr(alist->addrs[i].peer)); } + seq_printf(m, " info: fl=%lx rtt=%d\n", vlserver->flags, vlserver->rtt); + seq_printf(m, " probe: fl=%x e=%d ac=%d out=%d\n", + vlserver->probe.flags, vlserver->probe.error, + vlserver->probe.abort_code, + atomic_read(&vlserver->probe_outstanding)); return 0; } @@ -323,7 +380,7 @@ static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos) { struct afs_vl_seq_net_private *priv = m->private; struct afs_vlserver_list *vllist; - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); loff_t pos = *_pos; rcu_read_lock(); @@ -336,7 +393,7 @@ static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos) if (pos == 0) return SEQ_START_TOKEN; - if (!vllist || pos - 1 >= vllist->nr_servers) + if (pos - 1 >= vllist->nr_servers) return NULL; return &vllist->servers[pos - 1]; @@ -376,26 +433,51 @@ static const struct seq_operations afs_proc_cell_vlservers_ops = { */ static int afs_proc_servers_show(struct seq_file *m, void *v) { - struct afs_server *server; + struct afs_endpoint_state *estate; struct afs_addr_list *alist; + struct afs_server *server; + unsigned long failed; int i; if (v == SEQ_START_TOKEN) { - seq_puts(m, "UUID USE ADDR\n"); + seq_puts(m, "UUID REF ACT CELL\n"); return 0; } server = list_entry(v, struct afs_server, proc_link); - alist = rcu_dereference(server->addresses); - seq_printf(m, "%pU %3d %pISpc%s\n", + seq_printf(m, "%pU %3d %3d %s\n", &server->uuid, - atomic_read(&server->usage), - &alist->addrs[0].transport, - alist->preferred == 0 ? "*" : ""); - for (i = 1; i < alist->nr_addrs; i++) - seq_printf(m, " %pISpc%s\n", - &alist->addrs[i].transport, - alist->preferred == i ? "*" : ""); + refcount_read(&server->ref), + atomic_read(&server->active), + server->cell->name); + seq_printf(m, " - info: fl=%lx rtt=%u\n", + server->flags, server->rtt); + seq_printf(m, " - probe: last=%d\n", + (int)(jiffies - server->probed_at) / HZ); + + estate = rcu_dereference(server->endpoint_state); + if (!estate) + goto out; + failed = estate->failed_set; + seq_printf(m, " - ESTATE pq=%x np=%u rsp=%lx f=%lx\n", + estate->probe_seq, atomic_read(&estate->nr_probing), + estate->responsive_set, estate->failed_set); + + alist = estate->addresses; + seq_printf(m, " - ALIST v=%u ap=%u\n", + alist->version, alist->addr_pref_version); + for (i = 0; i < alist->nr_addrs; i++) { + const struct afs_address *addr = &alist->addrs[i]; + + seq_printf(m, " [%x] %pISpc%s rtt=%d err=%d p=%u\n", + i, rxrpc_kernel_remote_addr(addr->peer), + alist->preferred == i ? "*" : + test_bit(i, &failed) ? "!" : "", + rxrpc_kernel_get_srtt(addr->peer), + addr->last_error, addr->prio); + } + +out: return 0; } @@ -566,6 +648,7 @@ void afs_put_sysnames(struct afs_sysnames *sysnames) if (sysnames->subs[i] != afs_init_sysname && sysnames->subs[i] != sysnames->blank) kfree(sysnames->subs[i]); + kfree(sysnames); } } @@ -675,7 +758,11 @@ int afs_proc_init(struct afs_net *net) &afs_proc_sysname_ops, afs_proc_sysname_write, sizeof(struct seq_net_private), - NULL)) + NULL) || + !proc_create_net_single_write("addr_prefs", 0644, p, + afs_proc_addr_prefs_show, + afs_proc_addr_prefs_write, + NULL)) goto error_tree; net->proc_afs = p; diff --git a/fs/afs/protocol_afs.h b/fs/afs/protocol_afs.h new file mode 100644 index 000000000000..0c39358c8b70 --- /dev/null +++ b/fs/afs/protocol_afs.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* AFS protocol bits + * + * Copyright (C) 2021 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + + +#define AFSCAPABILITIESMAX 196 /* Maximum number of words in a capability set */ + +/* AFS3 Fileserver capabilities word 0 */ +#define AFS3_VICED_CAPABILITY_ERRORTRANS 0x0001 /* Uses UAE errors */ +#define AFS3_VICED_CAPABILITY_64BITFILES 0x0002 /* FetchData64 & StoreData64 supported */ +#define AFS3_VICED_CAPABILITY_WRITELOCKACL 0x0004 /* Can lock a file even without lock perm */ +#define AFS3_VICED_CAPABILITY_SANEACLS 0x0008 /* ACLs reviewed for sanity - don't use */ diff --git a/fs/afs/protocol_uae.h b/fs/afs/protocol_uae.h new file mode 100644 index 000000000000..1b3d1060bd34 --- /dev/null +++ b/fs/afs/protocol_uae.h @@ -0,0 +1,132 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Universal AFS Error codes (UAE). + * + * Copyright (C) 2003, Daria Phoebe Brashear + * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. + */ + +enum { + UAEPERM = 0x2f6df00, /* Operation not permitted */ + UAENOENT = 0x2f6df01, /* No such file or directory */ + UAESRCH = 0x2f6df02, /* No such process */ + UAEINTR = 0x2f6df03, /* Interrupted system call */ + UAEIO = 0x2f6df04, /* I/O error */ + UAENXIO = 0x2f6df05, /* No such device or address */ + UAE2BIG = 0x2f6df06, /* Arg list too long */ + UAENOEXEC = 0x2f6df07, /* Exec format error */ + UAEBADF = 0x2f6df08, /* Bad file number */ + UAECHILD = 0x2f6df09, /* No child processes */ + UAEAGAIN = 0x2f6df0a, /* Try again */ + UAENOMEM = 0x2f6df0b, /* Out of memory */ + UAEACCES = 0x2f6df0c, /* Permission denied */ + UAEFAULT = 0x2f6df0d, /* Bad address */ + UAENOTBLK = 0x2f6df0e, /* Block device required */ + UAEBUSY = 0x2f6df0f, /* Device or resource busy */ + UAEEXIST = 0x2f6df10, /* File exists */ + UAEXDEV = 0x2f6df11, /* Cross-device link */ + UAENODEV = 0x2f6df12, /* No such device */ + UAENOTDIR = 0x2f6df13, /* Not a directory */ + UAEISDIR = 0x2f6df14, /* Is a directory */ + UAEINVAL = 0x2f6df15, /* Invalid argument */ + UAENFILE = 0x2f6df16, /* File table overflow */ + UAEMFILE = 0x2f6df17, /* Too many open files */ + UAENOTTY = 0x2f6df18, /* Not a typewriter */ + UAETXTBSY = 0x2f6df19, /* Text file busy */ + UAEFBIG = 0x2f6df1a, /* File too large */ + UAENOSPC = 0x2f6df1b, /* No space left on device */ + UAESPIPE = 0x2f6df1c, /* Illegal seek */ + UAEROFS = 0x2f6df1d, /* Read-only file system */ + UAEMLINK = 0x2f6df1e, /* Too many links */ + UAEPIPE = 0x2f6df1f, /* Broken pipe */ + UAEDOM = 0x2f6df20, /* Math argument out of domain of func */ + UAERANGE = 0x2f6df21, /* Math result not representable */ + UAEDEADLK = 0x2f6df22, /* Resource deadlock would occur */ + UAENAMETOOLONG = 0x2f6df23, /* File name too long */ + UAENOLCK = 0x2f6df24, /* No record locks available */ + UAENOSYS = 0x2f6df25, /* Function not implemented */ + UAENOTEMPTY = 0x2f6df26, /* Directory not empty */ + UAELOOP = 0x2f6df27, /* Too many symbolic links encountered */ + UAEWOULDBLOCK = 0x2f6df28, /* Operation would block */ + UAENOMSG = 0x2f6df29, /* No message of desired type */ + UAEIDRM = 0x2f6df2a, /* Identifier removed */ + UAECHRNG = 0x2f6df2b, /* Channel number out of range */ + UAEL2NSYNC = 0x2f6df2c, /* Level 2 not synchronized */ + UAEL3HLT = 0x2f6df2d, /* Level 3 halted */ + UAEL3RST = 0x2f6df2e, /* Level 3 reset */ + UAELNRNG = 0x2f6df2f, /* Link number out of range */ + UAEUNATCH = 0x2f6df30, /* Protocol driver not attached */ + UAENOCSI = 0x2f6df31, /* No CSI structure available */ + UAEL2HLT = 0x2f6df32, /* Level 2 halted */ + UAEBADE = 0x2f6df33, /* Invalid exchange */ + UAEBADR = 0x2f6df34, /* Invalid request descriptor */ + UAEXFULL = 0x2f6df35, /* Exchange full */ + UAENOANO = 0x2f6df36, /* No anode */ + UAEBADRQC = 0x2f6df37, /* Invalid request code */ + UAEBADSLT = 0x2f6df38, /* Invalid slot */ + UAEBFONT = 0x2f6df39, /* Bad font file format */ + UAENOSTR = 0x2f6df3a, /* Device not a stream */ + UAENODATA = 0x2f6df3b, /* No data available */ + UAETIME = 0x2f6df3c, /* Timer expired */ + UAENOSR = 0x2f6df3d, /* Out of streams resources */ + UAENONET = 0x2f6df3e, /* Machine is not on the network */ + UAENOPKG = 0x2f6df3f, /* Package not installed */ + UAEREMOTE = 0x2f6df40, /* Object is remote */ + UAENOLINK = 0x2f6df41, /* Link has been severed */ + UAEADV = 0x2f6df42, /* Advertise error */ + UAESRMNT = 0x2f6df43, /* Srmount error */ + UAECOMM = 0x2f6df44, /* Communication error on send */ + UAEPROTO = 0x2f6df45, /* Protocol error */ + UAEMULTIHOP = 0x2f6df46, /* Multihop attempted */ + UAEDOTDOT = 0x2f6df47, /* RFS specific error */ + UAEBADMSG = 0x2f6df48, /* Not a data message */ + UAEOVERFLOW = 0x2f6df49, /* Value too large for defined data type */ + UAENOTUNIQ = 0x2f6df4a, /* Name not unique on network */ + UAEBADFD = 0x2f6df4b, /* File descriptor in bad state */ + UAEREMCHG = 0x2f6df4c, /* Remote address changed */ + UAELIBACC = 0x2f6df4d, /* Can not access a needed shared library */ + UAELIBBAD = 0x2f6df4e, /* Accessing a corrupted shared library */ + UAELIBSCN = 0x2f6df4f, /* .lib section in a.out corrupted */ + UAELIBMAX = 0x2f6df50, /* Attempting to link in too many shared libraries */ + UAELIBEXEC = 0x2f6df51, /* Cannot exec a shared library directly */ + UAEILSEQ = 0x2f6df52, /* Illegal byte sequence */ + UAERESTART = 0x2f6df53, /* Interrupted system call should be restarted */ + UAESTRPIPE = 0x2f6df54, /* Streams pipe error */ + UAEUSERS = 0x2f6df55, /* Too many users */ + UAENOTSOCK = 0x2f6df56, /* Socket operation on non-socket */ + UAEDESTADDRREQ = 0x2f6df57, /* Destination address required */ + UAEMSGSIZE = 0x2f6df58, /* Message too long */ + UAEPROTOTYPE = 0x2f6df59, /* Protocol wrong type for socket */ + UAENOPROTOOPT = 0x2f6df5a, /* Protocol not available */ + UAEPROTONOSUPPORT = 0x2f6df5b, /* Protocol not supported */ + UAESOCKTNOSUPPORT = 0x2f6df5c, /* Socket type not supported */ + UAEOPNOTSUPP = 0x2f6df5d, /* Operation not supported on transport endpoint */ + UAEPFNOSUPPORT = 0x2f6df5e, /* Protocol family not supported */ + UAEAFNOSUPPORT = 0x2f6df5f, /* Address family not supported by protocol */ + UAEADDRINUSE = 0x2f6df60, /* Address already in use */ + UAEADDRNOTAVAIL = 0x2f6df61, /* Cannot assign requested address */ + UAENETDOWN = 0x2f6df62, /* Network is down */ + UAENETUNREACH = 0x2f6df63, /* Network is unreachable */ + UAENETRESET = 0x2f6df64, /* Network dropped connection because of reset */ + UAECONNABORTED = 0x2f6df65, /* Software caused connection abort */ + UAECONNRESET = 0x2f6df66, /* Connection reset by peer */ + UAENOBUFS = 0x2f6df67, /* No buffer space available */ + UAEISCONN = 0x2f6df68, /* Transport endpoint is already connected */ + UAENOTCONN = 0x2f6df69, /* Transport endpoint is not connected */ + UAESHUTDOWN = 0x2f6df6a, /* Cannot send after transport endpoint shutdown */ + UAETOOMANYREFS = 0x2f6df6b, /* Too many references: cannot splice */ + UAETIMEDOUT = 0x2f6df6c, /* Connection timed out */ + UAECONNREFUSED = 0x2f6df6d, /* Connection refused */ + UAEHOSTDOWN = 0x2f6df6e, /* Host is down */ + UAEHOSTUNREACH = 0x2f6df6f, /* No route to host */ + UAEALREADY = 0x2f6df70, /* Operation already in progress */ + UAEINPROGRESS = 0x2f6df71, /* Operation now in progress */ + UAESTALE = 0x2f6df72, /* Stale NFS file handle */ + UAEUCLEAN = 0x2f6df73, /* Structure needs cleaning */ + UAENOTNAM = 0x2f6df74, /* Not a XENIX named type file */ + UAENAVAIL = 0x2f6df75, /* No XENIX semaphores available */ + UAEISNAM = 0x2f6df76, /* Is a named type file */ + UAEREMOTEIO = 0x2f6df77, /* Remote I/O error */ + UAEDQUOT = 0x2f6df78, /* Quota exceeded */ + UAENOMEDIUM = 0x2f6df79, /* No medium found */ + UAEMEDIUMTYPE = 0x2f6df7a, /* Wrong medium type */ +}; diff --git a/fs/afs/protocol_yfs.h b/fs/afs/protocol_yfs.h index d443e2bfa094..b2f06c1917c2 100644 --- a/fs/afs/protocol_yfs.h +++ b/fs/afs/protocol_yfs.h @@ -1,18 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* YFS protocol bits * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #define YFS_FS_SERVICE 2500 #define YFS_CM_SERVICE 2501 -#define YFSCBMAX 1024 +#define YFSCBMAX 1024 enum YFS_CM_Operations { YFSCBProbe = 206, /* probe client */ @@ -31,9 +27,9 @@ enum YFS_CM_Operations { }; enum YFS_FS_Operations { - YFSFETCHACL = 64131, /* YFS Fetch file ACL */ + YFSFETCHACL = 64131, /* YFS Fetch file AFS3 ACL */ YFSFETCHSTATUS = 64132, /* YFS Fetch file status */ - YFSSTOREACL = 64134, /* YFS Store file ACL */ + YFSSTOREACL = 64134, /* YFS Store file AFS3 ACL */ YFSSTORESTATUS = 64135, /* YFS Store file status */ YFSREMOVEFILE = 64136, /* YFS Remove a file */ YFSCREATEFILE = 64137, /* YFS Create a file */ @@ -49,11 +45,14 @@ enum YFS_FS_Operations { YFSRELEASELOCK = 64158, /* YFS Release a file lock */ YFSLOOKUP = 64161, /* YFS lookup file in directory */ YFSFLUSHCPS = 64165, - YFSFETCHOPAQUEACL = 64168, + YFSFETCHOPAQUEACL = 64168, /* YFS Fetch file YFS ACL */ YFSWHOAMI = 64170, YFSREMOVEACL = 64171, YFSREMOVEFILE2 = 64173, YFSSTOREOPAQUEACL2 = 64174, + YFSRENAME_REPLACE = 64176, + YFSRENAME_NOREPLACE = 64177, + YFSRENAME_EXCHANGE = 64187, YFSINLINEBULKSTATUS = 64536, /* YFS Fetch multiple file statuses with errors */ YFSFETCHDATA64 = 64537, /* YFS Fetch file data */ YFSSTOREDATA64 = 64538, /* YFS Store file data */ @@ -172,3 +171,9 @@ enum yfs_lock_type { yfs_LockMandatoryWrite = 0x101, yfs_LockMandatoryExtend = 0x102, }; + +/* RXYFS Viced Capability Flags */ +#define YFS_VICED_CAPABILITY_ERRORTRANS 0x0001 /* Deprecated v0.195 */ +#define YFS_VICED_CAPABILITY_64BITFILES 0x0002 /* Deprecated v0.195 */ +#define YFS_VICED_CAPABILITY_WRITELOCKACL 0x0004 /* Can lock a file even without lock perm */ +#define YFS_VICED_CAPABILITY_SANEACLS 0x0008 /* Deprecated v0.195 */ diff --git a/fs/afs/rotate.c b/fs/afs/rotate.c index c3ae324781f8..6a4e7da10fc4 100644 --- a/fs/afs/rotate.c +++ b/fs/afs/rotate.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Handle fileserver selection and rotation. * * Copyright (C) 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/kernel.h> @@ -17,56 +13,74 @@ #include <linux/sched/signal.h> #include "internal.h" #include "afs_fs.h" +#include "protocol_uae.h" -/* - * Begin an operation on the fileserver. - * - * Fileserver operations are serialised on the server by vnode, so we serialise - * them here also using the io_lock. - */ -bool afs_begin_vnode_operation(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - struct key *key) +void afs_clear_server_states(struct afs_operation *op) { - memset(fc, 0, sizeof(*fc)); - fc->vnode = vnode; - fc->key = key; - fc->ac.error = SHRT_MAX; - fc->error = -EDESTADDRREQ; - - if (mutex_lock_interruptible(&vnode->io_lock) < 0) { - fc->error = -EINTR; - fc->flags |= AFS_FS_CURSOR_STOP; - return false; - } + unsigned int i; - if (vnode->lock_state != AFS_VNODE_LOCK_NONE) - fc->flags |= AFS_FS_CURSOR_CUR_ONLY; - return true; + if (op->server_states) { + for (i = 0; i < op->server_list->nr_servers; i++) + afs_put_endpoint_state(op->server_states[i].endpoint_state, + afs_estate_trace_put_server_state); + kfree(op->server_states); + } } /* * Begin iteration through a server list, starting with the vnode's last used * server if possible, or the last recorded good server if not. */ -static bool afs_start_fs_iteration(struct afs_fs_cursor *fc, +static bool afs_start_fs_iteration(struct afs_operation *op, struct afs_vnode *vnode) { - struct afs_cb_interest *cbi; + struct afs_server *server; + void *cb_server; int i; - read_lock(&vnode->volume->servers_lock); - fc->server_list = afs_get_serverlist(vnode->volume->servers); - read_unlock(&vnode->volume->servers_lock); + trace_afs_rotate(op, afs_rotate_trace_start, 0); + + read_lock(&op->volume->servers_lock); + op->server_list = afs_get_serverlist( + rcu_dereference_protected(op->volume->servers, + lockdep_is_held(&op->volume->servers_lock))); + read_unlock(&op->volume->servers_lock); - fc->untried = (1UL << fc->server_list->nr_servers) - 1; - fc->index = READ_ONCE(fc->server_list->preferred); + op->server_states = kcalloc(op->server_list->nr_servers, sizeof(op->server_states[0]), + GFP_KERNEL); + if (!op->server_states) { + afs_op_nomem(op); + trace_afs_rotate(op, afs_rotate_trace_nomem, 0); + return false; + } + + rcu_read_lock(); + for (i = 0; i < op->server_list->nr_servers; i++) { + struct afs_endpoint_state *estate; + struct afs_server_state *s = &op->server_states[i]; + + server = op->server_list->servers[i].server; + estate = rcu_dereference(server->endpoint_state); + s->endpoint_state = afs_get_endpoint_state(estate, + afs_estate_trace_get_server_state); + s->probe_seq = estate->probe_seq; + s->untried_addrs = (1UL << estate->addresses->nr_addrs) - 1; + init_waitqueue_entry(&s->probe_waiter, current); + afs_get_address_preferences(op->net, estate->addresses); + } + rcu_read_unlock(); + + + op->untried_servers = (1UL << op->server_list->nr_servers) - 1; + op->server_index = -1; - cbi = vnode->cb_interest; - if (cbi) { + cb_server = vnode->cb_server; + if (cb_server) { /* See if the vnode's preferred record is still available */ - for (i = 0; i < fc->server_list->nr_servers; i++) { - if (fc->server_list->servers[i].cb_interest == cbi) { - fc->index = i; + for (i = 0; i < op->server_list->nr_servers; i++) { + server = op->server_list->servers[i].server; + if (server == cb_server) { + op->server_index = i; goto found_interest; } } @@ -75,21 +89,19 @@ static bool afs_start_fs_iteration(struct afs_fs_cursor *fc, * serving this vnode, then we can't switch to another server * and have to return an error. */ - if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) { - fc->error = -ESTALE; + if (op->flags & AFS_OPERATION_CUR_ONLY) { + afs_op_set_error(op, -ESTALE); + trace_afs_rotate(op, afs_rotate_trace_stale_lock, 0); return false; } /* Note that the callback promise is effectively broken */ write_seqlock(&vnode->cb_lock); - ASSERTCMP(cbi, ==, vnode->cb_interest); - vnode->cb_interest = NULL; - if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) + ASSERTCMP(cb_server, ==, vnode->cb_server); + vnode->cb_server = NULL; + if (afs_clear_cb_promise(vnode, afs_cb_promise_clear_rotate_server)) vnode->cb_break++; write_sequnlock(&vnode->cb_lock); - - afs_put_cb_interest(afs_v2net(vnode), cbi); - cbi = NULL; } found_interest: @@ -99,7 +111,7 @@ found_interest: /* * Post volume busy note. */ -static void afs_busy(struct afs_volume *volume, u32 abort_code) +static void afs_busy(struct afs_operation *op, u32 abort_code) { const char *m; @@ -110,18 +122,24 @@ static void afs_busy(struct afs_volume *volume, u32 abort_code) default: m = "busy"; break; } - pr_notice("kAFS: Volume %llu '%s' is %s\n", volume->vid, volume->name, m); + pr_notice("kAFS: Volume %llu '%s' on server %pU is %s\n", + op->volume->vid, op->volume->name, &op->server->uuid, m); } /* * Sleep and retry the operation to the same fileserver. */ -static bool afs_sleep_and_retry(struct afs_fs_cursor *fc) +static bool afs_sleep_and_retry(struct afs_operation *op) { - msleep_interruptible(1000); - if (signal_pending(current)) { - fc->error = -ERESTARTSYS; - return false; + trace_afs_rotate(op, afs_rotate_trace_busy_sleep, 0); + if (!(op->flags & AFS_OPERATION_UNINTR)) { + msleep_interruptible(1000); + if (signal_pending(current)) { + afs_op_set_error(op, -ERESTARTSYS); + return false; + } + } else { + msleep(1000); } return true; @@ -131,132 +149,224 @@ static bool afs_sleep_and_retry(struct afs_fs_cursor *fc) * Select the fileserver to use. May be called multiple times to rotate * through the fileservers. */ -bool afs_select_fileserver(struct afs_fs_cursor *fc) +bool afs_select_fileserver(struct afs_operation *op) { struct afs_addr_list *alist; struct afs_server *server; - struct afs_vnode *vnode = fc->vnode; - struct afs_error e; - u32 rtt; - int error = fc->ac.error, i; - - _enter("%lx[%d],%lx[%d],%d,%d", - fc->untried, fc->index, - fc->ac.tried, fc->ac.index, - error, fc->ac.abort_code); - - if (fc->flags & AFS_FS_CURSOR_STOP) { + struct afs_vnode *vnode = op->file[0].vnode; + unsigned long set, failed; + s32 abort_code = op->call_abort_code; + int best_prio = 0; + int error = op->call_error, addr_index, i, j; + + op->nr_iterations++; + + _enter("OP=%x+%x,%llx,%u{%lx},%u{%lx},%d,%d", + op->debug_id, op->nr_iterations, op->volume->vid, + op->server_index, op->untried_servers, + op->addr_index, op->addr_tried, + error, abort_code); + + if (op->flags & AFS_OPERATION_STOP) { + trace_afs_rotate(op, afs_rotate_trace_stopped, 0); _leave(" = f [stopped]"); return false; } - fc->nr_iterations++; - - /* Evaluate the result of the previous operation, if there was one. */ - switch (error) { - case SHRT_MAX: + if (op->nr_iterations == 0) goto start; + WRITE_ONCE(op->estate->addresses->addrs[op->addr_index].last_error, error); + trace_afs_rotate(op, afs_rotate_trace_iter, op->call_error); + + /* Evaluate the result of the previous operation, if there was one. */ + switch (op->call_error) { case 0: + clear_bit(AFS_SE_VOLUME_OFFLINE, + &op->server_list->servers[op->server_index].flags); + clear_bit(AFS_SE_VOLUME_BUSY, + &op->server_list->servers[op->server_index].flags); + op->cumul_error.responded = true; + + /* We succeeded, but we may need to redo the op from another + * server if we're looking at a set of RO volumes where some of + * the servers have not yet been brought up to date lest we + * regress the data. We only switch to the new version once + * >=50% of the servers are updated. + */ + error = afs_update_volume_state(op); + if (error != 0) { + if (error == 1) { + afs_sleep_and_retry(op); + goto restart_from_beginning; + } + afs_op_set_error(op, error); + goto failed; + } + fallthrough; default: /* Success or local failure. Stop. */ - fc->error = error; - fc->flags |= AFS_FS_CURSOR_STOP; + afs_op_set_error(op, error); + op->flags |= AFS_OPERATION_STOP; + trace_afs_rotate(op, afs_rotate_trace_stop, error); _leave(" = f [okay/local %d]", error); return false; case -ECONNABORTED: /* The far side rejected the operation on some grounds. This * might involve the server being busy or the volume having been moved. + * + * Note that various V* errors should not be sent to a cache manager + * by a fileserver as they should be translated to more modern UAE* + * errors instead. IBM AFS and OpenAFS fileservers, however, do leak + * these abort codes. */ - switch (fc->ac.abort_code) { + trace_afs_rotate(op, afs_rotate_trace_aborted, abort_code); + op->cumul_error.responded = true; + switch (abort_code) { case VNOVOL: /* This fileserver doesn't know about the volume. * - May indicate that the VL is wrong - retry once and compare * the results. * - May indicate that the fileserver couldn't attach to the vol. + * - The volume might have been temporarily removed so that it can + * be replaced by a volume restore. "vos" might have ended one + * transaction and has yet to create the next. + * - The volume might not be blessed or might not be in-service + * (administrative action). */ - if (fc->flags & AFS_FS_CURSOR_VNOVOL) { - fc->error = -EREMOTEIO; + if (op->flags & AFS_OPERATION_VNOVOL) { + afs_op_accumulate_error(op, -EREMOTEIO, abort_code); goto next_server; } - write_lock(&vnode->volume->servers_lock); - fc->server_list->vnovol_mask |= 1 << fc->index; - write_unlock(&vnode->volume->servers_lock); + write_lock(&op->volume->servers_lock); + op->server_list->vnovol_mask |= 1 << op->server_index; + write_unlock(&op->volume->servers_lock); - set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags); - error = afs_check_volume_status(vnode->volume, fc->key); - if (error < 0) - goto failed_set_error; + set_bit(AFS_VOLUME_NEEDS_UPDATE, &op->volume->flags); + error = afs_check_volume_status(op->volume, op); + if (error < 0) { + afs_op_set_error(op, error); + goto failed; + } - if (test_bit(AFS_VOLUME_DELETED, &vnode->volume->flags)) { - fc->error = -ENOMEDIUM; + if (test_bit(AFS_VOLUME_DELETED, &op->volume->flags)) { + afs_op_set_error(op, -ENOMEDIUM); goto failed; } /* If the server list didn't change, then assume that * it's the fileserver having trouble. */ - if (vnode->volume->servers == fc->server_list) { - fc->error = -EREMOTEIO; + if (rcu_access_pointer(op->volume->servers) == op->server_list) { + afs_op_accumulate_error(op, -EREMOTEIO, abort_code); goto next_server; } /* Try again */ - fc->flags |= AFS_FS_CURSOR_VNOVOL; + op->flags |= AFS_OPERATION_VNOVOL; _leave(" = t [vnovol]"); return true; - case VSALVAGE: /* TODO: Should this return an error or iterate? */ case VVOLEXISTS: - case VNOSERVICE: case VONLINE: - case VDISKFULL: - case VOVERQUOTA: - fc->error = afs_abort_to_error(fc->ac.abort_code); + /* These should not be returned from the fileserver. */ + pr_warn("Fileserver returned unexpected abort %d\n", + abort_code); + afs_op_accumulate_error(op, -EREMOTEIO, abort_code); goto next_server; + case VNOSERVICE: + /* Prior to AFS 3.2 VNOSERVICE was returned from the fileserver + * if the volume was neither in-service nor administratively + * blessed. All usage was replaced by VNOVOL because AFS 3.1 and + * earlier cache managers did not handle VNOSERVICE and assumed + * it was the client OSes errno 105. + * + * Starting with OpenAFS 1.4.8 VNOSERVICE was repurposed as the + * fileserver idle dead time error which was sent in place of + * RX_CALL_TIMEOUT (-3). The error was intended to be sent if the + * fileserver took too long to send a reply to the client. + * RX_CALL_TIMEOUT would have caused the cache manager to mark the + * server down whereas VNOSERVICE since AFS 3.2 would cause cache + * manager to temporarily (up to 15 minutes) mark the volume + * instance as unusable. + * + * The idle dead logic resulted in cache inconsistency since a + * state changing call that the cache manager assumed was dead + * could still be processed to completion by the fileserver. This + * logic was removed in OpenAFS 1.8.0 and VNOSERVICE is no longer + * returned. However, many 1.4.8 through 1.6.24 fileservers are + * still in existence. + * + * AuriStorFS fileservers have never returned VNOSERVICE. + * + * VNOSERVICE should be treated as an alias for RX_CALL_TIMEOUT. + */ + case RX_CALL_TIMEOUT: + afs_op_accumulate_error(op, -ETIMEDOUT, abort_code); + goto next_server; + + case VSALVAGING: /* This error should not be leaked to cache managers + * but is from OpenAFS demand attach fileservers. + * It should be treated as an alias for VOFFLINE. + */ + case VSALVAGE: /* VSALVAGE should be treated as a synonym of VOFFLINE */ case VOFFLINE: - if (!test_and_set_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags)) { - afs_busy(vnode->volume, fc->ac.abort_code); - clear_bit(AFS_VOLUME_BUSY, &vnode->volume->flags); - } - if (fc->flags & AFS_FS_CURSOR_NO_VSLEEP) { - fc->error = -EADV; - goto failed; + /* The volume is in use by the volserver or another volume utility + * for an operation that might alter the contents. The volume is + * expected to come back but it might take a long time (could be + * days). + */ + if (!test_and_set_bit(AFS_SE_VOLUME_OFFLINE, + &op->server_list->servers[op->server_index].flags)) { + afs_busy(op, abort_code); + clear_bit(AFS_SE_VOLUME_BUSY, + &op->server_list->servers[op->server_index].flags); } - if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) { - fc->error = -ESTALE; + if (op->flags & AFS_OPERATION_NO_VSLEEP) { + afs_op_set_error(op, -EADV); goto failed; } goto busy; - case VSALVAGING: - case VRESTARTING: + case VRESTARTING: /* The fileserver is either shutting down or starting up. */ case VBUSY: - /* Retry after going round all the servers unless we - * have a file lock we need to maintain. + /* The volume is in use by the volserver or another volume + * utility for an operation that is not expected to alter the + * contents of the volume. VBUSY does not need to be returned + * for a ROVOL or BACKVOL bound to an ITBusy volserver + * transaction. The fileserver is permitted to continue serving + * content from ROVOLs and BACKVOLs during an ITBusy transaction + * because the content will not change. However, many fileserver + * releases do return VBUSY for ROVOL and BACKVOL instances under + * many circumstances. + * + * Retry after going round all the servers unless we have a file + * lock we need to maintain. */ - if (fc->flags & AFS_FS_CURSOR_NO_VSLEEP) { - fc->error = -EBUSY; + if (op->flags & AFS_OPERATION_NO_VSLEEP) { + afs_op_set_error(op, -EBUSY); goto failed; } - if (!test_and_set_bit(AFS_VOLUME_BUSY, &vnode->volume->flags)) { - afs_busy(vnode->volume, fc->ac.abort_code); - clear_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags); + if (!test_and_set_bit(AFS_SE_VOLUME_BUSY, + &op->server_list->servers[op->server_index].flags)) { + afs_busy(op, abort_code); + clear_bit(AFS_SE_VOLUME_OFFLINE, + &op->server_list->servers[op->server_index].flags); } busy: - if (fc->flags & AFS_FS_CURSOR_CUR_ONLY) { - if (!afs_sleep_and_retry(fc)) + if (op->flags & AFS_OPERATION_CUR_ONLY) { + if (!afs_sleep_and_retry(op)) goto failed; - /* Retry with same server & address */ + /* Retry with same server & address */ _leave(" = t [vbusy]"); return true; } - fc->flags |= AFS_FS_CURSOR_VBUSY; + op->flags |= AFS_OPERATION_VBUSY; goto next_server; case VMOVED: @@ -267,17 +377,19 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) * We also limit the number of VMOVED hops we will * honour, just in case someone sets up a loop. */ - if (fc->flags & AFS_FS_CURSOR_VMOVED) { - fc->error = -EREMOTEIO; + if (op->flags & AFS_OPERATION_VMOVED) { + afs_op_set_error(op, -EREMOTEIO); goto failed; } - fc->flags |= AFS_FS_CURSOR_VMOVED; + op->flags |= AFS_OPERATION_VMOVED; - set_bit(AFS_VOLUME_WAIT, &vnode->volume->flags); - set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags); - error = afs_check_volume_status(vnode->volume, fc->key); - if (error < 0) - goto failed_set_error; + set_bit(AFS_VOLUME_WAIT, &op->volume->flags); + set_bit(AFS_VOLUME_NEEDS_UPDATE, &op->volume->flags); + error = afs_check_volume_status(op->volume, op); + if (error < 0) { + afs_op_set_error(op, error); + goto failed; + } /* If the server list didn't change, then the VLDB is * out of sync with the fileservers. This is hopefully @@ -288,25 +400,63 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) * * TODO: Retry a few times with sleeps. */ - if (vnode->volume->servers == fc->server_list) { - fc->error = -ENOMEDIUM; + if (rcu_access_pointer(op->volume->servers) == op->server_list) { + afs_op_accumulate_error(op, -ENOMEDIUM, abort_code); goto failed; } goto restart_from_beginning; + case UAEIO: + case VIO: + afs_op_accumulate_error(op, -EREMOTEIO, abort_code); + if (op->volume->type != AFSVL_RWVOL) + goto next_server; + goto failed; + + case VDISKFULL: + case UAENOSPC: + /* The partition is full. Only applies to RWVOLs. + * Translate locally and return ENOSPC. + * No replicas to failover to. + */ + afs_op_set_error(op, -ENOSPC); + goto failed_but_online; + + case VOVERQUOTA: + case UAEDQUOT: + /* Volume is full. Only applies to RWVOLs. + * Translate locally and return EDQUOT. + * No replicas to failover to. + */ + afs_op_set_error(op, -EDQUOT); + goto failed_but_online; + + case RX_INVALID_OPERATION: + case RXGEN_OPCODE: + /* Handle downgrading to an older operation. */ + afs_op_set_error(op, -ENOTSUPP); + if (op->flags & AFS_OPERATION_DOWNGRADE) { + op->flags &= ~AFS_OPERATION_DOWNGRADE; + goto go_again; + } + goto failed_but_online; + default: - clear_bit(AFS_VOLUME_OFFLINE, &vnode->volume->flags); - clear_bit(AFS_VOLUME_BUSY, &vnode->volume->flags); - fc->error = afs_abort_to_error(fc->ac.abort_code); + afs_op_accumulate_error(op, error, abort_code); + failed_but_online: + clear_bit(AFS_SE_VOLUME_OFFLINE, + &op->server_list->servers[op->server_index].flags); + clear_bit(AFS_SE_VOLUME_BUSY, + &op->server_list->servers[op->server_index].flags); goto failed; } case -ETIMEDOUT: case -ETIME: - if (fc->error != -EDESTADDRREQ) + if (afs_op_error(op) != -EDESTADDRREQ) goto iterate_address; - /* Fall through */ + fallthrough; case -ERFKILL: case -EADDRNOTAVAIL: case -ENETUNREACH: @@ -314,244 +464,255 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) case -EHOSTDOWN: case -ECONNREFUSED: _debug("no conn"); - fc->error = error; + afs_op_accumulate_error(op, error, 0); goto iterate_address; + case -ENETRESET: + pr_warn("kAFS: Peer reset %s (op=%x)\n", + op->type ? op->type->name : "???", op->debug_id); + fallthrough; case -ECONNRESET: _debug("call reset"); - fc->error = error; + afs_op_set_error(op, error); goto failed; } restart_from_beginning: + trace_afs_rotate(op, afs_rotate_trace_restart, 0); _debug("restart"); - afs_end_cursor(&fc->ac); - afs_put_cb_interest(afs_v2net(vnode), fc->cbi); - fc->cbi = NULL; - afs_put_serverlist(afs_v2net(vnode), fc->server_list); - fc->server_list = NULL; + op->estate = NULL; + op->server = NULL; + afs_clear_server_states(op); + op->server_states = NULL; + afs_put_serverlist(op->net, op->server_list); + op->server_list = NULL; start: _debug("start"); + ASSERTCMP(op->estate, ==, NULL); /* See if we need to do an update of the volume record. Note that the * volume may have moved or even have been deleted. */ - error = afs_check_volume_status(vnode->volume, fc->key); - if (error < 0) - goto failed_set_error; + error = afs_check_volume_status(op->volume, op); + trace_afs_rotate(op, afs_rotate_trace_check_vol_status, error); + if (error < 0) { + afs_op_set_error(op, error); + goto failed; + } - if (!afs_start_fs_iteration(fc, vnode)) + if (!afs_start_fs_iteration(op, vnode)) goto failed; - _debug("__ VOL %llx __", vnode->volume->vid); - error = afs_probe_fileservers(afs_v2net(vnode), fc->key, fc->server_list); - if (error < 0) - goto failed_set_error; + _debug("__ VOL %llx __", op->volume->vid); pick_server: - _debug("pick [%lx]", fc->untried); + _debug("pick [%lx]", op->untried_servers); + ASSERTCMP(op->estate, ==, NULL); - error = afs_wait_for_fs_probes(fc->server_list, fc->untried); - if (error < 0) - goto failed_set_error; + error = afs_wait_for_fs_probes(op, op->server_states, + !(op->flags & AFS_OPERATION_UNINTR)); + switch (error) { + case 0: /* No untried responsive servers and no outstanding probes */ + trace_afs_rotate(op, afs_rotate_trace_probe_none, 0); + goto no_more_servers; + case 1: /* Got a response */ + trace_afs_rotate(op, afs_rotate_trace_probe_response, 0); + break; + case 2: /* Probe data superseded */ + trace_afs_rotate(op, afs_rotate_trace_probe_superseded, 0); + goto restart_from_beginning; + default: + trace_afs_rotate(op, afs_rotate_trace_probe_error, error); + afs_op_set_error(op, error); + goto failed; + } - /* Pick the untried server with the lowest RTT. If we have outstanding - * callbacks, we stick with the server we're already using if we can. + /* Pick the untried server with the highest priority untried endpoint. + * If we have outstanding callbacks, we stick with the server we're + * already using if we can. */ - if (fc->cbi) { - _debug("cbi %u", fc->index); - if (test_bit(fc->index, &fc->untried)) + if (op->server) { + _debug("server %u", op->server_index); + if (test_bit(op->server_index, &op->untried_servers)) goto selected_server; - afs_put_cb_interest(afs_v2net(vnode), fc->cbi); - fc->cbi = NULL; - _debug("nocbi"); + op->server = NULL; + _debug("no server"); } - fc->index = -1; - rtt = U32_MAX; - for (i = 0; i < fc->server_list->nr_servers; i++) { - struct afs_server *s = fc->server_list->servers[i].server; - - if (!test_bit(i, &fc->untried) || !s->probe.responded) + rcu_read_lock(); + op->server_index = -1; + best_prio = -1; + for (i = 0; i < op->server_list->nr_servers; i++) { + struct afs_endpoint_state *es; + struct afs_server_entry *se = &op->server_list->servers[i]; + struct afs_addr_list *sal; + struct afs_server *s = se->server; + + if (!test_bit(i, &op->untried_servers) || + test_bit(AFS_SE_EXCLUDED, &se->flags) || + !test_bit(AFS_SERVER_FL_RESPONDING, &s->flags)) continue; - if (s->probe.rtt < rtt) { - fc->index = i; - rtt = s->probe.rtt; + es = op->server_states[i].endpoint_state; + sal = es->addresses; + + afs_get_address_preferences_rcu(op->net, sal); + for (j = 0; j < sal->nr_addrs; j++) { + if (es->failed_set & (1 << j)) + continue; + if (!sal->addrs[j].peer) + continue; + if (sal->addrs[j].prio > best_prio) { + op->server_index = i; + best_prio = sal->addrs[j].prio; + } } } + rcu_read_unlock(); - if (fc->index == -1) + if (op->server_index == -1) goto no_more_servers; selected_server: - _debug("use %d", fc->index); - __clear_bit(fc->index, &fc->untried); + trace_afs_rotate(op, afs_rotate_trace_selected_server, best_prio); + _debug("use %d prio %u", op->server_index, best_prio); + __clear_bit(op->server_index, &op->untried_servers); /* We're starting on a different fileserver from the list. We need to * check it, create a callback intercept, find its address list and * probe its capabilities before we use it. */ - ASSERTCMP(fc->ac.alist, ==, NULL); - server = fc->server_list->servers[fc->index].server; + ASSERTCMP(op->estate, ==, NULL); + server = op->server_list->servers[op->server_index].server; - if (!afs_check_server_record(fc, server)) + if (!afs_check_server_record(op, server, op->key)) goto failed; _debug("USING SERVER: %pU", &server->uuid); - /* Make sure we've got a callback interest record for this server. We - * have to link it in before we send the request as we can be sent a - * break request before we've finished decoding the reply and - * installing the vnode. - */ - error = afs_register_server_cb_interest(vnode, fc->server_list, - fc->index); - if (error < 0) - goto failed_set_error; - - fc->cbi = afs_get_cb_interest(vnode->cb_interest); - - read_lock(&server->fs_lock); - alist = rcu_dereference_protected(server->addresses, - lockdep_is_held(&server->fs_lock)); - afs_get_addrlist(alist); - read_unlock(&server->fs_lock); - - memset(&fc->ac, 0, sizeof(fc->ac)); - - if (!fc->ac.alist) - fc->ac.alist = alist; - else - afs_put_addrlist(alist); + op->flags |= AFS_OPERATION_RETRY_SERVER; + op->server = server; + if (vnode->cb_server != server) { + vnode->cb_server = server; + vnode->cb_v_check = atomic_read(&vnode->volume->cb_v_break); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_server_change); + } - fc->ac.index = -1; +retry_server: + op->addr_tried = 0; + op->addr_index = -1; iterate_address: - ASSERT(fc->ac.alist); /* Iterate over the current server's address list to try and find an * address on which it will respond to us. */ - if (!afs_iterate_addresses(&fc->ac)) - goto next_server; + op->estate = op->server_states[op->server_index].endpoint_state; + set = READ_ONCE(op->estate->responsive_set); + failed = READ_ONCE(op->estate->failed_set); + _debug("iterate ES=%x rs=%lx fs=%lx", op->estate->probe_seq, set, failed); + set &= ~(failed | op->addr_tried); + trace_afs_rotate(op, afs_rotate_trace_iterate_addr, set); + if (!set) + goto wait_for_more_probe_results; + + alist = op->estate->addresses; + best_prio = -1; + addr_index = 0; + for (i = 0; i < alist->nr_addrs; i++) { + if (!(set & (1 << i))) + continue; + if (alist->addrs[i].prio > best_prio) { + addr_index = i; + best_prio = alist->addrs[i].prio; + } + } + + alist->preferred = addr_index; - _debug("address [%u] %u/%u", fc->index, fc->ac.index, fc->ac.alist->nr_addrs); + op->addr_index = addr_index; + set_bit(addr_index, &op->addr_tried); + _debug("address [%u] %u/%u %pISp", + op->server_index, addr_index, alist->nr_addrs, + rxrpc_kernel_remote_addr(alist->addrs[op->addr_index].peer)); +go_again: + op->volsync.creation = TIME64_MIN; + op->volsync.update = TIME64_MIN; + op->call_responded = false; _leave(" = t"); return true; +wait_for_more_probe_results: + error = afs_wait_for_one_fs_probe(op->server, op->estate, op->addr_tried, + !(op->flags & AFS_OPERATION_UNINTR)); + if (error == 1) + goto iterate_address; + if (!error) + goto restart_from_beginning; + + /* We've now had a failure to respond on all of a server's addresses - + * immediately probe them again and consider retrying the server. + */ + trace_afs_rotate(op, afs_rotate_trace_probe_fileserver, 0); + afs_probe_fileserver(op->net, op->server); + if (op->flags & AFS_OPERATION_RETRY_SERVER) { + error = afs_wait_for_one_fs_probe(op->server, op->estate, op->addr_tried, + !(op->flags & AFS_OPERATION_UNINTR)); + switch (error) { + case 1: + op->flags &= ~AFS_OPERATION_RETRY_SERVER; + trace_afs_rotate(op, afs_rotate_trace_retry_server, 1); + goto retry_server; + case 0: + trace_afs_rotate(op, afs_rotate_trace_retry_server, 0); + goto restart_from_beginning; + case -ERESTARTSYS: + afs_op_set_error(op, error); + goto failed; + case -ETIME: + case -EDESTADDRREQ: + goto next_server; + } + } + next_server: + trace_afs_rotate(op, afs_rotate_trace_next_server, 0); _debug("next"); - afs_end_cursor(&fc->ac); + op->estate = NULL; goto pick_server; no_more_servers: /* That's all the servers poked to no good effect. Try again if some * of them were busy. */ - if (fc->flags & AFS_FS_CURSOR_VBUSY) + trace_afs_rotate(op, afs_rotate_trace_no_more_servers, 0); + if (op->flags & AFS_OPERATION_VBUSY) { + afs_sleep_and_retry(op); + op->flags &= ~AFS_OPERATION_VBUSY; goto restart_from_beginning; - - e.error = -EDESTADDRREQ; - e.responded = false; - for (i = 0; i < fc->server_list->nr_servers; i++) { - struct afs_server *s = fc->server_list->servers[i].server; - - afs_prioritise_error(&e, READ_ONCE(s->probe.error), - s->probe.abort_code); } -failed_set_error: - fc->error = error; -failed: - fc->flags |= AFS_FS_CURSOR_STOP; - afs_end_cursor(&fc->ac); - _leave(" = f [failed %d]", fc->error); - return false; -} - -/* - * Select the same fileserver we used for a vnode before and only that - * fileserver. We use this when we have a lock on that file, which is backed - * only by the fileserver we obtained it from. - */ -bool afs_select_current_fileserver(struct afs_fs_cursor *fc) -{ - struct afs_vnode *vnode = fc->vnode; - struct afs_cb_interest *cbi = vnode->cb_interest; - struct afs_addr_list *alist; - int error = fc->ac.error; - - _enter(""); - - switch (error) { - case SHRT_MAX: - if (!cbi) { - fc->error = -ESTALE; - fc->flags |= AFS_FS_CURSOR_STOP; - return false; - } - - fc->cbi = afs_get_cb_interest(vnode->cb_interest); - - read_lock(&cbi->server->fs_lock); - alist = rcu_dereference_protected(cbi->server->addresses, - lockdep_is_held(&cbi->server->fs_lock)); - afs_get_addrlist(alist); - read_unlock(&cbi->server->fs_lock); - if (!alist) { - fc->error = -ESTALE; - fc->flags |= AFS_FS_CURSOR_STOP; - return false; - } - - memset(&fc->ac, 0, sizeof(fc->ac)); - fc->ac.alist = alist; - fc->ac.index = -1; - goto iterate_address; - - case 0: - default: - /* Success or local failure. Stop. */ - fc->error = error; - fc->flags |= AFS_FS_CURSOR_STOP; - _leave(" = f [okay/local %d]", error); - return false; - - case -ECONNABORTED: - fc->error = afs_abort_to_error(fc->ac.abort_code); - fc->flags |= AFS_FS_CURSOR_STOP; - _leave(" = f [abort]"); - return false; - - case -ERFKILL: - case -EADDRNOTAVAIL: - case -ENETUNREACH: - case -EHOSTUNREACH: - case -EHOSTDOWN: - case -ECONNREFUSED: - case -ETIMEDOUT: - case -ETIME: - _debug("no conn"); - fc->error = error; - goto iterate_address; - } + rcu_read_lock(); + for (i = 0; i < op->server_list->nr_servers; i++) { + struct afs_endpoint_state *estate; -iterate_address: - /* Iterate over the current server's address list to try and find an - * address on which it will respond to us. - */ - if (afs_iterate_addresses(&fc->ac)) { - _leave(" = t"); - return true; + estate = op->server_states[i].endpoint_state; + error = READ_ONCE(estate->error); + if (error < 0) + afs_op_accumulate_error(op, error, estate->abort_code); } + rcu_read_unlock(); - afs_end_cursor(&fc->ac); +failed: + trace_afs_rotate(op, afs_rotate_trace_failed, 0); + op->flags |= AFS_OPERATION_STOP; + op->estate = NULL; + _leave(" = f [failed %d]", afs_op_error(op)); return false; } /* * Dump cursor state in the case of the error being EDESTADDRREQ. */ -static void afs_dump_edestaddrreq(const struct afs_fs_cursor *fc) +void afs_dump_edestaddrreq(const struct afs_operation *op) { static int count; int i; @@ -563,61 +724,40 @@ static void afs_dump_edestaddrreq(const struct afs_fs_cursor *fc) rcu_read_lock(); pr_notice("EDESTADDR occurred\n"); - pr_notice("FC: cbb=%x cbb2=%x fl=%hx err=%hd\n", - fc->cb_break, fc->cb_break_2, fc->flags, fc->error); - pr_notice("FC: ut=%lx ix=%d ni=%u\n", - fc->untried, fc->index, fc->nr_iterations); - - if (fc->server_list) { - const struct afs_server_list *sl = fc->server_list; - pr_notice("FC: SL nr=%u pr=%u vnov=%hx\n", - sl->nr_servers, sl->preferred, sl->vnovol_mask); + pr_notice("OP: cbb=%x cbb2=%x fl=%x err=%hd\n", + op->file[0].cb_break_before, + op->file[1].cb_break_before, op->flags, op->cumul_error.error); + pr_notice("OP: ut=%lx ix=%d ni=%u\n", + op->untried_servers, op->server_index, op->nr_iterations); + pr_notice("OP: call er=%d ac=%d r=%u\n", + op->call_error, op->call_abort_code, op->call_responded); + + if (op->server_list) { + const struct afs_server_list *sl = op->server_list; + + pr_notice("FC: SL nr=%u vnov=%hx\n", + sl->nr_servers, sl->vnovol_mask); for (i = 0; i < sl->nr_servers; i++) { const struct afs_server *s = sl->servers[i].server; + const struct afs_endpoint_state *e = + rcu_dereference(s->endpoint_state); + const struct afs_addr_list *a = e->addresses; + pr_notice("FC: server fl=%lx av=%u %pU\n", s->flags, s->addr_version, &s->uuid); - if (s->addresses) { - const struct afs_addr_list *a = - rcu_dereference(s->addresses); + pr_notice("FC: - pq=%x R=%lx F=%lx\n", + e->probe_seq, e->responsive_set, e->failed_set); + if (a) { pr_notice("FC: - av=%u nr=%u/%u/%u pr=%u\n", a->version, a->nr_ipv4, a->nr_addrs, a->max_addrs, a->preferred); - pr_notice("FC: - pr=%lx R=%lx F=%lx\n", - a->probed, a->responded, a->failed); - if (a == fc->ac.alist) + if (a == e->addresses) pr_notice("FC: - current\n"); } } } - pr_notice("AC: t=%lx ax=%u ac=%d er=%d r=%u ni=%u\n", - fc->ac.tried, fc->ac.index, fc->ac.abort_code, fc->ac.error, - fc->ac.responded, fc->ac.nr_iterations); + pr_notice("AC: t=%lx ax=%d\n", op->addr_tried, op->addr_index); rcu_read_unlock(); } - -/* - * Tidy up a filesystem cursor and unlock the vnode. - */ -int afs_end_vnode_operation(struct afs_fs_cursor *fc) -{ - struct afs_net *net = afs_v2net(fc->vnode); - - if (fc->error == -EDESTADDRREQ || - fc->error == -EADDRNOTAVAIL || - fc->error == -ENETUNREACH || - fc->error == -EHOSTUNREACH) - afs_dump_edestaddrreq(fc); - - mutex_unlock(&fc->vnode->io_lock); - - afs_end_cursor(&fc->ac); - afs_put_cb_interest(net, fc->cbi); - afs_put_serverlist(net, fc->server_list); - - if (fc->error == -ECONNABORTED) - fc->error = afs_abort_to_error(fc->ac.abort_code); - - return fc->error; -} diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 2c588f9bbbda..bf0e4ea0aafd 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Maintain an RxRPC server socket to do AFS communications through * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/slab.h> @@ -17,18 +13,28 @@ #include "internal.h" #include "afs_cm.h" #include "protocol_yfs.h" +#define RXRPC_TRACE_ONLY_DEFINE_ENUMS +#include <trace/events/rxrpc.h> struct workqueue_struct *afs_async_calls; +static void afs_deferred_free_worker(struct work_struct *work); static void afs_wake_up_call_waiter(struct sock *, struct rxrpc_call *, unsigned long); -static long afs_wait_for_call_to_complete(struct afs_call *, struct afs_addr_cursor *); static void afs_wake_up_async_call(struct sock *, struct rxrpc_call *, unsigned long); -static void afs_delete_async_call(struct work_struct *); static void afs_process_async_call(struct work_struct *); static void afs_rx_new_call(struct sock *, struct rxrpc_call *, unsigned long); static void afs_rx_discard_new_call(struct rxrpc_call *, unsigned long); +static void afs_rx_attach(struct rxrpc_call *rxcall, unsigned long user_call_ID); +static void afs_rx_notify_oob(struct sock *sk, struct sk_buff *oob); static int afs_deliver_cm_op_id(struct afs_call *); +static const struct rxrpc_kernel_ops afs_rxrpc_callback_ops = { + .notify_new_call = afs_rx_new_call, + .discard_new_call = afs_rx_discard_new_call, + .user_attach_call = afs_rx_attach, + .notify_oob = afs_rx_notify_oob, +}; + /* asynchronous incoming call initial processing */ static const struct afs_call_type afs_RXCMxxxx = { .name = "CB.xxxx", @@ -43,7 +49,6 @@ int afs_open_socket(struct afs_net *net) { struct sockaddr_rxrpc srx; struct socket *socket; - unsigned int min_level; int ret; _enter(""); @@ -53,6 +58,7 @@ int afs_open_socket(struct afs_net *net) goto error_1; socket->sk->sk_allocation = GFP_NOFS; + socket->sk->sk_user_data = net; /* bind the callback manager's address to make this a server socket */ memset(&srx, 0, sizeof(srx)); @@ -63,22 +69,29 @@ int afs_open_socket(struct afs_net *net) srx.transport.sin6.sin6_family = AF_INET6; srx.transport.sin6.sin6_port = htons(AFS_CM_PORT); - min_level = RXRPC_SECURITY_ENCRYPT; - ret = kernel_setsockopt(socket, SOL_RXRPC, RXRPC_MIN_SECURITY_LEVEL, - (void *)&min_level, sizeof(min_level)); + ret = rxrpc_sock_set_min_security_level(socket->sk, + RXRPC_SECURITY_ENCRYPT); + if (ret < 0) + goto error_2; + + ret = rxrpc_sock_set_manage_response(socket->sk, true); if (ret < 0) goto error_2; - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = afs_create_token_key(net, socket); + if (ret < 0) + pr_err("Couldn't create RxGK CM key: %d\n", ret); + + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); if (ret == -EADDRINUSE) { srx.transport.sin6.sin6_port = 0; - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); } if (ret < 0) goto error_2; srx.srx_service = YFS_CM_SERVICE; - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); if (ret < 0) goto error_2; @@ -89,8 +102,7 @@ int afs_open_socket(struct afs_net *net) * it sends back to us. */ - rxrpc_kernel_new_call_notification(socket, afs_rx_new_call, - afs_rx_discard_new_call); + rxrpc_kernel_set_notifications(socket, &afs_rxrpc_callback_ops); ret = kernel_listen(socket, INT_MAX); if (ret < 0) @@ -130,7 +142,9 @@ void afs_close_socket(struct afs_net *net) kernel_sock_shutdown(net->socket, SHUT_RDWR); flush_workqueue(afs_async_calls); + net->socket->sk->sk_user_data = NULL; sock_release(net->socket); + key_put(net->fs_cm_token_key); _debug("dework"); _leave(""); @@ -153,66 +167,92 @@ static struct afs_call *afs_alloc_call(struct afs_net *net, call->type = type; call->net = net; call->debug_id = atomic_inc_return(&rxrpc_debug_id); - atomic_set(&call->usage, 1); - INIT_WORK(&call->async_work, afs_process_async_call); + refcount_set(&call->ref, 1); + INIT_WORK(&call->async_work, type->async_rx ?: afs_process_async_call); + INIT_WORK(&call->work, call->type->work); + INIT_WORK(&call->free_work, afs_deferred_free_worker); init_waitqueue_head(&call->waitq); spin_lock_init(&call->state_lock); - call->_iter = &call->iter; + call->iter = &call->def_iter; o = atomic_inc_return(&net->nr_outstanding_calls); - trace_afs_call(call, afs_call_trace_alloc, 1, o, + trace_afs_call(call->debug_id, afs_call_trace_alloc, 1, o, __builtin_return_address(0)); return call; } +static void afs_free_call(struct afs_call *call) +{ + struct afs_net *net = call->net; + int o; + + ASSERT(!work_pending(&call->async_work)); + + rxrpc_kernel_put_peer(call->peer); + + if (call->rxcall) { + rxrpc_kernel_shutdown_call(net->socket, call->rxcall); + rxrpc_kernel_put_call(net->socket, call->rxcall); + call->rxcall = NULL; + } + if (call->type->destructor) + call->type->destructor(call); + + afs_unuse_server_notime(call->net, call->server, afs_server_trace_unuse_call); + kfree(call->request); + + o = atomic_read(&net->nr_outstanding_calls); + trace_afs_call(call->debug_id, afs_call_trace_free, 0, o, + __builtin_return_address(0)); + kfree(call); + + o = atomic_dec_return(&net->nr_outstanding_calls); + if (o == 0) + wake_up_var(&net->nr_outstanding_calls); +} + /* * Dispose of a reference on a call. */ void afs_put_call(struct afs_call *call) { struct afs_net *net = call->net; - int n = atomic_dec_return(&call->usage); - int o = atomic_read(&net->nr_outstanding_calls); + unsigned int debug_id = call->debug_id; + bool zero; + int r, o; - trace_afs_call(call, afs_call_trace_put, n + 1, o, + zero = __refcount_dec_and_test(&call->ref, &r); + o = atomic_read(&net->nr_outstanding_calls); + trace_afs_call(debug_id, afs_call_trace_put, r - 1, o, __builtin_return_address(0)); + if (zero) + afs_free_call(call); +} - ASSERTCMP(n, >=, 0); - if (n == 0) { - ASSERT(!work_pending(&call->async_work)); - ASSERT(call->type->name != NULL); - - if (call->rxcall) { - rxrpc_kernel_end_call(net->socket, call->rxcall); - call->rxcall = NULL; - } - if (call->type->destructor) - call->type->destructor(call); - - afs_put_server(call->net, call->cm_server); - afs_put_cb_interest(call->net, call->cbi); - afs_put_addrlist(call->alist); - kfree(call->request); - - trace_afs_call(call, afs_call_trace_free, 0, o, - __builtin_return_address(0)); - kfree(call); +static void afs_deferred_free_worker(struct work_struct *work) +{ + struct afs_call *call = container_of(work, struct afs_call, free_work); - o = atomic_dec_return(&net->nr_outstanding_calls); - if (o == 0) - wake_up_var(&net->nr_outstanding_calls); - } + afs_free_call(call); } -static struct afs_call *afs_get_call(struct afs_call *call, - enum afs_call_trace why) +/* + * Dispose of a reference on a call, deferring the cleanup to a workqueue + * to avoid lock recursion. + */ +void afs_deferred_put_call(struct afs_call *call) { - int u = atomic_inc_return(&call->usage); + struct afs_net *net = call->net; + unsigned int debug_id = call->debug_id; + bool zero; + int r, o; - trace_afs_call(call, why, u, - atomic_read(&call->net->nr_outstanding_calls), + zero = __refcount_dec_and_test(&call->ref, &r); + o = atomic_read(&net->nr_outstanding_calls); + trace_afs_call(debug_id, afs_call_trace_put, r - 1, o, __builtin_return_address(0)); - return call; + if (zero) + schedule_work(&call->free_work); } /* @@ -221,8 +261,6 @@ static struct afs_call *afs_get_call(struct afs_call *call, static void afs_queue_call_work(struct afs_call *call) { if (call->type->work) { - INIT_WORK(&call->work, call->type->work); - afs_get_call(call, afs_call_trace_work); if (!queue_work(afs_wq, &call->work)) afs_put_call(call); @@ -280,39 +318,6 @@ void afs_flat_call_destructor(struct afs_call *call) call->buffer = NULL; } -#define AFS_BVEC_MAX 8 - -/* - * Load the given bvec with the next few pages. - */ -static void afs_load_bvec(struct afs_call *call, struct msghdr *msg, - struct bio_vec *bv, pgoff_t first, pgoff_t last, - unsigned offset) -{ - struct page *pages[AFS_BVEC_MAX]; - unsigned int nr, n, i, to, bytes = 0; - - nr = min_t(pgoff_t, last - first + 1, AFS_BVEC_MAX); - n = find_get_pages_contig(call->mapping, first, nr, pages); - ASSERTCMP(n, ==, nr); - - msg->msg_flags |= MSG_MORE; - for (i = 0; i < nr; i++) { - to = PAGE_SIZE; - if (first + i >= last) { - to = call->last_to; - msg->msg_flags &= ~MSG_MORE; - } - bv[i].bv_page = pages[i]; - bv[i].bv_len = to - offset; - bv[i].bv_offset = offset; - bytes += to - offset; - offset = 0; - } - - iov_iter_bvec(&msg->msg_iter, WRITE, bv, nr, bytes); -} - /* * Advance the AFS call state when the RxRPC call ends the transmit phase. */ @@ -326,54 +331,19 @@ static void afs_notify_end_request_tx(struct sock *sock, } /* - * attach the data from a bunch of pages on an inode to a call + * Initiate a call and synchronously queue up the parameters for dispatch. Any + * error is stored into the call struct, which the caller must check for. */ -static int afs_send_pages(struct afs_call *call, struct msghdr *msg) +void afs_make_call(struct afs_call *call, gfp_t gfp) { - struct bio_vec bv[AFS_BVEC_MAX]; - unsigned int bytes, nr, loop, offset; - pgoff_t first = call->first, last = call->last; - int ret; - - offset = call->first_offset; - call->first_offset = 0; - - do { - afs_load_bvec(call, msg, bv, first, last, offset); - trace_afs_send_pages(call, msg, first, last, offset); - - offset = 0; - bytes = msg->msg_iter.count; - nr = msg->msg_iter.nr_segs; - - ret = rxrpc_kernel_send_data(call->net->socket, call->rxcall, msg, - bytes, afs_notify_end_request_tx); - for (loop = 0; loop < nr; loop++) - put_page(bv[loop].bv_page); - if (ret < 0) - break; - - first += nr; - } while (first <= last); - - trace_afs_sent_pages(call, call->first, last, first, ret); - return ret; -} - -/* - * initiate a call - */ -long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, - gfp_t gfp, bool async) -{ - struct sockaddr_rxrpc *srx = &ac->alist->addrs[ac->index]; struct rxrpc_call *rxcall; struct msghdr msg; struct kvec iov[1]; + size_t len; s64 tx_total_len; int ret; - _enter(",{%pISp},", &srx->transport); + _enter(",{%pISp+%u},", rxrpc_kernel_remote_addr(call->peer), call->service_id); ASSERT(call->type != NULL); ASSERT(call->type->name != NULL); @@ -382,43 +352,37 @@ long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, call, call->type->name, key_serial(call->key), atomic_read(&call->net->nr_outstanding_calls)); - call->async = async; - call->addr_ix = ac->index; - call->alist = afs_get_addrlist(ac->alist); + trace_afs_make_call(call); /* Work out the length we're going to transmit. This is awkward for * calls such as FS.StoreData where there's an extra injection of data * after the initial fixed part. */ tx_total_len = call->request_size; - if (call->send_pages) { - if (call->last == call->first) { - tx_total_len += call->last_to - call->first_offset; - } else { - /* It looks mathematically like you should be able to - * combine the following lines with the ones above, but - * unsigned arithmetic is fun when it wraps... - */ - tx_total_len += PAGE_SIZE - call->first_offset; - tx_total_len += call->last_to; - tx_total_len += (call->last - call->first - 1) * PAGE_SIZE; - } - } + if (call->write_iter) + tx_total_len += iov_iter_count(call->write_iter); /* If the call is going to be asynchronous, we need an extra ref for * the call to hold itself so the caller need not hang on to its ref. */ - if (call->async) + if (call->async) { afs_get_call(call, afs_call_trace_get); + call->drop_ref = true; + } /* create a call */ - rxcall = rxrpc_kernel_begin_call(call->net->socket, srx, call->key, + rxcall = rxrpc_kernel_begin_call(call->net->socket, call->peer, call->key, (unsigned long)call, - tx_total_len, gfp, - (async ? + tx_total_len, + call->max_lifespan, + gfp, + (call->async ? afs_wake_up_async_call : afs_wake_up_call_waiter), + call->service_id, call->upgrade, + (call->intr ? RXRPC_PREINTERRUPTIBLE : + RXRPC_UNINTERRUPTIBLE), call->debug_id); if (IS_ERR(rxcall)) { ret = PTR_ERR(rxcall); @@ -427,6 +391,7 @@ long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, } call->rxcall = rxcall; + call->issue_time = ktime_get_real(); /* send the request */ iov[0].iov_base = call->request; @@ -434,10 +399,10 @@ long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, msg.msg_name = NULL; msg.msg_namelen = 0; - iov_iter_kvec(&msg.msg_iter, WRITE, iov, 1, call->request_size); + iov_iter_kvec(&msg.msg_iter, ITER_SOURCE, iov, 1, call->request_size); msg.msg_control = NULL; msg.msg_controllen = 0; - msg.msg_flags = MSG_WAITALL | (call->send_pages ? MSG_MORE : 0); + msg.msg_flags = MSG_WAITALL | (call->write_iter ? MSG_MORE : 0); ret = rxrpc_kernel_send_data(call->net->socket, rxcall, &msg, call->request_size, @@ -445,67 +410,113 @@ long afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, if (ret < 0) goto error_do_abort; - if (call->send_pages) { - ret = afs_send_pages(call, &msg); + if (call->write_iter) { + msg.msg_iter = *call->write_iter; + msg.msg_flags &= ~MSG_MORE; + trace_afs_send_data(call, &msg); + + ret = rxrpc_kernel_send_data(call->net->socket, + call->rxcall, &msg, + iov_iter_count(&msg.msg_iter), + afs_notify_end_request_tx); + *call->write_iter = msg.msg_iter; + + trace_afs_sent_data(call, &msg, ret); if (ret < 0) goto error_do_abort; } /* Note that at this point, we may have received the reply or an abort * - and an asynchronous call may already have completed. + * + * afs_wait_for_call_to_complete(call) + * must be called to synchronously clean up. */ - if (call->async) { - afs_put_call(call); - return -EINPROGRESS; - } - - return afs_wait_for_call_to_complete(call, ac); + return; error_do_abort: - if (ret != -ECONNABORTED) { + if (ret != -ECONNABORTED) rxrpc_kernel_abort_call(call->net->socket, rxcall, - RX_USER_ABORT, ret, "KSD"); - } else { - iov_iter_kvec(&msg.msg_iter, READ, NULL, 0, 0); + RX_USER_ABORT, ret, + afs_abort_send_data_error); + if (call->async) { + afs_see_call(call, afs_call_trace_async_abort); + return; + } + + if (ret == -ECONNABORTED) { + len = 0; + iov_iter_kvec(&msg.msg_iter, ITER_DEST, NULL, 0, 0); rxrpc_kernel_recv_data(call->net->socket, rxcall, - &msg.msg_iter, false, + &msg.msg_iter, &len, false, &call->abort_code, &call->service_id); - ac->abort_code = call->abort_code; - ac->responded = true; + call->responded = true; } call->error = ret; trace_afs_call_done(call); error_kill_call: - if (call->type->done) - call->type->done(call); + if (call->async) + afs_see_call(call, afs_call_trace_async_kill); + if (call->type->immediate_cancel) + call->type->immediate_cancel(call); /* We need to dispose of the extra ref we grabbed for an async call. * The call, however, might be queued on afs_async_calls and we need to * make sure we don't get any more notifications that might requeue it. */ - if (call->rxcall) { - rxrpc_kernel_end_call(call->net->socket, call->rxcall); - call->rxcall = NULL; - } + if (call->rxcall) + rxrpc_kernel_shutdown_call(call->net->socket, call->rxcall); if (call->async) { if (cancel_work_sync(&call->async_work)) afs_put_call(call); - afs_put_call(call); + afs_set_call_complete(call, ret, 0); } - ac->error = ret; + call->error = ret; call->state = AFS_CALL_COMPLETE; - afs_put_call(call); _leave(" = %d", ret); - return ret; +} + +/* + * Log remote abort codes that indicate that we have a protocol disagreement + * with the server. + */ +static void afs_log_error(struct afs_call *call, s32 remote_abort) +{ + static int max = 0; + const char *msg; + int m; + + switch (remote_abort) { + case RX_EOF: msg = "unexpected EOF"; break; + case RXGEN_CC_MARSHAL: msg = "client marshalling"; break; + case RXGEN_CC_UNMARSHAL: msg = "client unmarshalling"; break; + case RXGEN_SS_MARSHAL: msg = "server marshalling"; break; + case RXGEN_SS_UNMARSHAL: msg = "server unmarshalling"; break; + case RXGEN_DECODE: msg = "opcode decode"; break; + case RXGEN_SS_XDRFREE: msg = "server XDR cleanup"; break; + case RXGEN_CC_XDRFREE: msg = "client XDR cleanup"; break; + case -32: msg = "insufficient data"; break; + default: + return; + } + + m = max; + if (m < 3) { + max = m + 1; + pr_notice("kAFS: Peer reported %s failure on %s [%pISp]\n", + msg, call->type->name, + rxrpc_kernel_remote_addr(call->peer)); + } } /* * deliver messages to a call */ -static void afs_deliver_to_call(struct afs_call *call) +void afs_deliver_to_call(struct afs_call *call) { enum afs_call_state state; + size_t len; u32 abort_code, remote_abort = 0; int ret; @@ -518,12 +529,13 @@ static void afs_deliver_to_call(struct afs_call *call) state == AFS_CALL_SV_AWAIT_ACK ) { if (state == AFS_CALL_SV_AWAIT_ACK) { - iov_iter_kvec(&call->iter, READ, NULL, 0, 0); + len = 0; + iov_iter_kvec(&call->def_iter, ITER_DEST, NULL, 0, 0); ret = rxrpc_kernel_recv_data(call->net->socket, - call->rxcall, &call->iter, - false, &remote_abort, + call->rxcall, &call->def_iter, + &len, false, &remote_abort, &call->service_id); - trace_afs_receive_data(call, &call->iter, false, ret); + trace_afs_receive_data(call, &call->def_iter, false, ret); if (ret == -EINPROGRESS || ret == -EAGAIN) return; @@ -535,21 +547,18 @@ static void afs_deliver_to_call(struct afs_call *call) return; } - if (call->want_reply_time && - rxrpc_kernel_get_reply_time(call->net->socket, - call->rxcall, - &call->reply_time)) - call->want_reply_time = false; - ret = call->type->deliver(call); state = READ_ONCE(call->state); + if (ret == 0 && call->unmarshalling_error) + ret = -EBADMSG; switch (ret) { case 0: + call->responded = true; afs_queue_call_work(call); if (state == AFS_CALL_CL_PROC_REPLY) { - if (call->cbi) + if (call->op) set_bit(AFS_SERVER_FL_MAY_HAVE_CB, - &call->cbi->server->flags); + &call->op->server->flags); goto call_complete; } ASSERTCMP(state, >, AFS_CALL_CL_PROC_REPLY); @@ -559,25 +568,37 @@ static void afs_deliver_to_call(struct afs_call *call) goto out; case -ECONNABORTED: ASSERTCMP(state, ==, AFS_CALL_COMPLETE); + call->responded = true; + afs_log_error(call, call->abort_code); goto done; case -ENOTSUPP: + call->responded = true; abort_code = RXGEN_OPCODE; rxrpc_kernel_abort_call(call->net->socket, call->rxcall, - abort_code, ret, "KIV"); + abort_code, ret, + afs_abort_op_not_supported); goto local_abort; case -EIO: pr_err("kAFS: Call %u in bad state %u\n", call->debug_id, state); - /* Fall through */ + fallthrough; case -ENODATA: case -EBADMSG: case -EMSGSIZE: - default: + case -ENOMEM: + case -EFAULT: abort_code = RXGEN_CC_UNMARSHAL; if (state != AFS_CALL_CL_AWAIT_REPLY) abort_code = RXGEN_SS_UNMARSHAL; rxrpc_kernel_abort_call(call->net->socket, call->rxcall, - abort_code, ret, "KUM"); + abort_code, ret, + afs_abort_unmarshal_error); + goto local_abort; + default: + abort_code = RX_CALL_DEAD; + rxrpc_kernel_abort_call(call->net->socket, call->rxcall, + abort_code, ret, + afs_abort_general_error); goto local_abort; } } @@ -585,8 +606,6 @@ static void afs_deliver_to_call(struct afs_call *call) done: if (call->type->done) call->type->done(call); - if (state == AFS_CALL_COMPLETE && call->incoming) - afs_put_call(call); out: _leave(""); return; @@ -595,104 +614,62 @@ local_abort: abort_code = 0; call_complete: afs_set_call_complete(call, ret, remote_abort); - state = AFS_CALL_COMPLETE; goto done; } /* - * wait synchronously for a call to complete + * Wait synchronously for a call to complete. */ -static long afs_wait_for_call_to_complete(struct afs_call *call, - struct afs_addr_cursor *ac) +void afs_wait_for_call_to_complete(struct afs_call *call) { - signed long rtt2, timeout; - long ret; - bool stalled = false; - u64 rtt; - u32 life, last_life; - - DECLARE_WAITQUEUE(myself, current); + bool rxrpc_complete = false; _enter(""); - rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall); - rtt2 = nsecs_to_jiffies64(rtt) * 2; - if (rtt2 < 2) - rtt2 = 2; - - timeout = rtt2; - last_life = rxrpc_kernel_check_life(call->net->socket, call->rxcall); - - add_wait_queue(&call->waitq, &myself); - for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); - - /* deliver any messages that are in the queue */ - if (!afs_check_call_state(call, AFS_CALL_COMPLETE) && - call->need_attention) { - call->need_attention = false; - __set_current_state(TASK_RUNNING); - afs_deliver_to_call(call); - continue; - } + if (!afs_check_call_state(call, AFS_CALL_COMPLETE)) { + DECLARE_WAITQUEUE(myself, current); + + add_wait_queue(&call->waitq, &myself); + for (;;) { + set_current_state(TASK_UNINTERRUPTIBLE); + + /* deliver any messages that are in the queue */ + if (!afs_check_call_state(call, AFS_CALL_COMPLETE) && + call->need_attention) { + call->need_attention = false; + __set_current_state(TASK_RUNNING); + afs_deliver_to_call(call); + continue; + } - if (afs_check_call_state(call, AFS_CALL_COMPLETE)) - break; + if (afs_check_call_state(call, AFS_CALL_COMPLETE)) + break; - life = rxrpc_kernel_check_life(call->net->socket, call->rxcall); - if (timeout == 0 && - life == last_life && signal_pending(current)) { - if (stalled) + if (!rxrpc_kernel_check_life(call->net->socket, call->rxcall)) { + /* rxrpc terminated the call. */ + rxrpc_complete = true; break; - __set_current_state(TASK_RUNNING); - rxrpc_kernel_probe_life(call->net->socket, call->rxcall); - timeout = rtt2; - stalled = true; - continue; - } + } - if (life != last_life) { - timeout = rtt2; - last_life = life; - stalled = false; + schedule(); } - timeout = schedule_timeout(timeout); + remove_wait_queue(&call->waitq, &myself); + __set_current_state(TASK_RUNNING); } - remove_wait_queue(&call->waitq, &myself); - __set_current_state(TASK_RUNNING); - - /* Kill off the call if it's still live. */ if (!afs_check_call_state(call, AFS_CALL_COMPLETE)) { - _debug("call interrupted"); - if (rxrpc_kernel_abort_call(call->net->socket, call->rxcall, - RX_USER_ABORT, -EINTR, "KWI")) - afs_set_call_complete(call, -EINTR, 0); - } - - spin_lock_bh(&call->state_lock); - ac->abort_code = call->abort_code; - ac->error = call->error; - spin_unlock_bh(&call->state_lock); - - ret = ac->error; - switch (ret) { - case 0: - if (call->ret_reply0) { - ret = (long)call->reply[0]; - call->reply[0] = NULL; + if (rxrpc_complete) { + afs_set_call_complete(call, call->error, call->abort_code); + } else { + /* Kill off the call if it's still live. */ + _debug("call interrupted"); + if (rxrpc_kernel_abort_call(call->net->socket, call->rxcall, + RX_USER_ABORT, -EINTR, + afs_abort_interrupted)) + afs_set_call_complete(call, -EINTR, 0); } - /* Fall through */ - case -ECONNABORTED: - ac->responded = true; - break; } - - _debug("call complete"); - afs_put_call(call); - _leave(" = %p", (void *)ret); - return ret; } /* @@ -708,44 +685,29 @@ static void afs_wake_up_call_waiter(struct sock *sk, struct rxrpc_call *rxcall, } /* - * wake up an asynchronous call + * Wake up an asynchronous call. The caller is holding the call notify + * spinlock around this, so we can't call afs_put_call(). */ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall, unsigned long call_user_ID) { struct afs_call *call = (struct afs_call *)call_user_ID; - int u; + int r; trace_afs_notify_call(rxcall, call); call->need_attention = true; - u = atomic_fetch_add_unless(&call->usage, 1, 0); - if (u != 0) { - trace_afs_call(call, afs_call_trace_wake, u, + if (__refcount_inc_not_zero(&call->ref, &r)) { + trace_afs_call(call->debug_id, afs_call_trace_wake, r + 1, atomic_read(&call->net->nr_outstanding_calls), __builtin_return_address(0)); if (!queue_work(afs_async_calls, &call->async_work)) - afs_put_call(call); + afs_deferred_put_call(call); } } /* - * Delete an asynchronous call. The work item carries a ref to the call struct - * that we need to release. - */ -static void afs_delete_async_call(struct work_struct *work) -{ - struct afs_call *call = container_of(work, struct afs_call, async_work); - - _enter(""); - - afs_put_call(call); - - _leave(""); -} - -/* * Perform I/O processing on an asynchronous call. The work item carries a ref * to the call struct that we either need to release or to pass on. */ @@ -760,16 +722,6 @@ static void afs_process_async_call(struct work_struct *work) afs_deliver_to_call(call); } - if (call->state == AFS_CALL_COMPLETE) { - /* We have two refs to release - one from the alloc and one - * queued with the work item - and we can't just deallocate the - * call because the work item may be queued again. - */ - call->async_work.func = afs_delete_async_call; - if (!queue_work(afs_async_calls, &call->async_work)) - afs_put_call(call); - } - afs_put_call(call); _leave(""); } @@ -796,6 +748,7 @@ void afs_charge_preallocation(struct work_struct *work) if (!call) break; + call->drop_ref = true; call->async = true; call->state = AFS_CALL_SV_AWAIT_OP_ID; init_waitqueue_head(&call->waitq); @@ -804,7 +757,6 @@ void afs_charge_preallocation(struct work_struct *work) if (rxrpc_kernel_charge_accept(net->socket, afs_wake_up_async_call, - afs_rx_attach, (unsigned long)call, GFP_KERNEL, call->debug_id) < 0) @@ -832,8 +784,14 @@ static void afs_rx_discard_new_call(struct rxrpc_call *rxcall, static void afs_rx_new_call(struct sock *sk, struct rxrpc_call *rxcall, unsigned long user_call_ID) { + struct afs_call *call = (struct afs_call *)user_call_ID; struct afs_net *net = afs_sock2net(sk); + call->peer = rxrpc_kernel_get_call_peer(sk->sk_socket, call->rxcall); + call->server = afs_find_server(call->peer); + if (!call->server) + trace_afs_cm_no_server(call, rxrpc_kernel_remote_srx(call->peer)); + queue_work(afs_wq, &net->charge_preallocation_work); } @@ -845,7 +803,7 @@ static int afs_deliver_cm_op_id(struct afs_call *call) { int ret; - _enter("{%zu}", iov_iter_count(call->_iter)); + _enter("{%zu}", iov_iter_count(call->iter)); /* the operation ID forms the first four bytes of the request data */ ret = afs_extract_data(call, true); @@ -860,9 +818,14 @@ static int afs_deliver_cm_op_id(struct afs_call *call) if (!afs_cm_incoming_call(call)) return -ENOTSUPP; + call->security_ix = rxrpc_kernel_query_call_security(call->rxcall, + &call->service_id, + &call->enctype); + trace_afs_cb_call(call); + call->work.func = call->type->work; - /* pass responsibility for the remainer of this message off to the + /* pass responsibility for the remainder of this message off to the * cache manager op */ return call->type->deliver(call); } @@ -894,7 +857,7 @@ void afs_send_empty_reply(struct afs_call *call) msg.msg_name = NULL; msg.msg_namelen = 0; - iov_iter_kvec(&msg.msg_iter, WRITE, NULL, 0, 0); + iov_iter_kvec(&msg.msg_iter, ITER_SOURCE, NULL, 0, 0); msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -908,7 +871,9 @@ void afs_send_empty_reply(struct afs_call *call) case -ENOMEM: _debug("oom"); rxrpc_kernel_abort_call(net->socket, call->rxcall, - RX_USER_ABORT, -ENOMEM, "KOO"); + RXGEN_SS_MARSHAL, -ENOMEM, + afs_abort_oom); + fallthrough; default: _leave(" [error]"); return; @@ -933,7 +898,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len) iov[0].iov_len = len; msg.msg_name = NULL; msg.msg_namelen = 0; - iov_iter_kvec(&msg.msg_iter, WRITE, iov, 1, len); + iov_iter_kvec(&msg.msg_iter, ITER_SOURCE, iov, 1, len); msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -949,7 +914,8 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len) if (n == -ENOMEM) { _debug("oom"); rxrpc_kernel_abort_call(net->socket, call->rxcall, - RX_USER_ABORT, -ENOMEM, "KOO"); + RXGEN_SS_MARSHAL, -ENOMEM, + afs_abort_oom); } _leave(" [error]"); } @@ -960,16 +926,18 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len) int afs_extract_data(struct afs_call *call, bool want_more) { struct afs_net *net = call->net; - struct iov_iter *iter = call->_iter; + struct iov_iter *iter = call->iter; enum afs_call_state state; u32 remote_abort = 0; int ret; - _enter("{%s,%zu},%d", call->type->name, iov_iter_count(iter), want_more); + _enter("{%s,%zu,%zu},%d", + call->type->name, call->iov_len, iov_iter_count(iter), want_more); ret = rxrpc_kernel_recv_data(net->socket, call->rxcall, iter, - want_more, &remote_abort, + &call->iov_len, want_more, &remote_abort, &call->service_id); + trace_afs_receive_data(call, call->iter, want_more, ret); if (ret == 0 || ret == -EAGAIN) return ret; @@ -998,9 +966,21 @@ int afs_extract_data(struct afs_call *call, bool want_more) /* * Log protocol error production. */ -noinline int afs_protocol_error(struct afs_call *call, int error, +noinline int afs_protocol_error(struct afs_call *call, enum afs_eproto_cause cause) { - trace_afs_protocol_error(call, error, cause); - return error; + trace_afs_protocol_error(call, cause); + if (call) + call->unmarshalling_error = true; + return -EBADMSG; +} + +/* + * Wake up OOB notification processing. + */ +static void afs_rx_notify_oob(struct sock *sk, struct sk_buff *oob) +{ + struct afs_net *net = sk->sk_user_data; + + schedule_work(&net->rx_oob_work); } diff --git a/fs/afs/security.c b/fs/afs/security.c index 5f58a9a17e69..55ddce94af03 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS security handling * * Copyright (C) 2007, 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/init.h> @@ -20,6 +16,31 @@ static DEFINE_HASHTABLE(afs_permits_cache, 10); static DEFINE_SPINLOCK(afs_permits_lock); +static DEFINE_MUTEX(afs_key_lock); + +/* + * Allocate a key to use as a placeholder for anonymous user security. + */ +static int afs_alloc_anon_key(struct afs_cell *cell) +{ + struct key *key; + + mutex_lock(&afs_key_lock); + key = cell->anonymous_key; + if (!key) { + key = rxrpc_get_null_key(cell->key_desc); + if (!IS_ERR(key)) + cell->anonymous_key = key; + } + mutex_unlock(&afs_key_lock); + + if (IS_ERR(key)) + return PTR_ERR(key); + + _debug("anon key %p{%x}", + cell->anonymous_key, key_serial(cell->anonymous_key)); + return 0; +} /* * get a key @@ -27,12 +48,47 @@ static DEFINE_SPINLOCK(afs_permits_lock); struct key *afs_request_key(struct afs_cell *cell) { struct key *key; + int ret; + + _enter("{%s}", cell->key_desc); + + _debug("key %s", cell->key_desc); + key = request_key_net(&key_type_rxrpc, cell->key_desc, + cell->net->net, NULL); + if (IS_ERR(key)) { + if (PTR_ERR(key) != -ENOKEY) { + _leave(" = %ld", PTR_ERR(key)); + return key; + } + + if (!cell->anonymous_key) { + ret = afs_alloc_anon_key(cell); + if (ret < 0) + return ERR_PTR(ret); + } + + /* act as anonymous user */ + _leave(" = {%x} [anon]", key_serial(cell->anonymous_key)); + return key_get(cell->anonymous_key); + } else { + /* act as authorised user */ + _leave(" = {%x} [auth]", key_serial(key)); + return key; + } +} + +/* + * Get a key when pathwalk is in rcuwalk mode. + */ +struct key *afs_request_key_rcu(struct afs_cell *cell) +{ + struct key *key; - _enter("{%x}", key_serial(cell->anonymous_key)); + _enter("{%s}", cell->key_desc); - _debug("key %s", cell->anonymous_key->description); - key = request_key(&key_type_rxrpc, cell->anonymous_key->description, - NULL); + _debug("key %s", cell->key_desc); + key = request_key_net_rcu(&key_type_rxrpc, cell->key_desc, + cell->net->net); if (IS_ERR(key)) { if (PTR_ERR(key) != -ENOKEY) { _leave(" = %ld", PTR_ERR(key)); @@ -40,6 +96,8 @@ struct key *afs_request_key(struct afs_cell *cell) } /* act as anonymous user */ + if (!cell->anonymous_key) + return NULL; /* Need to allocate */ _leave(" = {%x} [anon]", key_serial(cell->anonymous_key)); return key_get(cell->anonymous_key); } else { @@ -87,11 +145,9 @@ void afs_clear_permits(struct afs_vnode *vnode) permits = rcu_dereference_protected(vnode->permit_cache, lockdep_is_held(&vnode->lock)); RCU_INIT_POINTER(vnode->permit_cache, NULL); - vnode->cb_break++; spin_unlock(&vnode->lock); - if (permits) - afs_put_permits(permits); + afs_put_permits(permits); } /* @@ -118,10 +174,10 @@ static void afs_hash_permits(struct afs_permits *permits) * as the ACL *may* have changed. */ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, - unsigned int cb_break) + unsigned int cb_break, struct afs_status_cb *scb) { struct afs_permits *permits, *xpermits, *replacement, *zap, *new = NULL; - afs_access_t caller_access = READ_ONCE(vnode->status.caller_access); + afs_access_t caller_access = scb->status.caller_access; size_t size = 0; bool changed = false; int i, j; @@ -147,8 +203,7 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, break; } - if (afs_cb_is_broken(cb_break, vnode, - vnode->cb_interest)) { + if (afs_cb_is_broken(cb_break, vnode)) { changed = true; break; } @@ -178,7 +233,7 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, } } - if (afs_cb_is_broken(cb_break, vnode, vnode->cb_interest)) + if (afs_cb_is_broken(cb_break, vnode)) goto someone_else_changed_it; /* We need a ref on any permits list we want to copy as we'll have to @@ -197,8 +252,7 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, * yet. */ size++; - new = kzalloc(sizeof(struct afs_permits) + - sizeof(struct afs_permit) * size, GFP_NOFS); + new = kzalloc(struct_size(new, permits, size), GFP_NOFS); if (!new) goto out_put; @@ -255,14 +309,15 @@ found: kfree(new); + rcu_read_lock(); spin_lock(&vnode->lock); zap = rcu_access_pointer(vnode->permit_cache); - if (!afs_cb_is_broken(cb_break, vnode, vnode->cb_interest) && - zap == permits) + if (!afs_cb_is_broken(cb_break, vnode) && zap == permits) rcu_assign_pointer(vnode->permit_cache, replacement); else zap = replacement; spin_unlock(&vnode->lock); + rcu_read_unlock(); afs_put_permits(zap); out_put: afs_put_permits(permits); @@ -278,6 +333,40 @@ someone_else_changed_it: return; } +static bool afs_check_permit_rcu(struct afs_vnode *vnode, struct key *key, + afs_access_t *_access) +{ + const struct afs_permits *permits; + int i; + + _enter("{%llx:%llu},%x", + vnode->fid.vid, vnode->fid.vnode, key_serial(key)); + + /* check the permits to see if we've got one yet */ + if (key == vnode->volume->cell->anonymous_key) { + *_access = vnode->status.anon_access; + _leave(" = t [anon %x]", *_access); + return true; + } + + permits = rcu_dereference(vnode->permit_cache); + if (permits) { + for (i = 0; i < permits->nr_permits; i++) { + if (permits->permits[i].key < key) + continue; + if (permits->permits[i].key > key) + break; + + *_access = permits->permits[i].access; + _leave(" = %u [perm %x]", !permits->invalidated, *_access); + return !permits->invalidated; + } + } + + _leave(" = f"); + return false; +} + /* * check with the fileserver to see if the directory or parent directory is * permitted to be accessed with this authorisation, and if so, what access it @@ -322,13 +411,12 @@ int afs_check_permit(struct afs_vnode *vnode, struct key *key, */ _debug("no valid permit"); - ret = afs_fetch_status(vnode, key, false); + ret = afs_fetch_status(vnode, key, false, _access); if (ret < 0) { *_access = 0; _leave(" = %d", ret); return ret; } - *_access = vnode->status.caller_access; } _leave(" = 0 [access %x]", *_access); @@ -340,38 +428,48 @@ int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask) +int afs_permission(struct mnt_idmap *idmap, struct inode *inode, + int mask) { struct afs_vnode *vnode = AFS_FS_I(inode); - afs_access_t uninitialized_var(access); + afs_access_t access; struct key *key; - int ret; - - if (mask & MAY_NOT_BLOCK) - return -ECHILD; + int ret = 0; _enter("{{%llx:%llu},%lx},%x,", vnode->fid.vid, vnode->fid.vnode, vnode->flags, mask); - key = afs_request_key(vnode->volume->cell); - if (IS_ERR(key)) { - _leave(" = %ld [key]", PTR_ERR(key)); - return PTR_ERR(key); - } + if (mask & MAY_NOT_BLOCK) { + key = afs_request_key_rcu(vnode->volume->cell); + if (IS_ERR_OR_NULL(key)) + return -ECHILD; - ret = afs_validate(vnode, key); - if (ret < 0) - goto error; + ret = -ECHILD; + if (!afs_check_validity(vnode) || + !afs_check_permit_rcu(vnode, key, &access)) + goto error; + } else { + key = afs_request_key(vnode->volume->cell); + if (IS_ERR(key)) { + _leave(" = %ld [key]", PTR_ERR(key)); + return PTR_ERR(key); + } - /* check the permits to see if we've got one yet */ - ret = afs_check_permit(vnode, key, &access); - if (ret < 0) - goto error; + ret = afs_validate(vnode, key); + if (ret < 0) + goto error; + + /* check the permits to see if we've got one yet */ + ret = afs_check_permit(vnode, key, &access); + if (ret < 0) + goto error; + } /* interpret the access mask */ _debug("REQ %x ACC %x on %s", mask, access, S_ISDIR(inode->i_mode) ? "dir" : "file"); + ret = 0; if (S_ISDIR(inode->i_mode)) { if (mask & (MAY_EXEC | MAY_READ | MAY_CHDIR)) { if (!(access & AFS_ACE_LOOKUP)) diff --git a/fs/afs/server.c b/fs/afs/server.c index 642afa2e9783..c4428ebddb1d 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS server record management * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/sched.h> @@ -16,158 +12,78 @@ #include "protocol_yfs.h" static unsigned afs_server_gc_delay = 10; /* Server record timeout in seconds */ -static unsigned afs_server_update_delay = 30; /* Time till VLDB recheck in secs */ +static atomic_t afs_server_debug_id; -static void afs_inc_servers_outstanding(struct afs_net *net) -{ - atomic_inc(&net->servers_outstanding); -} - -static void afs_dec_servers_outstanding(struct afs_net *net) -{ - if (atomic_dec_and_test(&net->servers_outstanding)) - wake_up_var(&net->servers_outstanding); -} +static void __afs_put_server(struct afs_net *, struct afs_server *); +static void afs_server_timer(struct timer_list *timer); +static void afs_server_destroyer(struct work_struct *work); /* * Find a server by one of its addresses. */ -struct afs_server *afs_find_server(struct afs_net *net, - const struct sockaddr_rxrpc *srx) +struct afs_server *afs_find_server(const struct rxrpc_peer *peer) { - const struct sockaddr_in6 *a = &srx->transport.sin6, *b; - const struct afs_addr_list *alist; - struct afs_server *server = NULL; - unsigned int i; - bool ipv6 = true; - int seq = 0, diff; - - if (srx->transport.sin6.sin6_addr.s6_addr32[0] == 0 || - srx->transport.sin6.sin6_addr.s6_addr32[1] == 0 || - srx->transport.sin6.sin6_addr.s6_addr32[2] == htonl(0xffff)) - ipv6 = false; - - rcu_read_lock(); - - do { - if (server) - afs_put_server(net, server); - server = NULL; - read_seqbegin_or_lock(&net->fs_addr_lock, &seq); - - if (ipv6) { - hlist_for_each_entry_rcu(server, &net->fs_addresses6, addr6_link) { - alist = rcu_dereference(server->addresses); - for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) { - b = &alist->addrs[i].transport.sin6; - diff = ((u16 __force)a->sin6_port - - (u16 __force)b->sin6_port); - if (diff == 0) - diff = memcmp(&a->sin6_addr, - &b->sin6_addr, - sizeof(struct in6_addr)); - if (diff == 0) - goto found; - } - } - } else { - hlist_for_each_entry_rcu(server, &net->fs_addresses4, addr4_link) { - alist = rcu_dereference(server->addresses); - for (i = 0; i < alist->nr_ipv4; i++) { - b = &alist->addrs[i].transport.sin6; - diff = ((u16 __force)a->sin6_port - - (u16 __force)b->sin6_port); - if (diff == 0) - diff = ((u32 __force)a->sin6_addr.s6_addr32[3] - - (u32 __force)b->sin6_addr.s6_addr32[3]); - if (diff == 0) - goto found; - } - } - } + struct afs_server *server = (struct afs_server *)rxrpc_kernel_get_peer_data(peer); - server = NULL; - found: - if (server && !atomic_inc_not_zero(&server->usage)) - server = NULL; - - } while (need_seqretry(&net->fs_addr_lock, seq)); - - done_seqretry(&net->fs_addr_lock, seq); - - rcu_read_unlock(); - return server; + if (!server) + return NULL; + return afs_use_server(server, false, afs_server_trace_use_cm_call); } /* - * Look up a server by its UUID + * Look up a server by its UUID and mark it active. The caller must hold + * cell->fs_lock. */ -struct afs_server *afs_find_server_by_uuid(struct afs_net *net, const uuid_t *uuid) +static struct afs_server *afs_find_server_by_uuid(struct afs_cell *cell, const uuid_t *uuid) { - struct afs_server *server = NULL; + struct afs_server *server; struct rb_node *p; - int diff, seq = 0; + int diff; _enter("%pU", uuid); - do { - /* Unfortunately, rbtree walking doesn't give reliable results - * under just the RCU read lock, so we have to check for - * changes. - */ - if (server) - afs_put_server(net, server); - server = NULL; - - read_seqbegin_or_lock(&net->fs_lock, &seq); - - p = net->fs_servers.rb_node; - while (p) { - server = rb_entry(p, struct afs_server, uuid_rb); - - diff = memcmp(uuid, &server->uuid, sizeof(*uuid)); - if (diff < 0) { - p = p->rb_left; - } else if (diff > 0) { - p = p->rb_right; - } else { - afs_get_server(server); - break; - } + p = cell->fs_servers.rb_node; + while (p) { + server = rb_entry(p, struct afs_server, uuid_rb); - server = NULL; + diff = memcmp(uuid, &server->uuid, sizeof(*uuid)); + if (diff < 0) { + p = p->rb_left; + } else if (diff > 0) { + p = p->rb_right; + } else { + if (test_bit(AFS_SERVER_FL_UNCREATED, &server->flags)) + return NULL; /* Need a write lock */ + afs_use_server(server, true, afs_server_trace_use_by_uuid); + return server; } - } while (need_seqretry(&net->fs_lock, seq)); - - done_seqretry(&net->fs_lock, seq); + } - _leave(" = %p", server); - return server; + return NULL; } /* - * Install a server record in the namespace tree + * Install a server record in the cell tree. The caller must hold an exclusive + * lock on cell->fs_lock. */ -static struct afs_server *afs_install_server(struct afs_net *net, - struct afs_server *candidate) +static struct afs_server *afs_install_server(struct afs_cell *cell, + struct afs_server **candidate) { - const struct afs_addr_list *alist; struct afs_server *server; + struct afs_net *net = cell->net; struct rb_node **pp, *p; - int ret = -EEXIST, diff; + int diff; _enter("%p", candidate); - write_seqlock(&net->fs_lock); - /* Firstly install the server in the UUID lookup tree */ - pp = &net->fs_servers.rb_node; + pp = &cell->fs_servers.rb_node; p = NULL; while (*pp) { p = *pp; _debug("- consider %p", p); server = rb_entry(p, struct afs_server, uuid_rb); - diff = memcmp(&candidate->uuid, &server->uuid, sizeof(uuid_t)); + diff = memcmp(&(*candidate)->uuid, &server->uuid, sizeof(uuid_t)); if (diff < 0) pp = &(*pp)->rb_left; else if (diff > 0) @@ -176,90 +92,77 @@ static struct afs_server *afs_install_server(struct afs_net *net, goto exists; } - server = candidate; + server = *candidate; + *candidate = NULL; rb_link_node(&server->uuid_rb, p, pp); - rb_insert_color(&server->uuid_rb, &net->fs_servers); + rb_insert_color(&server->uuid_rb, &cell->fs_servers); + write_seqlock(&net->fs_lock); hlist_add_head_rcu(&server->proc_link, &net->fs_proc); + write_sequnlock(&net->fs_lock); - write_seqlock(&net->fs_addr_lock); - alist = rcu_dereference_protected(server->addresses, - lockdep_is_held(&net->fs_addr_lock.lock)); - - /* Secondly, if the server has any IPv4 and/or IPv6 addresses, install - * it in the IPv4 and/or IPv6 reverse-map lists. - * - * TODO: For speed we want to use something other than a flat list - * here; even sorting the list in terms of lowest address would help a - * bit, but anything we might want to do gets messy and memory - * intensive. - */ - if (alist->nr_ipv4 > 0) - hlist_add_head_rcu(&server->addr4_link, &net->fs_addresses4); - if (alist->nr_addrs > alist->nr_ipv4) - hlist_add_head_rcu(&server->addr6_link, &net->fs_addresses6); - - write_sequnlock(&net->fs_addr_lock); - ret = 0; + afs_get_cell(cell, afs_cell_trace_get_server); exists: - afs_get_server(server); - write_sequnlock(&net->fs_lock); + afs_use_server(server, true, afs_server_trace_use_install); return server; } /* - * allocate a new server record + * Allocate a new server record and mark it as active but uncreated. */ -static struct afs_server *afs_alloc_server(struct afs_net *net, - const uuid_t *uuid, - struct afs_addr_list *alist) +static struct afs_server *afs_alloc_server(struct afs_cell *cell, const uuid_t *uuid) { struct afs_server *server; + struct afs_net *net = cell->net; _enter(""); server = kzalloc(sizeof(struct afs_server), GFP_KERNEL); if (!server) - goto enomem; + return NULL; - atomic_set(&server->usage, 1); - RCU_INIT_POINTER(server->addresses, alist); - server->addr_version = alist->version; + refcount_set(&server->ref, 1); + atomic_set(&server->active, 0); + __set_bit(AFS_SERVER_FL_UNCREATED, &server->flags); + server->debug_id = atomic_inc_return(&afs_server_debug_id); server->uuid = *uuid; - server->flags = (1UL << AFS_SERVER_FL_NEW); - server->update_at = ktime_get_real_seconds() + afs_server_update_delay; rwlock_init(&server->fs_lock); - INIT_HLIST_HEAD(&server->cb_volumes); - rwlock_init(&server->cb_break_lock); + INIT_WORK(&server->destroyer, &afs_server_destroyer); + timer_setup(&server->timer, afs_server_timer, 0); + INIT_LIST_HEAD(&server->volumes); init_waitqueue_head(&server->probe_wq); + mutex_init(&server->cm_token_lock); + INIT_LIST_HEAD(&server->probe_link); + INIT_HLIST_NODE(&server->proc_link); spin_lock_init(&server->probe_lock); + server->cell = cell; + server->rtt = UINT_MAX; + server->service_id = FS_SERVICE; + server->probe_counter = 1; + server->probed_at = jiffies - LONG_MAX / 2; afs_inc_servers_outstanding(net); _leave(" = %p", server); return server; - -enomem: - _leave(" = NULL [nomem]"); - return NULL; } /* * Look up an address record for a server */ -static struct afs_addr_list *afs_vl_lookup_addrs(struct afs_cell *cell, - struct key *key, const uuid_t *uuid) +static struct afs_addr_list *afs_vl_lookup_addrs(struct afs_server *server, + struct key *key) { struct afs_vl_cursor vc; struct afs_addr_list *alist = NULL; int ret; ret = -ERESTARTSYS; - if (afs_begin_vlserver_operation(&vc, cell, key)) { + if (afs_begin_vlserver_operation(&vc, server->cell, key)) { while (afs_select_vlserver(&vc)) { if (test_bit(AFS_VLSERVER_FL_IS_YFS, &vc.server->flags)) - alist = afs_yfsvl_get_endpoints(&vc, uuid); + alist = afs_yfsvl_get_endpoints(&vc, &server->uuid); else - alist = afs_vl_get_addrs_u(&vc, uuid); + alist = afs_vl_get_addrs_u(&vc, &server->uuid); } ret = afs_end_vlserver_operation(&vc); @@ -269,242 +172,365 @@ static struct afs_addr_list *afs_vl_lookup_addrs(struct afs_cell *cell, } /* - * Get or create a fileserver record. + * Get or create a fileserver record and return it with an active-use count on + * it. */ struct afs_server *afs_lookup_server(struct afs_cell *cell, struct key *key, - const uuid_t *uuid) + const uuid_t *uuid, u32 addr_version) { - struct afs_addr_list *alist; - struct afs_server *server, *candidate; + struct afs_addr_list *alist = NULL; + struct afs_server *server, *candidate = NULL; + bool creating = false; + int ret; _enter("%p,%pU", cell->net, uuid); - server = afs_find_server_by_uuid(cell->net, uuid); - if (server) + down_read(&cell->fs_lock); + server = afs_find_server_by_uuid(cell, uuid); + /* Won't see servers marked uncreated. */ + up_read(&cell->fs_lock); + + if (server) { + timer_delete_sync(&server->timer); + if (test_bit(AFS_SERVER_FL_CREATING, &server->flags)) + goto wait_for_creation; + if (server->addr_version != addr_version) + set_bit(AFS_SERVER_FL_NEEDS_UPDATE, &server->flags); return server; + } - alist = afs_vl_lookup_addrs(cell, key, uuid); - if (IS_ERR(alist)) - return ERR_CAST(alist); - - candidate = afs_alloc_server(cell->net, uuid, alist); + candidate = afs_alloc_server(cell, uuid); if (!candidate) { - afs_put_addrlist(alist); + afs_put_addrlist(alist, afs_alist_trace_put_server_oom); return ERR_PTR(-ENOMEM); } - server = afs_install_server(cell->net, candidate); - if (server != candidate) { - afs_put_addrlist(alist); + down_write(&cell->fs_lock); + server = afs_install_server(cell, &candidate); + if (test_bit(AFS_SERVER_FL_CREATING, &server->flags)) { + /* We need to wait for creation to complete. */ + up_write(&cell->fs_lock); + goto wait_for_creation; + } + if (test_bit(AFS_SERVER_FL_UNCREATED, &server->flags)) { + set_bit(AFS_SERVER_FL_CREATING, &server->flags); + clear_bit(AFS_SERVER_FL_UNCREATED, &server->flags); + creating = true; + } + up_write(&cell->fs_lock); + timer_delete_sync(&server->timer); + + /* If we get to create the server, we look up the addresses and then + * immediately dispatch an asynchronous probe to each interface on the + * fileserver. This will make sure the repeat-probing service is + * started. + */ + if (creating) { + alist = afs_vl_lookup_addrs(server, key); + if (IS_ERR(alist)) { + ret = PTR_ERR(alist); + goto create_failed; + } + + ret = afs_fs_probe_fileserver(cell->net, server, alist, key); + if (ret) + goto create_failed; + + clear_and_wake_up_bit(AFS_SERVER_FL_CREATING, &server->flags); + } + +out: + afs_put_addrlist(alist, afs_alist_trace_put_server_create); + if (candidate) { + kfree(rcu_access_pointer(server->endpoint_state)); kfree(candidate); + afs_dec_servers_outstanding(cell->net); + } + return server ?: ERR_PTR(ret); + +wait_for_creation: + afs_see_server(server, afs_server_trace_wait_create); + wait_on_bit(&server->flags, AFS_SERVER_FL_CREATING, TASK_UNINTERRUPTIBLE); + if (test_bit_acquire(AFS_SERVER_FL_UNCREATED, &server->flags)) { + /* Barrier: read flag before error */ + ret = READ_ONCE(server->create_error); + afs_put_server(cell->net, server, afs_server_trace_unuse_create_fail); + server = NULL; + goto out; } - _leave(" = %p{%d}", server, atomic_read(&server->usage)); - return server; + ret = 0; + goto out; + +create_failed: + down_write(&cell->fs_lock); + + WRITE_ONCE(server->create_error, ret); + smp_wmb(); /* Barrier: set error before flag. */ + set_bit(AFS_SERVER_FL_UNCREATED, &server->flags); + + clear_and_wake_up_bit(AFS_SERVER_FL_CREATING, &server->flags); + + if (test_bit(AFS_SERVER_FL_UNCREATED, &server->flags)) { + clear_bit(AFS_SERVER_FL_UNCREATED, &server->flags); + creating = true; + } + afs_unuse_server(cell->net, server, afs_server_trace_unuse_create_fail); + server = NULL; + + up_write(&cell->fs_lock); + goto out; } /* - * Set the server timer to fire after a given delay, assuming it's not already - * set for an earlier time. + * Set/reduce a server's timer. */ -static void afs_set_server_timer(struct afs_net *net, time64_t delay) +static void afs_set_server_timer(struct afs_server *server, unsigned int delay_secs) { - if (net->live) { - afs_inc_servers_outstanding(net); - if (timer_reduce(&net->fs_timer, jiffies + delay * HZ)) - afs_dec_servers_outstanding(net); - } + mod_timer(&server->timer, jiffies + delay_secs * HZ); } /* - * Server management timer. We have an increment on fs_outstanding that we - * need to pass along to the work item. + * Get a reference on a server object. */ -void afs_servers_timer(struct timer_list *timer) +struct afs_server *afs_get_server(struct afs_server *server, + enum afs_server_trace reason) { - struct afs_net *net = container_of(timer, struct afs_net, fs_timer); + unsigned int a; + int r; - _enter(""); - if (!queue_work(afs_wq, &net->fs_manager)) - afs_dec_servers_outstanding(net); + __refcount_inc(&server->ref, &r); + a = atomic_read(&server->active); + trace_afs_server(server->debug_id, r + 1, a, reason); + return server; +} + +/* + * Get an active count on a server object and maybe remove from the inactive + * list. + */ +struct afs_server *afs_use_server(struct afs_server *server, bool activate, + enum afs_server_trace reason) +{ + unsigned int a; + int r; + + __refcount_inc(&server->ref, &r); + a = atomic_inc_return(&server->active); + if (a == 1 && activate && + !test_bit(AFS_SERVER_FL_EXPIRED, &server->flags)) + timer_delete(&server->timer); + + trace_afs_server(server->debug_id, r + 1, a, reason); + return server; } /* * Release a reference on a server record. */ -void afs_put_server(struct afs_net *net, struct afs_server *server) +void afs_put_server(struct afs_net *net, struct afs_server *server, + enum afs_server_trace reason) { - unsigned int usage; + unsigned int a, debug_id; + bool zero; + int r; if (!server) return; - server->put_time = ktime_get_real_seconds(); + debug_id = server->debug_id; + a = atomic_read(&server->active); + zero = __refcount_dec_and_test(&server->ref, &r); + trace_afs_server(debug_id, r - 1, a, reason); + if (unlikely(zero)) + __afs_put_server(net, server); +} - usage = atomic_dec_return(&server->usage); +/* + * Drop an active count on a server object without updating the last-unused + * time. + */ +void afs_unuse_server_notime(struct afs_net *net, struct afs_server *server, + enum afs_server_trace reason) +{ + if (!server) + return; + + if (atomic_dec_and_test(&server->active)) { + if (test_bit(AFS_SERVER_FL_EXPIRED, &server->flags) || + READ_ONCE(server->cell->state) >= AFS_CELL_REMOVING) + schedule_work(&server->destroyer); + } - _enter("{%u}", usage); + afs_put_server(net, server, reason); +} - if (likely(usage > 0)) +/* + * Drop an active count on a server object. + */ +void afs_unuse_server(struct afs_net *net, struct afs_server *server, + enum afs_server_trace reason) +{ + if (!server) return; - afs_set_server_timer(net, afs_server_gc_delay); + if (atomic_dec_and_test(&server->active)) { + if (!test_bit(AFS_SERVER_FL_EXPIRED, &server->flags) && + READ_ONCE(server->cell->state) < AFS_CELL_REMOVING) { + time64_t unuse_time = ktime_get_real_seconds(); + + server->unuse_time = unuse_time; + afs_set_server_timer(server, afs_server_gc_delay); + } else { + schedule_work(&server->destroyer); + } + } + + afs_put_server(net, server, reason); } static void afs_server_rcu(struct rcu_head *rcu) { struct afs_server *server = container_of(rcu, struct afs_server, rcu); - afs_put_addrlist(rcu_access_pointer(server->addresses)); + trace_afs_server(server->debug_id, refcount_read(&server->ref), + atomic_read(&server->active), afs_server_trace_free); + afs_put_endpoint_state(rcu_access_pointer(server->endpoint_state), + afs_estate_trace_put_server); + afs_put_cell(server->cell, afs_cell_trace_put_server); + kfree(server->cm_rxgk_appdata.data); kfree(server); } -/* - * destroy a dead server - */ -static void afs_destroy_server(struct afs_net *net, struct afs_server *server) +static void __afs_put_server(struct afs_net *net, struct afs_server *server) { - struct afs_addr_list *alist = rcu_access_pointer(server->addresses); - struct afs_addr_cursor ac = { - .alist = alist, - .index = alist->preferred, - .error = 0, - }; - _enter("%p", server); - - if (test_bit(AFS_SERVER_FL_MAY_HAVE_CB, &server->flags)) - afs_fs_give_up_all_callbacks(net, server, &ac, NULL); - - wait_var_event(&server->probe_outstanding, - atomic_read(&server->probe_outstanding) == 0); - call_rcu(&server->rcu, afs_server_rcu); afs_dec_servers_outstanding(net); } +static void afs_give_up_callbacks(struct afs_net *net, struct afs_server *server) +{ + struct afs_endpoint_state *estate = rcu_access_pointer(server->endpoint_state); + struct afs_addr_list *alist = estate->addresses; + + afs_fs_give_up_all_callbacks(net, server, &alist->addrs[alist->preferred], NULL); +} + /* - * Garbage collect any expired servers. + * Check to see if the server record has expired. */ -static void afs_gc_servers(struct afs_net *net, struct afs_server *gc_list) +static bool afs_has_server_expired(const struct afs_server *server) { - struct afs_server *server; - bool deleted; - int usage; - - while ((server = gc_list)) { - gc_list = server->gc_next; - - write_seqlock(&net->fs_lock); - usage = 1; - deleted = atomic_try_cmpxchg(&server->usage, &usage, 0); - if (deleted) { - rb_erase(&server->uuid_rb, &net->fs_servers); - hlist_del_rcu(&server->proc_link); - } - write_sequnlock(&net->fs_lock); - - if (deleted) { - write_seqlock(&net->fs_addr_lock); - if (!hlist_unhashed(&server->addr4_link)) - hlist_del_rcu(&server->addr4_link); - if (!hlist_unhashed(&server->addr6_link)) - hlist_del_rcu(&server->addr6_link); - write_sequnlock(&net->fs_addr_lock); - afs_destroy_server(net, server); - } + time64_t expires_at; + + if (atomic_read(&server->active)) + return false; + + if (server->cell->net->live || + server->cell->state >= AFS_CELL_REMOVING) { + trace_afs_server(server->debug_id, refcount_read(&server->ref), + 0, afs_server_trace_purging); + return true; } + + expires_at = server->unuse_time; + if (!test_bit(AFS_SERVER_FL_VL_FAIL, &server->flags) && + !test_bit(AFS_SERVER_FL_NOT_FOUND, &server->flags)) + expires_at += afs_server_gc_delay; + + return ktime_get_real_seconds() > expires_at; } /* - * Manage the records of servers known to be within a network namespace. This - * includes garbage collecting unused servers. - * - * Note also that we were given an increment on net->servers_outstanding by - * whoever queued us that we need to deal with before returning. + * Remove a server record from it's parent cell's database. */ -void afs_manage_servers(struct work_struct *work) +static bool afs_remove_server_from_cell(struct afs_server *server) { - struct afs_net *net = container_of(work, struct afs_net, fs_manager); - struct afs_server *gc_list = NULL; - struct rb_node *cursor; - time64_t now = ktime_get_real_seconds(), next_manage = TIME64_MAX; - bool purging = !net->live; + struct afs_cell *cell = server->cell; - _enter(""); + down_write(&cell->fs_lock); - /* Trawl the server list looking for servers that have expired from - * lack of use. - */ - read_seqlock_excl(&net->fs_lock); - - for (cursor = rb_first(&net->fs_servers); cursor; cursor = rb_next(cursor)) { - struct afs_server *server = - rb_entry(cursor, struct afs_server, uuid_rb); - int usage = atomic_read(&server->usage); - - _debug("manage %pU %u", &server->uuid, usage); - - ASSERTCMP(usage, >=, 1); - ASSERTIFCMP(purging, usage, ==, 1); - - if (usage == 1) { - time64_t expire_at = server->put_time; - - if (!test_bit(AFS_SERVER_FL_VL_FAIL, &server->flags) && - !test_bit(AFS_SERVER_FL_NOT_FOUND, &server->flags)) - expire_at += afs_server_gc_delay; - if (purging || expire_at <= now) { - server->gc_next = gc_list; - gc_list = server; - } else if (expire_at < next_manage) { - next_manage = expire_at; - } - } + if (!afs_has_server_expired(server)) { + up_write(&cell->fs_lock); + return false; } - read_sequnlock_excl(&net->fs_lock); + set_bit(AFS_SERVER_FL_EXPIRED, &server->flags); + _debug("expire %pU %u", &server->uuid, atomic_read(&server->active)); + afs_see_server(server, afs_server_trace_see_expired); + rb_erase(&server->uuid_rb, &cell->fs_servers); + up_write(&cell->fs_lock); + return true; +} - /* Update the timer on the way out. We have to pass an increment on - * servers_outstanding in the namespace that we are in to the timer or - * the work scheduler. - */ - if (!purging && next_manage < TIME64_MAX) { - now = ktime_get_real_seconds(); +static void afs_server_destroyer(struct work_struct *work) +{ + struct afs_endpoint_state *estate; + struct afs_server *server = container_of(work, struct afs_server, destroyer); + struct afs_net *net = server->cell->net; - if (next_manage - now <= 0) { - if (queue_work(afs_wq, &net->fs_manager)) - afs_inc_servers_outstanding(net); - } else { - afs_set_server_timer(net, next_manage - now); - } - } + afs_see_server(server, afs_server_trace_see_destroyer); - afs_gc_servers(net, gc_list); + if (test_bit(AFS_SERVER_FL_EXPIRED, &server->flags)) + return; - afs_dec_servers_outstanding(net); - _leave(" [%d]", atomic_read(&net->servers_outstanding)); + if (!afs_remove_server_from_cell(server)) + return; + + timer_shutdown_sync(&server->timer); + cancel_work(&server->destroyer); + + if (test_bit(AFS_SERVER_FL_MAY_HAVE_CB, &server->flags)) + afs_give_up_callbacks(net, server); + + /* Unbind the rxrpc_peer records from the server. */ + estate = rcu_access_pointer(server->endpoint_state); + if (estate) + afs_set_peer_appdata(server, estate->addresses, NULL); + + write_seqlock(&net->fs_lock); + list_del_init(&server->probe_link); + if (!hlist_unhashed(&server->proc_link)) + hlist_del_rcu(&server->proc_link); + write_sequnlock(&net->fs_lock); + + afs_put_server(net, server, afs_server_trace_destroy); } -static void afs_queue_server_manager(struct afs_net *net) +static void afs_server_timer(struct timer_list *timer) { - afs_inc_servers_outstanding(net); - if (!queue_work(afs_wq, &net->fs_manager)) - afs_dec_servers_outstanding(net); + struct afs_server *server = container_of(timer, struct afs_server, timer); + + afs_see_server(server, afs_server_trace_see_timer); + if (!test_bit(AFS_SERVER_FL_EXPIRED, &server->flags)) + schedule_work(&server->destroyer); } /* - * Purge list of servers. + * Wake up all the servers in a cell so that they can purge themselves. */ -void afs_purge_servers(struct afs_net *net) +void afs_purge_servers(struct afs_cell *cell) { - _enter(""); + struct afs_server *server; + struct rb_node *rb; - if (del_timer_sync(&net->fs_timer)) - atomic_dec(&net->servers_outstanding); + down_read(&cell->fs_lock); + for (rb = rb_first(&cell->fs_servers); rb; rb = rb_next(rb)) { + server = rb_entry(rb, struct afs_server, uuid_rb); + afs_see_server(server, afs_server_trace_see_purge); + schedule_work(&server->destroyer); + } + up_read(&cell->fs_lock); +} - afs_queue_server_manager(net); +/* + * Wait for outstanding servers. + */ +void afs_wait_for_servers(struct afs_net *net) +{ + _enter(""); - _debug("wait"); + atomic_dec(&net->servers_outstanding); wait_var_event(&net->servers_outstanding, !atomic_read(&net->servers_outstanding)); _leave(""); @@ -513,32 +539,43 @@ void afs_purge_servers(struct afs_net *net) /* * Get an update for a server's address list. */ -static noinline bool afs_update_server_record(struct afs_fs_cursor *fc, struct afs_server *server) +static noinline bool afs_update_server_record(struct afs_operation *op, + struct afs_server *server, + struct key *key) { - struct afs_addr_list *alist, *discard; + struct afs_endpoint_state *estate; + struct afs_addr_list *alist; + bool has_addrs; _enter(""); - alist = afs_vl_lookup_addrs(fc->vnode->volume->cell, fc->key, - &server->uuid); + trace_afs_server(server->debug_id, refcount_read(&server->ref), + atomic_read(&server->active), + afs_server_trace_update); + + alist = afs_vl_lookup_addrs(server, op->key); if (IS_ERR(alist)) { - fc->ac.error = PTR_ERR(alist); - _leave(" = f [%d]", fc->ac.error); + rcu_read_lock(); + estate = rcu_dereference(server->endpoint_state); + has_addrs = estate->addresses; + rcu_read_unlock(); + + if ((PTR_ERR(alist) == -ERESTARTSYS || + PTR_ERR(alist) == -EINTR) && + (op->flags & AFS_OPERATION_UNINTR) && + has_addrs) { + _leave(" = t [intr]"); + return true; + } + afs_op_set_error(op, PTR_ERR(alist)); + _leave(" = f [%d]", afs_op_error(op)); return false; } - discard = alist; - if (server->addr_version != alist->version) { - write_lock(&server->fs_lock); - discard = rcu_dereference_protected(server->addresses, - lockdep_is_held(&server->fs_lock)); - rcu_assign_pointer(server->addresses, alist); - server->addr_version = alist->version; - write_unlock(&server->fs_lock); - } + if (server->addr_version != alist->version) + afs_fs_probe_fileserver(op->net, server, alist, key); - server->update_at = ktime_get_real_seconds() + afs_server_update_delay; - afs_put_addrlist(discard); + afs_put_addrlist(alist, afs_alist_trace_put_server_update); _leave(" = t"); return true; } @@ -546,10 +583,9 @@ static noinline bool afs_update_server_record(struct afs_fs_cursor *fc, struct a /* * See if a server's address list needs updating. */ -bool afs_check_server_record(struct afs_fs_cursor *fc, struct afs_server *server) +bool afs_check_server_record(struct afs_operation *op, struct afs_server *server, + struct key *key) { - time64_t now = ktime_get_real_seconds(); - long diff; bool success; int ret, retries = 0; @@ -558,24 +594,29 @@ bool afs_check_server_record(struct afs_fs_cursor *fc, struct afs_server *server ASSERT(server); retry: - diff = READ_ONCE(server->update_at) - now; - if (diff > 0) { - _leave(" = t [not now %ld]", diff); - return true; - } + if (test_bit(AFS_SERVER_FL_UPDATING, &server->flags)) + goto wait; + if (test_bit(AFS_SERVER_FL_NEEDS_UPDATE, &server->flags)) + goto update; + _leave(" = t [good]"); + return true; +update: if (!test_and_set_bit_lock(AFS_SERVER_FL_UPDATING, &server->flags)) { - success = afs_update_server_record(fc, server); + clear_bit(AFS_SERVER_FL_NEEDS_UPDATE, &server->flags); + success = afs_update_server_record(op, server, key); clear_bit_unlock(AFS_SERVER_FL_UPDATING, &server->flags); wake_up_bit(&server->flags, AFS_SERVER_FL_UPDATING); _leave(" = %d", success); return success; } +wait: ret = wait_on_bit(&server->flags, AFS_SERVER_FL_UPDATING, - TASK_INTERRUPTIBLE); + (op->flags & AFS_OPERATION_UNINTR) ? + TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE); if (ret == -ERESTARTSYS) { - fc->ac.error = ret; + afs_op_set_error(op, ret); _leave(" = f [intr]"); return false; } diff --git a/fs/afs/server_list.c b/fs/afs/server_list.c index 155dc14caef9..20d5474837df 100644 --- a/fs/afs/server_list.c +++ b/fs/afs/server_list.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS fileserver list management. * * Copyright (C) 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -18,43 +14,73 @@ void afs_put_serverlist(struct afs_net *net, struct afs_server_list *slist) int i; if (slist && refcount_dec_and_test(&slist->usage)) { - for (i = 0; i < slist->nr_servers; i++) { - afs_put_cb_interest(net, slist->servers[i].cb_interest); - afs_put_server(net, slist->servers[i].server); - } - kfree(slist); + for (i = 0; i < slist->nr_servers; i++) + afs_unuse_server(net, slist->servers[i].server, + afs_server_trace_unuse_slist); + kfree_rcu(slist, rcu); } } /* * Build a server list from a VLDB record. */ -struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, +struct afs_server_list *afs_alloc_server_list(struct afs_volume *volume, struct key *key, - struct afs_vldb_entry *vldb, - u8 type_mask) + struct afs_vldb_entry *vldb) { struct afs_server_list *slist; struct afs_server *server; - int ret = -ENOMEM, nr_servers = 0, i, j; - - for (i = 0; i < vldb->nr_servers; i++) - if (vldb->fs_mask[i] & type_mask) - nr_servers++; + unsigned int type_mask = 1 << volume->type; + bool use_newrepsites = false; + int ret = -ENOMEM, nr_servers = 0, newrep = 0, i, j, usable = 0; + + /* Work out if we're going to restrict to NEWREPSITE-marked servers or + * not. If at least one site is marked as NEWREPSITE, then it's likely + * that "vos release" is busy updating RO sites. We cut over from one + * to the other when >=50% of the sites have been updated. Sites that + * are in the process of being updated are marked DONTUSE. + */ + for (i = 0; i < vldb->nr_servers; i++) { + if (!(vldb->fs_mask[i] & type_mask)) + continue; + nr_servers++; + if (vldb->vlsf_flags[i] & AFS_VLSF_DONTUSE) + continue; + usable++; + if (vldb->vlsf_flags[i] & AFS_VLSF_NEWREPSITE) + newrep++; + } slist = kzalloc(struct_size(slist, servers, nr_servers), GFP_KERNEL); if (!slist) goto error; + if (newrep) { + if (newrep < usable / 2) { + slist->ro_replicating = AFS_RO_REPLICATING_USE_OLD; + } else { + slist->ro_replicating = AFS_RO_REPLICATING_USE_NEW; + use_newrepsites = true; + } + } + refcount_set(&slist->usage, 1); rwlock_init(&slist->lock); /* Make sure a records exists for each server in the list. */ for (i = 0; i < vldb->nr_servers; i++) { + unsigned long se_flags = 0; + bool newrepsite = vldb->vlsf_flags[i] & AFS_VLSF_NEWREPSITE; + if (!(vldb->fs_mask[i] & type_mask)) continue; + if (vldb->vlsf_flags[i] & AFS_VLSF_DONTUSE) + __set_bit(AFS_SE_EXCLUDED, &se_flags); + if (newrep && (newrepsite ^ use_newrepsites)) + __set_bit(AFS_SE_EXCLUDED, &se_flags); - server = afs_lookup_server(cell, key, &vldb->fs_server[i]); + server = afs_lookup_server(volume->cell, key, &vldb->fs_server[i], + vldb->addr_version[i]); if (IS_ERR(server)) { ret = PTR_ERR(server); if (ret == -ENOENT || @@ -71,7 +97,8 @@ struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, break; if (j < slist->nr_servers) { if (slist->servers[j].server == server) { - afs_put_server(cell->net, server); + afs_unuse_server_notime(volume->cell->net, server, + afs_server_trace_unuse_slist_isort); continue; } @@ -81,6 +108,9 @@ struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, } slist->servers[j].server = server; + slist->servers[j].volume = volume; + slist->servers[j].flags = se_flags; + slist->servers[j].cb_expires_at = AFS_NO_CB_PROMISE; slist->nr_servers++; } @@ -92,7 +122,7 @@ struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, return slist; error_2: - afs_put_serverlist(cell->net, slist); + afs_put_serverlist(volume->cell->net, slist); error: return ERR_PTR(ret); } @@ -103,53 +133,117 @@ error: bool afs_annotate_server_list(struct afs_server_list *new, struct afs_server_list *old) { - struct afs_server *cur; - int i, j; + unsigned long mask = 1UL << AFS_SE_EXCLUDED; + int i; - if (old->nr_servers != new->nr_servers) + if (old->nr_servers != new->nr_servers || + old->ro_replicating != new->ro_replicating) goto changed; - for (i = 0; i < old->nr_servers; i++) + for (i = 0; i < old->nr_servers; i++) { if (old->servers[i].server != new->servers[i].server) goto changed; - + if ((old->servers[i].flags & mask) != (new->servers[i].flags & mask)) + goto changed; + } return false; - changed: - /* Maintain the same preferred server as before if possible. */ - cur = old->servers[old->preferred].server; - for (j = 0; j < new->nr_servers; j++) { - if (new->servers[j].server == cur) { - new->preferred = j; - break; + return true; +} + +/* + * Attach a volume to the servers it is going to use. + */ +void afs_attach_volume_to_servers(struct afs_volume *volume, struct afs_server_list *slist) +{ + struct afs_server_entry *se, *pe; + struct afs_server *server; + struct list_head *p; + unsigned int i; + + down_write(&volume->cell->vs_lock); + + for (i = 0; i < slist->nr_servers; i++) { + se = &slist->servers[i]; + server = se->server; + + list_for_each(p, &server->volumes) { + pe = list_entry(p, struct afs_server_entry, slink); + if (volume->vid <= pe->volume->vid) + break; } + list_add_tail(&se->slink, p); } - /* Keep the old callback interest records where possible so that we - * maintain callback interception. - */ - i = 0; - j = 0; - while (i < old->nr_servers && j < new->nr_servers) { - if (new->servers[j].server == old->servers[i].server) { - struct afs_cb_interest *cbi = old->servers[i].cb_interest; - if (cbi) { - new->servers[j].cb_interest = cbi; - refcount_inc(&cbi->usage); - } - i++; - j++; - continue; - } + slist->attached = true; + up_write(&volume->cell->vs_lock); +} - if (new->servers[j].server < old->servers[i].server) { - j++; +/* + * Reattach a volume to the servers it is going to use when server list is + * replaced. We try to switch the attachment points to avoid rewalking the + * lists. + */ +void afs_reattach_volume_to_servers(struct afs_volume *volume, struct afs_server_list *new, + struct afs_server_list *old) +{ + unsigned int n = 0, o = 0; + + down_write(&volume->cell->vs_lock); + + while (n < new->nr_servers || o < old->nr_servers) { + struct afs_server_entry *pn = n < new->nr_servers ? &new->servers[n] : NULL; + struct afs_server_entry *po = o < old->nr_servers ? &old->servers[o] : NULL; + struct afs_server_entry *s; + struct list_head *p; + int diff; + + if (pn && po && pn->server == po->server) { + pn->cb_expires_at = po->cb_expires_at; + list_replace(&po->slink, &pn->slink); + n++; + o++; continue; } - i++; - continue; + if (pn && po) + diff = memcmp(&pn->server->uuid, &po->server->uuid, + sizeof(pn->server->uuid)); + else + diff = pn ? -1 : 1; + + if (diff < 0) { + list_for_each(p, &pn->server->volumes) { + s = list_entry(p, struct afs_server_entry, slink); + if (volume->vid <= s->volume->vid) + break; + } + list_add_tail(&pn->slink, p); + n++; + } else { + list_del(&po->slink); + o++; + } } - return true; + up_write(&volume->cell->vs_lock); +} + +/* + * Detach a volume from the servers it has been using. + */ +void afs_detach_volume_from_servers(struct afs_volume *volume, struct afs_server_list *slist) +{ + unsigned int i; + + if (!slist->attached) + return; + + down_write(&volume->cell->vs_lock); + + for (i = 0; i < slist->nr_servers; i++) + list_del(&slist->servers[i].slink); + + slist->attached = false; + up_write(&volume->cell->vs_lock); } diff --git a/fs/afs/super.c b/fs/afs/super.c index dcd07fe99871..d672b7ab57ae 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -1,6 +1,6 @@ /* AFS superblock handling * - * Copyright (c) 2002, 2007 Red Hat, Inc. All rights reserved. + * Copyright (c) 2002, 2007, 2018 Red Hat, Inc. All rights reserved. * * This software may be freely redistributed under the terms of the * GNU General Public License. @@ -21,7 +21,7 @@ #include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> -#include <linux/parser.h> +#include <linux/fs_parser.h> #include <linux/statfs.h> #include <linux/sched.h> #include <linux/nsproxy.h> @@ -30,21 +30,23 @@ #include "internal.h" static void afs_i_init_once(void *foo); -static struct dentry *afs_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data); static void afs_kill_super(struct super_block *sb); static struct inode *afs_alloc_inode(struct super_block *sb); static void afs_destroy_inode(struct inode *inode); +static void afs_free_inode(struct inode *inode); static int afs_statfs(struct dentry *dentry, struct kstatfs *buf); static int afs_show_devname(struct seq_file *m, struct dentry *root); static int afs_show_options(struct seq_file *m, struct dentry *root); +static int afs_init_fs_context(struct fs_context *fc); +static const struct fs_parameter_spec afs_fs_parameters[]; struct file_system_type afs_fs_type = { - .owner = THIS_MODULE, - .name = "afs", - .mount = afs_mount, - .kill_sb = afs_kill_super, - .fs_flags = 0, + .owner = THIS_MODULE, + .name = "afs", + .init_fs_context = afs_init_fs_context, + .parameters = afs_fs_parameters, + .kill_sb = afs_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE, }; MODULE_ALIAS_FS("afs"); @@ -53,8 +55,10 @@ int afs_net_id; static const struct super_operations afs_super_ops = { .statfs = afs_statfs, .alloc_inode = afs_alloc_inode, + .write_inode = netfs_unpin_writeback, .drop_inode = afs_drop_inode, .destroy_inode = afs_destroy_inode, + .free_inode = afs_free_inode, .evict_inode = afs_evict_inode, .show_devname = afs_show_devname, .show_options = afs_show_options, @@ -63,22 +67,27 @@ static const struct super_operations afs_super_ops = { static struct kmem_cache *afs_inode_cachep; static atomic_t afs_count_active_inodes; -enum { - afs_no_opt, - afs_opt_cell, - afs_opt_dyn, - afs_opt_rwpath, - afs_opt_vol, - afs_opt_autocell, +enum afs_param { + Opt_autocell, + Opt_dyn, + Opt_flock, + Opt_source, }; -static const match_table_t afs_options_list = { - { afs_opt_cell, "cell=%s" }, - { afs_opt_dyn, "dyn" }, - { afs_opt_rwpath, "rwpath" }, - { afs_opt_vol, "vol=%s" }, - { afs_opt_autocell, "autocell" }, - { afs_no_opt, NULL }, +static const struct constant_table afs_param_flock[] = { + {"local", afs_flock_mode_local }, + {"openafs", afs_flock_mode_openafs }, + {"strict", afs_flock_mode_strict }, + {"write", afs_flock_mode_write }, + {} +}; + +static const struct fs_parameter_spec afs_fs_parameters[] = { + fsparam_flag ("autocell", Opt_autocell), + fsparam_flag ("dyn", Opt_dyn), + fsparam_enum ("flock", Opt_flock, afs_param_flock), + fsparam_string("source", Opt_source), + {} }; /* @@ -181,166 +190,226 @@ static int afs_show_devname(struct seq_file *m, struct dentry *root) static int afs_show_options(struct seq_file *m, struct dentry *root) { struct afs_super_info *as = AFS_FS_S(root->d_sb); + const char *p = NULL; if (as->dyn_root) seq_puts(m, ",dyn"); - if (test_bit(AFS_VNODE_AUTOCELL, &AFS_FS_I(d_inode(root))->flags)) - seq_puts(m, ",autocell"); - return 0; -} - -/* - * parse the mount options - * - this function has been shamelessly adapted from the ext3 fs which - * shamelessly adapted it from the msdos fs - */ -static int afs_parse_options(struct afs_mount_params *params, - char *options, const char **devname) -{ - struct afs_cell *cell; - substring_t args[MAX_OPT_ARGS]; - char *p; - int token; - - _enter("%s", options); - - options[PAGE_SIZE - 1] = 0; - - while ((p = strsep(&options, ","))) { - if (!*p) - continue; - - token = match_token(p, afs_options_list, args); - switch (token) { - case afs_opt_cell: - rcu_read_lock(); - cell = afs_lookup_cell_rcu(params->net, - args[0].from, - args[0].to - args[0].from); - rcu_read_unlock(); - if (IS_ERR(cell)) - return PTR_ERR(cell); - afs_put_cell(params->net, params->cell); - params->cell = cell; - break; - - case afs_opt_rwpath: - params->rwpath = true; - break; - - case afs_opt_vol: - *devname = args[0].from; - break; - - case afs_opt_autocell: - params->autocell = true; - break; - - case afs_opt_dyn: - params->dyn_root = true; - break; - - default: - printk(KERN_ERR "kAFS:" - " Unknown or invalid mount option: '%s'\n", p); - return -EINVAL; - } + switch (as->flock_mode) { + case afs_flock_mode_unset: break; + case afs_flock_mode_local: p = "local"; break; + case afs_flock_mode_openafs: p = "openafs"; break; + case afs_flock_mode_strict: p = "strict"; break; + case afs_flock_mode_write: p = "write"; break; } + if (p) + seq_printf(m, ",flock=%s", p); - _leave(" = 0"); return 0; } /* - * parse a device name to get cell name, volume name, volume type and R/W - * selector - * - this can be one of the following: + * Parse the source name to get cell name, volume name, volume type and R/W + * selector. + * + * This can be one of the following: * "%[cell:]volume[.]" R/W volume - * "#[cell:]volume[.]" R/O or R/W volume (rwpath=0), - * or R/W (rwpath=1) volume + * "#[cell:]volume[.]" R/O or R/W volume (R/O parent), + * or R/W (R/W parent) volume * "%[cell:]volume.readonly" R/O volume * "#[cell:]volume.readonly" R/O volume * "%[cell:]volume.backup" Backup volume * "#[cell:]volume.backup" Backup volume */ -static int afs_parse_device_name(struct afs_mount_params *params, - const char *name) +static int afs_parse_source(struct fs_context *fc, struct fs_parameter *param) { + struct afs_fs_context *ctx = fc->fs_private; struct afs_cell *cell; - const char *cellname, *suffix; + const char *cellname, *suffix, *name = param->string; int cellnamesz; _enter(",%s", name); + if (fc->source) + return invalf(fc, "kAFS: Multiple sources not supported"); + if (!name) { printk(KERN_ERR "kAFS: no volume name specified\n"); return -EINVAL; } if ((name[0] != '%' && name[0] != '#') || !name[1]) { + /* To use dynroot, we don't want to have to provide a source */ + if (strcmp(name, "none") == 0) { + ctx->no_cell = true; + return 0; + } printk(KERN_ERR "kAFS: unparsable volume name\n"); return -EINVAL; } /* determine the type of volume we're looking for */ - params->type = AFSVL_ROVOL; - params->force = false; - if (params->rwpath || name[0] == '%') { - params->type = AFSVL_RWVOL; - params->force = true; + if (name[0] == '%') { + ctx->type = AFSVL_RWVOL; + ctx->force = true; } name++; /* split the cell name out if there is one */ - params->volname = strchr(name, ':'); - if (params->volname) { + ctx->volname = strchr(name, ':'); + if (ctx->volname) { cellname = name; - cellnamesz = params->volname - name; - params->volname++; + cellnamesz = ctx->volname - name; + ctx->volname++; } else { - params->volname = name; + ctx->volname = name; cellname = NULL; cellnamesz = 0; } /* the volume type is further affected by a possible suffix */ - suffix = strrchr(params->volname, '.'); + suffix = strrchr(ctx->volname, '.'); if (suffix) { if (strcmp(suffix, ".readonly") == 0) { - params->type = AFSVL_ROVOL; - params->force = true; + ctx->type = AFSVL_ROVOL; + ctx->force = true; } else if (strcmp(suffix, ".backup") == 0) { - params->type = AFSVL_BACKVOL; - params->force = true; + ctx->type = AFSVL_BACKVOL; + ctx->force = true; } else if (suffix[1] == 0) { } else { suffix = NULL; } } - params->volnamesz = suffix ? - suffix - params->volname : strlen(params->volname); + ctx->volnamesz = suffix ? + suffix - ctx->volname : strlen(ctx->volname); _debug("cell %*.*s [%p]", - cellnamesz, cellnamesz, cellname ?: "", params->cell); + cellnamesz, cellnamesz, cellname ?: "", ctx->cell); /* lookup the cell record */ - if (cellname || !params->cell) { - cell = afs_lookup_cell(params->net, cellname, cellnamesz, - NULL, false); + if (cellname) { + cell = afs_lookup_cell(ctx->net, cellname, cellnamesz, + NULL, AFS_LOOKUP_CELL_DIRECT_MOUNT, + afs_cell_trace_use_lookup_mount); if (IS_ERR(cell)) { - printk(KERN_ERR "kAFS: unable to lookup cell '%*.*s'\n", + pr_err("kAFS: unable to lookup cell '%*.*s'\n", cellnamesz, cellnamesz, cellname ?: ""); return PTR_ERR(cell); } - afs_put_cell(params->net, params->cell); - params->cell = cell; + afs_unuse_cell(ctx->cell, afs_cell_trace_unuse_parse); + afs_see_cell(cell, afs_cell_trace_see_source); + ctx->cell = cell; } _debug("CELL:%s [%p] VOLUME:%*.*s SUFFIX:%s TYPE:%d%s", - params->cell->name, params->cell, - params->volnamesz, params->volnamesz, params->volname, - suffix ?: "-", params->type, params->force ? " FORCE" : ""); + ctx->cell->name, ctx->cell, + ctx->volnamesz, ctx->volnamesz, ctx->volname, + suffix ?: "-", ctx->type, ctx->force ? " FORCE" : ""); + + fc->source = param->string; + param->string = NULL; + return 0; +} + +/* + * Parse a single mount parameter. + */ +static int afs_parse_param(struct fs_context *fc, struct fs_parameter *param) +{ + struct fs_parse_result result; + struct afs_fs_context *ctx = fc->fs_private; + int opt; + + opt = fs_parse(fc, afs_fs_parameters, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_source: + return afs_parse_source(fc, param); + + case Opt_autocell: + ctx->autocell = true; + break; + + case Opt_dyn: + ctx->dyn_root = true; + break; + + case Opt_flock: + ctx->flock_mode = result.uint_32; + break; + + default: + return -EINVAL; + } + + _leave(" = 0"); + return 0; +} + +/* + * Validate the options, get the cell key and look up the volume. + */ +static int afs_validate_fc(struct fs_context *fc) +{ + struct afs_fs_context *ctx = fc->fs_private; + struct afs_volume *volume; + struct afs_cell *cell; + struct key *key; + int ret; + + if (!ctx->dyn_root) { + if (ctx->no_cell) { + pr_warn("kAFS: Can only specify source 'none' with -o dyn\n"); + return -EINVAL; + } + + if (!ctx->cell) { + pr_warn("kAFS: No cell specified\n"); + return -EDESTADDRREQ; + } + + reget_key: + /* We try to do the mount securely. */ + key = afs_request_key(ctx->cell); + if (IS_ERR(key)) + return PTR_ERR(key); + + ctx->key = key; + + if (ctx->volume) { + afs_put_volume(ctx->volume, afs_volume_trace_put_validate_fc); + ctx->volume = NULL; + } + + if (test_bit(AFS_CELL_FL_CHECK_ALIAS, &ctx->cell->flags)) { + ret = afs_cell_detect_alias(ctx->cell, key); + if (ret < 0) + return ret; + if (ret == 1) { + _debug("switch to alias"); + key_put(ctx->key); + ctx->key = NULL; + cell = afs_use_cell(ctx->cell->alias_of, + afs_cell_trace_use_fc_alias); + afs_unuse_cell(ctx->cell, afs_cell_trace_unuse_fc); + ctx->cell = cell; + goto reget_key; + } + } + + volume = afs_create_volume(ctx); + if (IS_ERR(volume)) + return PTR_ERR(volume); + + ctx->volume = volume; + if (volume->type != AFSVL_RWVOL) { + ctx->flock_mode = afs_flock_mode_local; + fc->sb_flags |= SB_RDONLY; + } + } return 0; } @@ -348,42 +417,37 @@ static int afs_parse_device_name(struct afs_mount_params *params, /* * check a superblock to see if it's the one we're looking for */ -static int afs_test_super(struct super_block *sb, void *data) +static int afs_test_super(struct super_block *sb, struct fs_context *fc) { - struct afs_super_info *as1 = data; + struct afs_fs_context *ctx = fc->fs_private; struct afs_super_info *as = AFS_FS_S(sb); - return (as->net_ns == as1->net_ns && + return (as->net_ns == fc->net_ns && as->volume && - as->volume->vid == as1->volume->vid && + as->volume->vid == ctx->volume->vid && + as->cell == ctx->cell && !as->dyn_root); } -static int afs_dynroot_test_super(struct super_block *sb, void *data) +static int afs_dynroot_test_super(struct super_block *sb, struct fs_context *fc) { - struct afs_super_info *as1 = data; struct afs_super_info *as = AFS_FS_S(sb); - return (as->net_ns == as1->net_ns && + return (as->net_ns == fc->net_ns && as->dyn_root); } -static int afs_set_super(struct super_block *sb, void *data) +static int afs_set_super(struct super_block *sb, struct fs_context *fc) { - struct afs_super_info *as = data; - - sb->s_fs_info = as; return set_anon_super(sb, NULL); } /* * fill in the superblock */ -static int afs_fill_super(struct super_block *sb, - struct afs_mount_params *params) +static int afs_fill_super(struct super_block *sb, struct afs_fs_context *ctx) { struct afs_super_info *as = AFS_FS_S(sb); - struct afs_fid fid; struct inode *inode = NULL; int ret; @@ -392,6 +456,7 @@ static int afs_fill_super(struct super_block *sb, /* fill in the superblock */ sb->s_blocksize = PAGE_SIZE; sb->s_blocksize_bits = PAGE_SHIFT; + sb->s_maxbytes = MAX_LFS_FILESIZE; sb->s_magic = AFS_FS_MAGIC; sb->s_op = &afs_super_ops; if (!as->dyn_root) @@ -399,40 +464,29 @@ static int afs_fill_super(struct super_block *sb, ret = super_setup_bdi(sb); if (ret) return ret; - sb->s_bdi->ra_pages = VM_MAX_READAHEAD * 1024 / PAGE_SIZE; /* allocate the root inode and dentry */ if (as->dyn_root) { - inode = afs_iget_pseudo_dir(sb, true); - sb->s_flags |= SB_RDONLY; + inode = afs_dynroot_iget_root(sb); } else { sprintf(sb->s_id, "%llu", as->volume->vid); afs_activate_volume(as->volume); - fid.vid = as->volume->vid; - fid.vnode = 1; - fid.vnode_hi = 0; - fid.unique = 1; - inode = afs_iget(sb, params->key, &fid, NULL, NULL, NULL); + inode = afs_root_iget(sb, ctx->key); } if (IS_ERR(inode)) return PTR_ERR(inode); - if (params->autocell || params->dyn_root) - set_bit(AFS_VNODE_AUTOCELL, &AFS_FS_I(inode)->flags); - ret = -ENOMEM; sb->s_root = d_make_root(inode); if (!sb->s_root) goto error; if (as->dyn_root) { - sb->s_d_op = &afs_dynroot_dentry_operations; - ret = afs_dynroot_populate(sb); - if (ret < 0) - goto error; + set_default_d_op(sb, &afs_dynroot_dentry_operations); } else { - sb->s_d_op = &afs_fs_dentry_operations; + set_default_d_op(sb, &afs_fs_dentry_operations); + rcu_assign_pointer(as->volume->sb, sb); } _leave(" = 0"); @@ -443,17 +497,22 @@ error: return ret; } -static struct afs_super_info *afs_alloc_sbi(struct afs_mount_params *params) +static struct afs_super_info *afs_alloc_sbi(struct fs_context *fc) { + struct afs_fs_context *ctx = fc->fs_private; struct afs_super_info *as; as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); if (as) { - as->net_ns = get_net(params->net_ns); - if (params->dyn_root) + as->net_ns = get_net(fc->net_ns); + as->flock_mode = ctx->flock_mode; + if (ctx->dyn_root) { as->dyn_root = true; - else - as->cell = afs_get_cell(params->cell); + } else { + as->cell = afs_use_cell(ctx->cell, afs_cell_trace_use_sbi); + as->volume = afs_get_volume(ctx->volume, + afs_volume_trace_get_alloc_sbi); + } } return as; } @@ -461,8 +520,8 @@ static struct afs_super_info *afs_alloc_sbi(struct afs_mount_params *params) static void afs_destroy_sbi(struct afs_super_info *as) { if (as) { - afs_put_volume(as->cell, as->volume); - afs_put_cell(afs_net(as->net_ns), as->cell); + afs_put_volume(as->volume, afs_volume_trace_put_destroy_sbi); + afs_unuse_cell(as->cell, afs_cell_trace_unuse_sbi); put_net(as->net_ns); kfree(as); } @@ -471,16 +530,12 @@ static void afs_destroy_sbi(struct afs_super_info *as) static void afs_kill_super(struct super_block *sb) { struct afs_super_info *as = AFS_FS_S(sb); - struct afs_net *net = afs_net(as->net_ns); - if (as->dyn_root) - afs_dynroot_depopulate(sb); - /* Clear the callback interests (which will do ilookup5) before * deactivating the superblock. */ if (as->volume) - afs_clear_callback_interests(net, as->volume->servers); + rcu_assign_pointer(as->volume->sb, NULL); kill_anon_super(sb); if (as->volume) afs_deactivate_volume(as->volume); @@ -488,111 +543,102 @@ static void afs_kill_super(struct super_block *sb) } /* - * get an AFS superblock + * Get an AFS superblock and root directory. */ -static struct dentry *afs_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *options) +static int afs_get_tree(struct fs_context *fc) { - struct afs_mount_params params; + struct afs_fs_context *ctx = fc->fs_private; struct super_block *sb; - struct afs_volume *candidate; - struct key *key; struct afs_super_info *as; int ret; - _enter(",,%s,%p", dev_name, options); - - memset(¶ms, 0, sizeof(params)); - - ret = -EINVAL; - if (current->nsproxy->net_ns != &init_net) + ret = afs_validate_fc(fc); + if (ret) goto error; - params.net_ns = current->nsproxy->net_ns; - params.net = afs_net(params.net_ns); - - /* parse the options and device name */ - if (options) { - ret = afs_parse_options(¶ms, options, &dev_name); - if (ret < 0) - goto error; - } - if (!params.dyn_root) { - ret = afs_parse_device_name(¶ms, dev_name); - if (ret < 0) - goto error; - - /* try and do the mount securely */ - key = afs_request_key(params.cell); - if (IS_ERR(key)) { - _leave(" = %ld [key]", PTR_ERR(key)); - ret = PTR_ERR(key); - goto error; - } - params.key = key; - } + _enter(""); /* allocate a superblock info record */ ret = -ENOMEM; - as = afs_alloc_sbi(¶ms); + as = afs_alloc_sbi(fc); if (!as) - goto error_key; - - if (!params.dyn_root) { - /* Assume we're going to need a volume record; at the very - * least we can use it to update the volume record if we have - * one already. This checks that the volume exists within the - * cell. - */ - candidate = afs_create_volume(¶ms); - if (IS_ERR(candidate)) { - ret = PTR_ERR(candidate); - goto error_as; - } - - as->volume = candidate; - } + goto error; + fc->s_fs_info = as; /* allocate a deviceless superblock */ - sb = sget(fs_type, - as->dyn_root ? afs_dynroot_test_super : afs_test_super, - afs_set_super, flags, as); + sb = sget_fc(fc, + as->dyn_root ? afs_dynroot_test_super : afs_test_super, + afs_set_super); if (IS_ERR(sb)) { ret = PTR_ERR(sb); - goto error_as; + goto error; } if (!sb->s_root) { /* initial superblock/root creation */ _debug("create"); - ret = afs_fill_super(sb, ¶ms); + ret = afs_fill_super(sb, ctx); if (ret < 0) goto error_sb; - as = NULL; sb->s_flags |= SB_ACTIVE; } else { _debug("reuse"); ASSERTCMP(sb->s_flags, &, SB_ACTIVE); - afs_destroy_sbi(as); - as = NULL; } - afs_put_cell(params.net, params.cell); - key_put(params.key); + fc->root = dget(sb->s_root); + trace_afs_get_tree(as->cell, as->volume); _leave(" = 0 [%p]", sb); - return dget(sb->s_root); + return 0; error_sb: deactivate_locked_super(sb); - goto error_key; -error_as: - afs_destroy_sbi(as); -error_key: - key_put(params.key); error: - afs_put_cell(params.net, params.cell); _leave(" = %d", ret); - return ERR_PTR(ret); + return ret; +} + +static void afs_free_fc(struct fs_context *fc) +{ + struct afs_fs_context *ctx = fc->fs_private; + + afs_destroy_sbi(fc->s_fs_info); + afs_put_volume(ctx->volume, afs_volume_trace_put_free_fc); + afs_unuse_cell(ctx->cell, afs_cell_trace_unuse_fc); + key_put(ctx->key); + kfree(ctx); +} + +static const struct fs_context_operations afs_context_ops = { + .free = afs_free_fc, + .parse_param = afs_parse_param, + .get_tree = afs_get_tree, +}; + +/* + * Set up the filesystem mount context. + */ +static int afs_init_fs_context(struct fs_context *fc) +{ + struct afs_fs_context *ctx; + struct afs_cell *cell; + + ctx = kzalloc(sizeof(struct afs_fs_context), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->type = AFSVL_ROVOL; + ctx->net = afs_net(fc->net_ns); + + /* Default to the workstation cell. */ + cell = afs_find_cell(ctx->net, NULL, 0, afs_cell_trace_use_fc); + if (IS_ERR(cell)) + cell = NULL; + ctx->cell = cell; + + fc->fs_private = ctx; + fc->ops = &afs_context_ops; + return 0; } /* @@ -605,8 +651,8 @@ static void afs_i_init_once(void *_vnode) struct afs_vnode *vnode = _vnode; memset(vnode, 0, sizeof(*vnode)); - inode_init_once(&vnode->vfs_inode); - mutex_init(&vnode->io_lock); + inode_init_once(&vnode->netfs.inode); + INIT_LIST_HEAD(&vnode->io_lock_waiters); init_rwsem(&vnode->validate_lock); spin_lock_init(&vnode->wb_lock); spin_lock_init(&vnode->lock); @@ -614,6 +660,7 @@ static void afs_i_init_once(void *_vnode) INIT_LIST_HEAD(&vnode->pending_locks); INIT_LIST_HEAD(&vnode->granted_locks); INIT_DELAYED_WORK(&vnode->lock_work, afs_lock_work); + INIT_LIST_HEAD(&vnode->cb_mmap_link); seqlock_init(&vnode->cb_lock); } @@ -624,7 +671,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb) { struct afs_vnode *vnode; - vnode = kmem_cache_alloc(afs_inode_cachep, GFP_KERNEL); + vnode = alloc_inode_sb(sb, afs_inode_cachep, GFP_KERNEL); if (!vnode) return NULL; @@ -633,28 +680,27 @@ static struct inode *afs_alloc_inode(struct super_block *sb) /* Reset anything that shouldn't leak from one inode to the next. */ memset(&vnode->fid, 0, sizeof(vnode->fid)); memset(&vnode->status, 0, sizeof(vnode->status)); + afs_vnode_set_cache(vnode, NULL); vnode->volume = NULL; vnode->lock_key = NULL; vnode->permit_cache = NULL; - vnode->cb_interest = NULL; -#ifdef CONFIG_AFS_FSCACHE - vnode->cache = NULL; -#endif + vnode->directory = NULL; + vnode->directory_size = 0; vnode->flags = 1 << AFS_VNODE_UNSET; - vnode->cb_type = 0; vnode->lock_state = AFS_VNODE_LOCK_NONE; - _leave(" = %p", &vnode->vfs_inode); - return &vnode->vfs_inode; + init_rwsem(&vnode->rmdir_lock); + INIT_WORK(&vnode->cb_work, afs_invalidate_mmap_work); + + _leave(" = %p", &vnode->netfs.inode); + return &vnode->netfs.inode; } -static void afs_i_callback(struct rcu_head *head) +static void afs_free_inode(struct inode *inode) { - struct inode *inode = container_of(head, struct inode, i_rcu); - struct afs_vnode *vnode = AFS_FS_I(inode); - kmem_cache_free(afs_inode_cachep, vnode); + kmem_cache_free(afs_inode_cachep, AFS_FS_I(inode)); } /* @@ -668,23 +714,38 @@ static void afs_destroy_inode(struct inode *inode) _debug("DESTROY INODE %p", inode); - ASSERTCMP(vnode->cb_interest, ==, NULL); - - call_rcu(&inode->i_rcu, afs_i_callback); atomic_dec(&afs_count_active_inodes); } +static void afs_get_volume_status_success(struct afs_operation *op) +{ + struct afs_volume_status *vs = &op->volstatus.vs; + struct kstatfs *buf = op->volstatus.buf; + + if (vs->max_quota == 0) + buf->f_blocks = vs->part_max_blocks; + else + buf->f_blocks = vs->max_quota; + + if (buf->f_blocks > vs->blocks_in_use) + buf->f_bavail = buf->f_bfree = + buf->f_blocks - vs->blocks_in_use; +} + +static const struct afs_operation_ops afs_get_volume_status_operation = { + .issue_afs_rpc = afs_fs_get_volume_status, + .issue_yfs_rpc = yfs_fs_get_volume_status, + .success = afs_get_volume_status_success, +}; + /* * return information about an AFS volume */ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct afs_super_info *as = AFS_FS_S(dentry->d_sb); - struct afs_fs_cursor fc; - struct afs_volume_status vs; + struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - struct key *key; - int ret; buf->f_type = dentry->d_sb->s_magic; buf->f_bsize = AFS_BLOCK_SIZE; @@ -697,32 +758,13 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } - key = afs_request_key(vnode->volume->cell); - if (IS_ERR(key)) - return PTR_ERR(key); + op = afs_alloc_operation(NULL, as->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, key)) { - fc.flags |= AFS_FS_CURSOR_NO_VSLEEP; - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_get_volume_status(&fc, &vs); - } - - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } - - key_put(key); - - if (ret == 0) { - if (vs.max_quota == 0) - buf->f_blocks = vs.part_max_blocks; - else - buf->f_blocks = vs.max_quota; - buf->f_bavail = buf->f_bfree = buf->f_blocks - vs.blocks_in_use; - } - - return ret; + afs_op_set_vnode(op, 0, vnode); + op->nr_files = 1; + op->volstatus.buf = buf; + op->ops = &afs_get_volume_status_operation; + return afs_do_sync_operation(op); } diff --git a/fs/afs/validation.c b/fs/afs/validation.c new file mode 100644 index 000000000000..0ba8336c9025 --- /dev/null +++ b/fs/afs/validation.c @@ -0,0 +1,484 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* vnode and volume validity verification. + * + * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/sched.h> +#include "internal.h" + +/* + * Data validation is managed through a number of mechanisms from the server: + * + * (1) On first contact with a server (such as if it has just been rebooted), + * the server sends us a CB.InitCallBackState* request. + * + * (2) On a RW volume, in response to certain vnode (inode)-accessing RPC + * calls, the server maintains a time-limited per-vnode promise that it + * will send us a CB.CallBack request if a third party alters the vnodes + * accessed. + * + * Note that a vnode-level callbacks may also be sent for other reasons, + * such as filelock release. + * + * (3) On a RO (or Backup) volume, in response to certain vnode-accessing RPC + * calls, each server maintains a time-limited per-volume promise that it + * will send us a CB.CallBack request if the RO volume is updated to a + * snapshot of the RW volume ("vos release"). This is an atomic event + * that cuts over all instances of the RO volume across multiple servers + * simultaneously. + * + * Note that a volume-level callbacks may also be sent for other reasons, + * such as the volumeserver taking over control of the volume from the + * fileserver. + * + * Note also that each server maintains an independent time limit on an + * independent callback. + * + * (4) Certain RPC calls include a volume information record "VolSync" in + * their reply. This contains a creation date for the volume that should + * remain unchanged for a RW volume (but will be changed if the volume is + * restored from backup) or will be bumped to the time of snapshotting + * when a RO volume is released. + * + * In order to track this events, the following are provided: + * + * ->cb_v_break. A counter of events that might mean that the contents of + * a volume have been altered since we last checked a vnode. + * + * ->cb_v_check. A counter of the number of events that we've sent a + * query to the server for. Everything's up to date if this equals + * cb_v_break. + * + * ->cb_scrub. A counter of the number of regression events for which we + * have to completely wipe the cache. + * + * ->cb_ro_snapshot. A counter of the number of times that we've + * recognised that a RO volume has been updated. + * + * ->cb_break. A counter of events that might mean that the contents of a + * vnode have been altered. + * + * ->cb_expires_at. The time at which the callback promise expires or + * AFS_NO_CB_PROMISE if we have no promise. + * + * The way we manage things is: + * + * (1) When a volume-level CB.CallBack occurs, we increment ->cb_v_break on + * the volume and reset ->cb_expires_at (ie. set AFS_NO_CB_PROMISE) on the + * volume and volume's server record. + * + * (2) When a CB.InitCallBackState occurs, we treat this as a volume-level + * callback break on all the volumes that have been using that volume + * (ie. increment ->cb_v_break and reset ->cb_expires_at). + * + * (3) When a vnode-level CB.CallBack occurs, we increment ->cb_break on the + * vnode and reset its ->cb_expires_at. If the vnode is mmapped, we also + * dispatch a work item to unmap all PTEs to the vnode's pagecache to + * force reentry to the filesystem for revalidation. + * + * (4) When entering the filesystem, we call afs_validate() to check the + * validity of a vnode. This first checks to see if ->cb_v_check and + * ->cb_v_break match, and if they don't, we lock volume->cb_check_lock + * exclusively and perform an FS.FetchStatus on the vnode. + * + * After checking the volume, we check the vnode. If there's a mismatch + * between the volume counters and the vnode's mirrors of those counters, + * we lock vnode->validate_lock and issue an FS.FetchStatus on the vnode. + * + * (5) When the reply from FS.FetchStatus arrives, the VolSync record is + * parsed: + * + * (A) If the Creation timestamp has changed on a RW volume or regressed + * on a RO volume, we try to increment ->cb_scrub; if it advances on a + * RO volume, we assume "vos release" happened and try to increment + * ->cb_ro_snapshot. + * + * (B) If the Update timestamp has regressed, we try to increment + * ->cb_scrub. + * + * Note that in both of these cases, we only do the increment if we can + * cmpxchg the value of the timestamp from the value we noted before the + * op. This tries to prevent parallel ops from fighting one another. + * + * volume->cb_v_check is then set to ->cb_v_break. + * + * (6) The AFSCallBack record included in the FS.FetchStatus reply is also + * parsed and used to set the promise in ->cb_expires_at for the vnode, + * the volume and the volume's server record. + * + * (7) If ->cb_scrub is seen to have advanced, we invalidate the pagecache for + * the vnode. + */ + +/* + * Check the validity of a vnode/inode and its parent volume. + */ +bool afs_check_validity(const struct afs_vnode *vnode) +{ + const struct afs_volume *volume = vnode->volume; + enum afs_vnode_invalid_trace trace = afs_vnode_valid_trace; + time64_t cb_expires_at = atomic64_read(&vnode->cb_expires_at); + time64_t deadline = ktime_get_real_seconds() + 10; + + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) + return true; + + if (atomic_read(&volume->cb_v_check) != atomic_read(&volume->cb_v_break)) + trace = afs_vnode_invalid_trace_cb_v_break; + else if (cb_expires_at == AFS_NO_CB_PROMISE) + trace = afs_vnode_invalid_trace_no_cb_promise; + else if (cb_expires_at <= deadline) + trace = afs_vnode_invalid_trace_expired; + else if (volume->cb_expires_at <= deadline) + trace = afs_vnode_invalid_trace_vol_expired; + else if (vnode->cb_ro_snapshot != atomic_read(&volume->cb_ro_snapshot)) + trace = afs_vnode_invalid_trace_cb_ro_snapshot; + else if (vnode->cb_scrub != atomic_read(&volume->cb_scrub)) + trace = afs_vnode_invalid_trace_cb_scrub; + else if (test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + trace = afs_vnode_invalid_trace_zap_data; + else + return true; + trace_afs_vnode_invalid(vnode, trace); + return false; +} + +/* + * See if the server we've just talked to is currently excluded. + */ +static bool __afs_is_server_excluded(struct afs_operation *op, struct afs_volume *volume) +{ + const struct afs_server_entry *se; + const struct afs_server_list *slist; + bool is_excluded = true; + int i; + + rcu_read_lock(); + + slist = rcu_dereference(volume->servers); + for (i = 0; i < slist->nr_servers; i++) { + se = &slist->servers[i]; + if (op->server == se->server) { + is_excluded = test_bit(AFS_SE_EXCLUDED, &se->flags); + break; + } + } + + rcu_read_unlock(); + return is_excluded; +} + +/* + * Update the volume's server list when the creation time changes and see if + * the server we've just talked to is currently excluded. + */ +static int afs_is_server_excluded(struct afs_operation *op, struct afs_volume *volume) +{ + int ret; + + if (__afs_is_server_excluded(op, volume)) + return 1; + + set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags); + ret = afs_check_volume_status(op->volume, op); + if (ret < 0) + return ret; + + return __afs_is_server_excluded(op, volume); +} + +/* + * Handle a change to the volume creation time in the VolSync record. + */ +static int afs_update_volume_creation_time(struct afs_operation *op, struct afs_volume *volume) +{ + unsigned int snap; + time64_t cur = volume->creation_time; + time64_t old = op->pre_volsync.creation; + time64_t new = op->volsync.creation; + int ret; + + _enter("%llx,%llx,%llx->%llx", volume->vid, cur, old, new); + + if (cur == TIME64_MIN) { + volume->creation_time = new; + return 0; + } + + if (new == cur) + return 0; + + /* Try to advance the creation timestamp from what we had before the + * operation to what we got back from the server. This should + * hopefully ensure that in a race between multiple operations only one + * of them will do this. + */ + if (cur != old) + return 0; + + /* If the creation time changes in an unexpected way, we need to scrub + * our caches. For a RW vol, this will only change if the volume is + * restored from a backup; for a RO/Backup vol, this will advance when + * the volume is updated to a new snapshot (eg. "vos release"). + */ + if (volume->type == AFSVL_RWVOL) + goto regressed; + if (volume->type == AFSVL_BACKVOL) { + if (new < old) + goto regressed; + goto advance; + } + + /* We have an RO volume, we need to query the VL server and look at the + * server flags to see if RW->RO replication is in progress. + */ + ret = afs_is_server_excluded(op, volume); + if (ret < 0) + return ret; + if (ret > 0) { + snap = atomic_read(&volume->cb_ro_snapshot); + trace_afs_cb_v_break(volume->vid, snap, afs_cb_break_volume_excluded); + return ret; + } + +advance: + snap = atomic_inc_return(&volume->cb_ro_snapshot); + trace_afs_cb_v_break(volume->vid, snap, afs_cb_break_for_vos_release); + volume->creation_time = new; + return 0; + +regressed: + atomic_inc(&volume->cb_scrub); + trace_afs_cb_v_break(volume->vid, 0, afs_cb_break_for_creation_regress); + volume->creation_time = new; + return 0; +} + +/* + * Handle a change to the volume update time in the VolSync record. + */ +static void afs_update_volume_update_time(struct afs_operation *op, struct afs_volume *volume) +{ + enum afs_cb_break_reason reason = afs_cb_break_no_break; + time64_t cur = volume->update_time; + time64_t old = op->pre_volsync.update; + time64_t new = op->volsync.update; + + _enter("%llx,%llx,%llx->%llx", volume->vid, cur, old, new); + + if (cur == TIME64_MIN) { + volume->update_time = new; + return; + } + + if (new == cur) + return; + + /* If the volume update time changes in an unexpected way, we need to + * scrub our caches. For a RW vol, this will advance on every + * modification op; for a RO/Backup vol, this will advance when the + * volume is updated to a new snapshot (eg. "vos release"). + */ + if (new < old) + reason = afs_cb_break_for_update_regress; + + /* Try to advance the update timestamp from what we had before the + * operation to what we got back from the server. This should + * hopefully ensure that in a race between multiple operations only one + * of them will do this. + */ + if (cur == old) { + if (reason == afs_cb_break_for_update_regress) { + atomic_inc(&volume->cb_scrub); + trace_afs_cb_v_break(volume->vid, 0, reason); + } + volume->update_time = new; + } +} + +static int afs_update_volume_times(struct afs_operation *op, struct afs_volume *volume) +{ + int ret = 0; + + if (likely(op->volsync.creation == volume->creation_time && + op->volsync.update == volume->update_time)) + return 0; + + mutex_lock(&volume->volsync_lock); + if (op->volsync.creation != volume->creation_time) { + ret = afs_update_volume_creation_time(op, volume); + if (ret < 0) + goto out; + } + if (op->volsync.update != volume->update_time) + afs_update_volume_update_time(op, volume); +out: + mutex_unlock(&volume->volsync_lock); + return ret; +} + +/* + * Update the state of a volume, including recording the expiration time of the + * callback promise. Returns 1 to redo the operation from the start. + */ +int afs_update_volume_state(struct afs_operation *op) +{ + struct afs_server_list *slist = op->server_list; + struct afs_server_entry *se = &slist->servers[op->server_index]; + struct afs_callback *cb = &op->file[0].scb.callback; + struct afs_volume *volume = op->volume; + unsigned int cb_v_break = atomic_read(&volume->cb_v_break); + unsigned int cb_v_check = atomic_read(&volume->cb_v_check); + int ret; + + _enter("%llx", op->volume->vid); + + if (op->volsync.creation != TIME64_MIN || op->volsync.update != TIME64_MIN) { + ret = afs_update_volume_times(op, volume); + if (ret != 0) { + _leave(" = %d", ret); + return ret; + } + } + + if (op->cb_v_break == cb_v_break && + (op->file[0].scb.have_cb || op->file[1].scb.have_cb)) { + time64_t expires_at = cb->expires_at; + + if (!op->file[0].scb.have_cb) + expires_at = op->file[1].scb.callback.expires_at; + + se->cb_expires_at = expires_at; + volume->cb_expires_at = expires_at; + } + if (cb_v_check < op->cb_v_break) + atomic_cmpxchg(&volume->cb_v_check, cb_v_check, op->cb_v_break); + return 0; +} + +/* + * mark the data attached to an inode as obsolete due to a write on the server + * - might also want to ditch all the outstanding writes and dirty pages + */ +static void afs_zap_data(struct afs_vnode *vnode) +{ + _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); + + afs_invalidate_cache(vnode, 0); + + /* nuke all the non-dirty pages that aren't locked, mapped or being + * written back in a regular file and completely discard the pages in a + * directory or symlink */ + if (S_ISREG(vnode->netfs.inode.i_mode)) + filemap_invalidate_inode(&vnode->netfs.inode, true, 0, LLONG_MAX); + else + filemap_invalidate_inode(&vnode->netfs.inode, false, 0, LLONG_MAX); +} + +/* + * validate a vnode/inode + * - there are several things we need to check + * - parent dir data changes (rm, rmdir, rename, mkdir, create, link, + * symlink) + * - parent dir metadata changed (security changes) + * - dentry data changed (write, truncate) + * - dentry metadata changed (security changes) + */ +int afs_validate(struct afs_vnode *vnode, struct key *key) +{ + struct afs_volume *volume = vnode->volume; + unsigned int cb_ro_snapshot, cb_scrub; + time64_t deadline = ktime_get_real_seconds() + 10; + bool zap = false, locked_vol = false; + int ret; + + _enter("{v={%llx:%llu} fl=%lx},%x", + vnode->fid.vid, vnode->fid.vnode, vnode->flags, + key_serial(key)); + + if (afs_check_validity(vnode)) + return test_bit(AFS_VNODE_DELETED, &vnode->flags) ? -ESTALE : 0; + + ret = down_write_killable(&vnode->validate_lock); + if (ret < 0) + goto error; + + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { + ret = -ESTALE; + goto error_unlock; + } + + /* Validate a volume after the v_break has changed or the volume + * callback expired. We only want to do this once per volume per + * v_break change. The actual work will be done when parsing the + * status fetch reply. + */ + if (volume->cb_expires_at <= deadline || + atomic_read(&volume->cb_v_check) != atomic_read(&volume->cb_v_break)) { + ret = mutex_lock_interruptible(&volume->cb_check_lock); + if (ret < 0) + goto error_unlock; + locked_vol = true; + } + + cb_ro_snapshot = atomic_read(&volume->cb_ro_snapshot); + cb_scrub = atomic_read(&volume->cb_scrub); + if (vnode->cb_ro_snapshot != cb_ro_snapshot || + vnode->cb_scrub != cb_scrub) + unmap_mapping_pages(vnode->netfs.inode.i_mapping, 0, 0, false); + + if (vnode->cb_ro_snapshot != cb_ro_snapshot || + vnode->cb_scrub != cb_scrub || + volume->cb_expires_at <= deadline || + atomic_read(&volume->cb_v_check) != atomic_read(&volume->cb_v_break) || + atomic64_read(&vnode->cb_expires_at) <= deadline + ) { + ret = afs_fetch_status(vnode, key, false, NULL); + if (ret < 0) { + if (ret == -ENOENT) { + set_bit(AFS_VNODE_DELETED, &vnode->flags); + ret = -ESTALE; + } + goto error_unlock; + } + + _debug("new promise [fl=%lx]", vnode->flags); + } + + /* We can drop the volume lock now as. */ + if (locked_vol) { + mutex_unlock(&volume->cb_check_lock); + locked_vol = false; + } + + cb_ro_snapshot = atomic_read(&volume->cb_ro_snapshot); + cb_scrub = atomic_read(&volume->cb_scrub); + _debug("vnode inval %x==%x %x==%x", + vnode->cb_ro_snapshot, cb_ro_snapshot, + vnode->cb_scrub, cb_scrub); + if (vnode->cb_scrub != cb_scrub) + zap = true; + vnode->cb_ro_snapshot = cb_ro_snapshot; + vnode->cb_scrub = cb_scrub; + + /* if the vnode's data version number changed then its contents are + * different */ + zap |= test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); + if (zap) + afs_zap_data(vnode); + up_write(&vnode->validate_lock); + _leave(" = 0"); + return 0; + +error_unlock: + if (locked_vol) + mutex_unlock(&volume->cb_check_lock); + up_write(&vnode->validate_lock); +error: + _leave(" = %d", ret); + return ret; +} diff --git a/fs/afs/vl_alias.c b/fs/afs/vl_alias.c new file mode 100644 index 000000000000..fc9676abd252 --- /dev/null +++ b/fs/afs/vl_alias.c @@ -0,0 +1,340 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* AFS cell alias detection + * + * Copyright (C) 2020 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/slab.h> +#include <linux/sched.h> +#include <linux/namei.h> +#include <keys/rxrpc-type.h> +#include "internal.h" + +/* + * Sample a volume. + */ +static struct afs_volume *afs_sample_volume(struct afs_cell *cell, struct key *key, + const char *name, unsigned int namelen) +{ + struct afs_volume *volume; + struct afs_fs_context fc = { + .type = 0, /* Explicitly leave it to the VLDB */ + .volnamesz = namelen, + .volname = name, + .net = cell->net, + .cell = cell, + .key = key, /* This might need to be something */ + }; + + volume = afs_create_volume(&fc); + _leave(" = %p", volume); + return volume; +} + +/* + * Compare the address lists of a pair of fileservers. + */ +static int afs_compare_fs_alists(const struct afs_server *server_a, + const struct afs_server *server_b) +{ + const struct afs_addr_list *la, *lb; + int a = 0, b = 0, addr_matches = 0; + + la = rcu_dereference(server_a->endpoint_state)->addresses; + lb = rcu_dereference(server_b->endpoint_state)->addresses; + + while (a < la->nr_addrs && b < lb->nr_addrs) { + unsigned long pa = (unsigned long)la->addrs[a].peer; + unsigned long pb = (unsigned long)lb->addrs[b].peer; + long diff = pa - pb; + + if (diff < 0) { + a++; + } else if (diff > 0) { + b++; + } else { + addr_matches++; + a++; + b++; + } + } + + return addr_matches; +} + +/* + * Compare the fileserver lists of two volumes. The server lists are sorted in + * order of ascending UUID. + */ +static int afs_compare_volume_slists(const struct afs_volume *vol_a, + const struct afs_volume *vol_b) +{ + const struct afs_server_list *la, *lb; + int i, a = 0, b = 0, uuid_matches = 0, addr_matches = 0; + + la = rcu_dereference(vol_a->servers); + lb = rcu_dereference(vol_b->servers); + + for (i = 0; i < AFS_MAXTYPES; i++) + if (vol_a->vids[i] != vol_b->vids[i]) + return 0; + + while (a < la->nr_servers && b < lb->nr_servers) { + const struct afs_server *server_a = la->servers[a].server; + const struct afs_server *server_b = lb->servers[b].server; + int diff = memcmp(&server_a->uuid, &server_b->uuid, sizeof(uuid_t)); + + if (diff < 0) { + a++; + } else if (diff > 0) { + b++; + } else { + uuid_matches++; + addr_matches += afs_compare_fs_alists(server_a, server_b); + a++; + b++; + } + } + + _leave(" = %d [um %d]", addr_matches, uuid_matches); + return addr_matches; +} + +/* + * Compare root.cell volumes. + */ +static int afs_compare_cell_roots(struct afs_cell *cell) +{ + struct afs_cell *p; + + _enter(""); + + rcu_read_lock(); + + hlist_for_each_entry_rcu(p, &cell->net->proc_cells, proc_link) { + if (p == cell || p->alias_of) + continue; + if (!p->root_volume) + continue; /* Ignore cells that don't have a root.cell volume. */ + + if (afs_compare_volume_slists(cell->root_volume, p->root_volume) != 0) + goto is_alias; + } + + rcu_read_unlock(); + _leave(" = 0"); + return 0; + +is_alias: + rcu_read_unlock(); + cell->alias_of = afs_use_cell(p, afs_cell_trace_use_alias); + return 1; +} + +/* + * Query the new cell for a volume from a cell we're already using. + */ +static int afs_query_for_alias_one(struct afs_cell *cell, struct key *key, + struct afs_cell *p) +{ + struct afs_volume *volume, *pvol = NULL; + int ret; + + /* Arbitrarily pick a volume from the list. */ + read_seqlock_excl(&p->volume_lock); + if (!RB_EMPTY_ROOT(&p->volumes)) + pvol = afs_get_volume(rb_entry(p->volumes.rb_node, + struct afs_volume, cell_node), + afs_volume_trace_get_query_alias); + read_sequnlock_excl(&p->volume_lock); + if (!pvol) + return 0; + + _enter("%s:%s", cell->name, pvol->name); + + /* And see if it's in the new cell. */ + volume = afs_sample_volume(cell, key, pvol->name, pvol->name_len); + if (IS_ERR(volume)) { + afs_put_volume(pvol, afs_volume_trace_put_query_alias); + if (PTR_ERR(volume) != -ENOMEDIUM) + return PTR_ERR(volume); + /* That volume is not in the new cell, so not an alias */ + return 0; + } + + /* The new cell has a like-named volume also - compare volume ID, + * server and address lists. + */ + ret = 0; + if (pvol->vid == volume->vid) { + rcu_read_lock(); + if (afs_compare_volume_slists(volume, pvol)) + ret = 1; + rcu_read_unlock(); + } + + afs_put_volume(volume, afs_volume_trace_put_query_alias); + afs_put_volume(pvol, afs_volume_trace_put_query_alias); + return ret; +} + +/* + * Query the new cell for volumes we know exist in cells we're already using. + */ +static int afs_query_for_alias(struct afs_cell *cell, struct key *key) +{ + struct afs_cell *p; + + _enter("%s", cell->name); + + if (mutex_lock_interruptible(&cell->net->proc_cells_lock) < 0) + return -ERESTARTSYS; + + hlist_for_each_entry(p, &cell->net->proc_cells, proc_link) { + if (p == cell || p->alias_of) + continue; + if (RB_EMPTY_ROOT(&p->volumes)) + continue; + if (p->root_volume) + continue; /* Ignore cells that have a root.cell volume. */ + afs_use_cell(p, afs_cell_trace_use_check_alias); + mutex_unlock(&cell->net->proc_cells_lock); + + if (afs_query_for_alias_one(cell, key, p) != 0) + goto is_alias; + + if (mutex_lock_interruptible(&cell->net->proc_cells_lock) < 0) { + afs_unuse_cell(p, afs_cell_trace_unuse_check_alias); + return -ERESTARTSYS; + } + + afs_unuse_cell(p, afs_cell_trace_unuse_check_alias); + } + + mutex_unlock(&cell->net->proc_cells_lock); + _leave(" = 0"); + return 0; + +is_alias: + cell->alias_of = p; /* Transfer our ref */ + return 1; +} + +/* + * Look up a VLDB record for a volume. + */ +static char *afs_vl_get_cell_name(struct afs_cell *cell, struct key *key) +{ + struct afs_vl_cursor vc; + char *cell_name = ERR_PTR(-EDESTADDRREQ); + bool skipped = false, not_skipped = false; + int ret; + + if (!afs_begin_vlserver_operation(&vc, cell, key)) + return ERR_PTR(-ERESTARTSYS); + + while (afs_select_vlserver(&vc)) { + if (!test_bit(AFS_VLSERVER_FL_IS_YFS, &vc.server->flags)) { + vc.call_error = -EOPNOTSUPP; + skipped = true; + continue; + } + not_skipped = true; + cell_name = afs_yfsvl_get_cell_name(&vc); + } + + ret = afs_end_vlserver_operation(&vc); + if (skipped && !not_skipped) + ret = -EOPNOTSUPP; + return ret < 0 ? ERR_PTR(ret) : cell_name; +} + +static int yfs_check_canonical_cell_name(struct afs_cell *cell, struct key *key) +{ + struct afs_cell *master; + size_t name_len; + char *cell_name; + + cell_name = afs_vl_get_cell_name(cell, key); + if (IS_ERR(cell_name)) + return PTR_ERR(cell_name); + + if (strcmp(cell_name, cell->name) == 0) { + kfree(cell_name); + return 0; + } + + name_len = strlen(cell_name); + if (!name_len || name_len > AFS_MAXCELLNAME) + master = ERR_PTR(-EOPNOTSUPP); + else + master = afs_lookup_cell(cell->net, cell_name, name_len, NULL, + AFS_LOOKUP_CELL_ALIAS_CHECK, + afs_cell_trace_use_lookup_canonical); + kfree(cell_name); + if (IS_ERR(master)) + return PTR_ERR(master); + + cell->alias_of = master; /* Transfer our ref */ + return 1; +} + +static int afs_do_cell_detect_alias(struct afs_cell *cell, struct key *key) +{ + struct afs_volume *root_volume; + int ret; + + _enter("%s", cell->name); + + ret = yfs_check_canonical_cell_name(cell, key); + if (ret != -EOPNOTSUPP) + return ret; + + /* Try and get the root.cell volume for comparison with other cells */ + root_volume = afs_sample_volume(cell, key, "root.cell", 9); + if (!IS_ERR(root_volume)) { + cell->root_volume = root_volume; + return afs_compare_cell_roots(cell); + } + + if (PTR_ERR(root_volume) != -ENOMEDIUM) + return PTR_ERR(root_volume); + + /* Okay, this cell doesn't have an root.cell volume. We need to + * locate some other random volume and use that to check. + */ + return afs_query_for_alias(cell, key); +} + +/* + * Check to see if a new cell is an alias of a cell we already have. At this + * point we have the cell's volume server list. + * + * Returns 0 if we didn't detect an alias, 1 if we found an alias and an error + * if we had problems gathering the data required. In the case the we did + * detect an alias, cell->alias_of is set to point to the assumed master. + */ +int afs_cell_detect_alias(struct afs_cell *cell, struct key *key) +{ + struct afs_net *net = cell->net; + int ret; + + if (mutex_lock_interruptible(&net->cells_alias_lock) < 0) + return -ERESTARTSYS; + + if (test_bit(AFS_CELL_FL_CHECK_ALIAS, &cell->flags)) { + ret = afs_do_cell_detect_alias(cell, key); + if (ret >= 0) + clear_bit_unlock(AFS_CELL_FL_CHECK_ALIAS, &cell->flags); + } else { + ret = cell->alias_of ? 1 : 0; + } + + mutex_unlock(&net->cells_alias_lock); + + if (ret == 1) + pr_notice("kAFS: Cell %s is an alias of %s\n", + cell->name, cell->alias_of->name); + return ret; +} diff --git a/fs/afs/vl_list.c b/fs/afs/vl_list.c index b4f1a84519b9..9b1c20daac53 100644 --- a/fs/afs/vl_list.c +++ b/fs/afs/vl_list.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS vlserver list management. * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -17,15 +13,19 @@ struct afs_vlserver *afs_alloc_vlserver(const char *name, size_t name_len, unsigned short port) { struct afs_vlserver *vlserver; + static atomic_t debug_ids; vlserver = kzalloc(struct_size(vlserver, name, name_len + 1), GFP_KERNEL); if (vlserver) { - atomic_set(&vlserver->usage, 1); + refcount_set(&vlserver->ref, 1); rwlock_init(&vlserver->lock); init_waitqueue_head(&vlserver->probe_wq); spin_lock_init(&vlserver->probe_lock); + vlserver->debug_id = atomic_inc_return(&debug_ids); + vlserver->rtt = UINT_MAX; vlserver->name_len = name_len; + vlserver->service_id = VL_SERVICE; vlserver->port = port; memcpy(vlserver->name, name, name_len); } @@ -36,19 +36,16 @@ static void afs_vlserver_rcu(struct rcu_head *rcu) { struct afs_vlserver *vlserver = container_of(rcu, struct afs_vlserver, rcu); - afs_put_addrlist(rcu_access_pointer(vlserver->addresses)); + afs_put_addrlist(rcu_access_pointer(vlserver->addresses), + afs_alist_trace_put_vlserver); kfree_rcu(vlserver, rcu); } void afs_put_vlserver(struct afs_net *net, struct afs_vlserver *vlserver) { - if (vlserver) { - unsigned int u = atomic_dec_return(&vlserver->usage); - //_debug("VL PUT %p{%u}", vlserver, u); - - if (u == 0) - call_rcu(&vlserver->rcu, afs_vlserver_rcu); - } + if (vlserver && + refcount_dec_and_test(&vlserver->ref)) + call_rcu(&vlserver->rcu, afs_vlserver_rcu); } struct afs_vlserver_list *afs_alloc_vlserver_list(unsigned int nr_servers) @@ -57,7 +54,7 @@ struct afs_vlserver_list *afs_alloc_vlserver_list(unsigned int nr_servers) vllist = kzalloc(struct_size(vllist, servers, nr_servers), GFP_KERNEL); if (vllist) { - atomic_set(&vllist->usage, 1); + refcount_set(&vllist->ref, 1); rwlock_init(&vllist->lock); } @@ -67,10 +64,7 @@ struct afs_vlserver_list *afs_alloc_vlserver_list(unsigned int nr_servers) void afs_put_vlserverlist(struct afs_net *net, struct afs_vlserver_list *vllist) { if (vllist) { - unsigned int u = atomic_dec_return(&vllist->usage); - - //_debug("VLLS PUT %p{%u}", vllist, u); - if (u == 0) { + if (refcount_dec_and_test(&vllist->ref)) { int i; for (i = 0; i < vllist->nr_servers; i++) { @@ -93,14 +87,15 @@ static u16 afs_extract_le16(const u8 **_b) /* * Build a VL server address list from a DNS queried server list. */ -static struct afs_addr_list *afs_extract_vl_addrs(const u8 **_b, const u8 *end, +static struct afs_addr_list *afs_extract_vl_addrs(struct afs_net *net, + const u8 **_b, const u8 *end, u8 nr_addrs, u16 port) { struct afs_addr_list *alist; const u8 *b = *_b; int ret = -EINVAL; - alist = afs_alloc_addrlist(nr_addrs, VL_SERVICE, port); + alist = afs_alloc_addrlist(nr_addrs); if (!alist) return ERR_PTR(-ENOMEM); if (nr_addrs == 0) @@ -119,7 +114,9 @@ static struct afs_addr_list *afs_extract_vl_addrs(const u8 **_b, const u8 *end, goto error; } memcpy(x, b, 4); - afs_merge_fs_addr4(alist, x[0], port); + ret = afs_merge_fs_addr4(net, alist, x[0], port); + if (ret < 0) + goto error; b += 4; break; @@ -129,7 +126,9 @@ static struct afs_addr_list *afs_extract_vl_addrs(const u8 **_b, const u8 *end, goto error; } memcpy(x, b, 16); - afs_merge_fs_addr6(alist, x, port); + ret = afs_merge_fs_addr6(net, alist, x, port); + if (ret < 0) + goto error; b += 16; break; @@ -150,7 +149,7 @@ static struct afs_addr_list *afs_extract_vl_addrs(const u8 **_b, const u8 *end, error: *_b = b; - afs_put_addrlist(alist); + afs_put_addrlist(alist, afs_alist_trace_put_parse_error); return ERR_PTR(ret); } @@ -232,18 +231,16 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell, if (bs.status > NR__dns_lookup_status) bs.status = NR__dns_lookup_status; + /* See if we can update an old server record */ server = NULL; - if (previous) { - /* See if we can update an old server record */ - for (i = 0; i < previous->nr_servers; i++) { - struct afs_vlserver *p = previous->servers[i].server; - - if (p->name_len == bs.name_len && - p->port == bs.port && - strncasecmp(b, p->name, bs.name_len) == 0) { - server = afs_get_vlserver(p); - break; - } + for (i = 0; i < previous->nr_servers; i++) { + struct afs_vlserver *p = previous->servers[i].server; + + if (p->name_len == bs.name_len && + p->port == bs.port && + strncasecmp(b, p->name, bs.name_len) == 0) { + server = afs_get_vlserver(p); + break; } } @@ -259,7 +256,7 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell, /* Extract the addresses - note that we can't skip this as we * have to advance the payload pointer. */ - addrs = afs_extract_vl_addrs(&b, end, bs.nr_addrs, bs.port); + addrs = afs_extract_vl_addrs(cell->net, &b, end, bs.nr_addrs, bs.port); if (IS_ERR(addrs)) { ret = PTR_ERR(addrs); goto error_2; @@ -267,7 +264,7 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell, if (vllist->nr_servers >= nr_servers) { _debug("skip %u >= %u", vllist->nr_servers, nr_servers); - afs_put_addrlist(addrs); + afs_put_addrlist(addrs, afs_alist_trace_put_parse_empty); afs_put_vlserver(cell->net, server); continue; } @@ -276,7 +273,7 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell, addrs->status = bs.status; if (addrs->nr_addrs == 0) { - afs_put_addrlist(addrs); + afs_put_addrlist(addrs, afs_alist_trace_put_parse_empty); if (!rcu_access_pointer(server->addresses)) { afs_put_vlserver(cell->net, server); continue; @@ -285,10 +282,10 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell, struct afs_addr_list *old = addrs; write_lock(&server->lock); - rcu_swap_protected(server->addresses, old, - lockdep_is_held(&server->lock)); + old = rcu_replace_pointer(server->addresses, old, + lockdep_is_held(&server->lock)); write_unlock(&server->lock); - afs_put_addrlist(old); + afs_put_addrlist(old, afs_alist_trace_put_vlserver_old); } diff --git a/fs/afs/vl_probe.c b/fs/afs/vl_probe.c index f402ee8171a1..3d2e0c925460 100644 --- a/fs/afs/vl_probe.c +++ b/fs/afs/vl_probe.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS vlserver probing * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/sched.h> @@ -15,15 +11,33 @@ #include "internal.h" #include "protocol_yfs.h" -static bool afs_vl_probe_done(struct afs_vlserver *server) + +/* + * Handle the completion of a set of probes. + */ +static void afs_finished_vl_probe(struct afs_vlserver *server) { - if (!atomic_dec_and_test(&server->probe_outstanding)) - return false; + if (!(server->probe.flags & AFS_VLSERVER_PROBE_RESPONDED)) { + server->rtt = UINT_MAX; + clear_bit(AFS_VLSERVER_FL_RESPONDING, &server->flags); + } - wake_up_var(&server->probe_outstanding); clear_bit_unlock(AFS_VLSERVER_FL_PROBING, &server->flags); wake_up_bit(&server->flags, AFS_VLSERVER_FL_PROBING); - return true; +} + +/* + * Handle the completion of a probe RPC call. + */ +static void afs_done_one_vl_probe(struct afs_vlserver *server, bool wake_up) +{ + if (atomic_dec_and_test(&server->probe_outstanding)) { + afs_finished_vl_probe(server); + wake_up = true; + } + + if (wake_up) + wake_up_all(&server->probe_wq); } /* @@ -32,13 +46,13 @@ static bool afs_vl_probe_done(struct afs_vlserver *server) */ void afs_vlserver_probe_result(struct afs_call *call) { - struct afs_addr_list *alist = call->alist; - struct afs_vlserver *server = call->reply[0]; - unsigned int server_index = (long)call->reply[1]; - unsigned int index = call->addr_ix; - unsigned int rtt = UINT_MAX; + struct afs_addr_list *alist = call->vl_probe; + struct afs_vlserver *server = call->vlserver; + struct afs_address *addr = &alist->addrs[call->probe_index]; + unsigned int server_index = call->server_index; + unsigned int rtt_us = 0; + unsigned int index = call->probe_index; bool have_result = false; - u64 _rtt; int ret = call->error; _enter("%s,%u,%u,%d,%d", server->name, server_index, index, ret, call->abort_code); @@ -50,15 +64,20 @@ void afs_vlserver_probe_result(struct afs_call *call) server->probe.error = 0; goto responded; case -ECONNABORTED: - if (!server->probe.responded) { + if (!(server->probe.flags & AFS_VLSERVER_PROBE_RESPONDED)) { server->probe.abort_code = call->abort_code; server->probe.error = ret; } goto responded; case -ENOMEM: case -ENONET: - server->probe.local_failure = true; - afs_io_error(call, afs_io_error_vl_probe_fail); + case -EKEYEXPIRED: + case -EKEYREVOKED: + case -EKEYREJECTED: + server->probe.flags |= AFS_VLSERVER_PROBE_LOCAL_FAILURE; + if (server->probe.error == 0) + server->probe.error = ret; + trace_afs_io_error(call->debug_id, ret, afs_io_error_vl_probe_fail); goto out; case -ECONNRESET: /* Responded, but call expired. */ case -ERFKILL: @@ -71,61 +90,53 @@ void afs_vlserver_probe_result(struct afs_call *call) case -ETIME: default: clear_bit(index, &alist->responded); - set_bit(index, &alist->failed); - if (!server->probe.responded && + set_bit(index, &alist->probe_failed); + if (!(server->probe.flags & AFS_VLSERVER_PROBE_RESPONDED) && (server->probe.error == 0 || server->probe.error == -ETIMEDOUT || server->probe.error == -ETIME)) server->probe.error = ret; - afs_io_error(call, afs_io_error_vl_probe_fail); + trace_afs_io_error(call->debug_id, ret, afs_io_error_vl_probe_fail); goto out; } responded: set_bit(index, &alist->responded); - clear_bit(index, &alist->failed); + clear_bit(index, &alist->probe_failed); if (call->service_id == YFS_VL_SERVICE) { - server->probe.is_yfs = true; + server->probe.flags |= AFS_VLSERVER_PROBE_IS_YFS; set_bit(AFS_VLSERVER_FL_IS_YFS, &server->flags); - alist->addrs[index].srx_service = call->service_id; + server->service_id = call->service_id; } else { - server->probe.not_yfs = true; - if (!server->probe.is_yfs) { + server->probe.flags |= AFS_VLSERVER_PROBE_NOT_YFS; + if (!(server->probe.flags & AFS_VLSERVER_PROBE_IS_YFS)) { clear_bit(AFS_VLSERVER_FL_IS_YFS, &server->flags); - alist->addrs[index].srx_service = call->service_id; + server->service_id = call->service_id; } } - /* Get the RTT and scale it to fit into a 32-bit value that represents - * over a minute of time so that we can access it with one instruction - * on a 32-bit system. - */ - _rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall); - _rtt /= 64; - rtt = (_rtt > UINT_MAX) ? UINT_MAX : _rtt; - if (rtt < server->probe.rtt) { - server->probe.rtt = rtt; + rtt_us = rxrpc_kernel_get_srtt(addr->peer); + if (rtt_us < server->probe.rtt) { + server->probe.rtt = rtt_us; + server->rtt = rtt_us; alist->preferred = index; - have_result = true; } smp_wmb(); /* Set rtt before responded. */ - server->probe.responded = true; + server->probe.flags |= AFS_VLSERVER_PROBE_RESPONDED; set_bit(AFS_VLSERVER_FL_PROBED, &server->flags); + set_bit(AFS_VLSERVER_FL_RESPONDING, &server->flags); + have_result = true; out: spin_unlock(&server->probe_lock); - _debug("probe [%u][%u] %pISpc rtt=%u ret=%d", - server_index, index, &alist->addrs[index].transport, - (unsigned int)rtt, ret); + trace_afs_vl_probe(server, false, alist, index, call->error, call->abort_code, rtt_us); + _debug("probe [%u][%u] %pISpc rtt=%d ret=%d", + server_index, index, rxrpc_kernel_remote_addr(addr->peer), + rtt_us, ret); - have_result |= afs_vl_probe_done(server); - if (have_result) { - server->probe.have_result = true; - wake_up_var(&server->probe.have_result); - wake_up_all(&server->probe_wq); - } + afs_done_one_vl_probe(server, have_result); } /* @@ -138,34 +149,52 @@ static bool afs_do_probe_vlserver(struct afs_net *net, unsigned int server_index, struct afs_error *_e) { - struct afs_addr_cursor ac = { - .index = 0, - }; + struct afs_addr_list *alist; + struct afs_call *call; + unsigned long unprobed; + unsigned int index, i; bool in_progress = false; - int err; + int best_prio; _enter("%s", server->name); read_lock(&server->lock); - ac.alist = rcu_dereference_protected(server->addresses, - lockdep_is_held(&server->lock)); + alist = rcu_dereference_protected(server->addresses, + lockdep_is_held(&server->lock)); + afs_get_addrlist(alist, afs_alist_trace_get_vlprobe); read_unlock(&server->lock); - atomic_set(&server->probe_outstanding, ac.alist->nr_addrs); + atomic_set(&server->probe_outstanding, alist->nr_addrs); memset(&server->probe, 0, sizeof(server->probe)); server->probe.rtt = UINT_MAX; - for (ac.index = 0; ac.index < ac.alist->nr_addrs; ac.index++) { - err = afs_vl_get_capabilities(net, &ac, key, server, - server_index, true); - if (err == -EINPROGRESS) + unprobed = (1UL << alist->nr_addrs) - 1; + while (unprobed) { + best_prio = -1; + index = 0; + for (i = 0; i < alist->nr_addrs; i++) { + if (test_bit(i, &unprobed) && + alist->addrs[i].prio > best_prio) { + index = i; + best_prio = alist->addrs[i].prio; + } + } + __clear_bit(index, &unprobed); + + trace_afs_vl_probe(server, true, alist, index, 0, 0, 0); + call = afs_vl_get_capabilities(net, alist, index, key, server, + server_index); + if (!IS_ERR(call)) { + afs_prioritise_error(_e, call->error, call->abort_code); + afs_put_call(call); in_progress = true; - else - afs_prioritise_error(_e, err, ac.abort_code); + } else { + afs_prioritise_error(_e, PTR_ERR(call), 0); + afs_done_one_vl_probe(server, false); + } } - if (!in_progress) - afs_vl_probe_done(server); + afs_put_addrlist(alist, afs_alist_trace_put_vlprobe); return in_progress; } @@ -176,12 +205,10 @@ int afs_send_vl_probes(struct afs_net *net, struct key *key, struct afs_vlserver_list *vllist) { struct afs_vlserver *server; - struct afs_error e; + struct afs_error e = {}; bool in_progress = false; int i; - e.error = 0; - e.responded = false; for (i = 0; i < vllist->nr_servers; i++) { server = vllist->servers[i].server; if (test_bit(AFS_VLSERVER_FL_PROBED, &server->flags)) @@ -203,7 +230,7 @@ int afs_wait_for_vl_probes(struct afs_vlserver_list *vllist, { struct wait_queue_entry *waits; struct afs_vlserver *server; - unsigned int rtt = UINT_MAX; + unsigned int rtt = UINT_MAX, rtt_s; bool have_responders = false; int pref = -1, i; @@ -215,7 +242,7 @@ int afs_wait_for_vl_probes(struct afs_vlserver_list *vllist, server = vllist->servers[i].server; if (!test_bit(AFS_VLSERVER_FL_PROBING, &server->flags)) __clear_bit(i, &untried); - if (server->probe.responded) + if (server->probe.flags & AFS_VLSERVER_PROBE_RESPONDED) have_responders = true; } } @@ -241,7 +268,7 @@ int afs_wait_for_vl_probes(struct afs_vlserver_list *vllist, for (i = 0; i < vllist->nr_servers; i++) { if (test_bit(i, &untried)) { server = vllist->servers[i].server; - if (server->probe.responded) + if (server->probe.flags & AFS_VLSERVER_PROBE_RESPONDED) goto stop; if (test_bit(AFS_VLSERVER_FL_PROBING, &server->flags)) still_probing = true; @@ -259,10 +286,11 @@ stop: for (i = 0; i < vllist->nr_servers; i++) { if (test_bit(i, &untried)) { server = vllist->servers[i].server; - if (server->probe.responded && - server->probe.rtt < rtt) { + rtt_s = READ_ONCE(server->rtt); + if (test_bit(AFS_VLSERVER_FL_RESPONDING, &server->flags) && + rtt_s < rtt) { pref = i; - rtt = server->probe.rtt; + rtt = rtt_s; } remove_wait_queue(&server->probe_wq, &waits[i]); diff --git a/fs/afs/vl_rotate.c b/fs/afs/vl_rotate.c index 7adde83a0648..6ad9688d8f4b 100644 --- a/fs/afs/vl_rotate.c +++ b/fs/afs/vl_rotate.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Handle vlserver selection and rotation. * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/kernel.h> @@ -21,18 +17,21 @@ bool afs_begin_vlserver_operation(struct afs_vl_cursor *vc, struct afs_cell *cell, struct key *key) { + static atomic_t debug_ids; + memset(vc, 0, sizeof(*vc)); vc->cell = cell; vc->key = key; - vc->error = -EDESTADDRREQ; - vc->ac.error = SHRT_MAX; + vc->cumul_error.error = -EDESTADDRREQ; + vc->nr_iterations = -1; if (signal_pending(current)) { - vc->error = -EINTR; + vc->cumul_error.error = -EINTR; vc->flags |= AFS_VL_CURSOR_STOP; return false; } + vc->debug_id = atomic_inc_return(&debug_ids); return true; } @@ -43,11 +42,35 @@ bool afs_begin_vlserver_operation(struct afs_vl_cursor *vc, struct afs_cell *cel static bool afs_start_vl_iteration(struct afs_vl_cursor *vc) { struct afs_cell *cell = vc->cell; + unsigned int dns_lookup_count; + + if (cell->dns_source == DNS_RECORD_UNAVAILABLE || + cell->dns_expiry <= ktime_get_real_seconds()) { + dns_lookup_count = smp_load_acquire(&cell->dns_lookup_count); + set_bit(AFS_CELL_FL_DO_LOOKUP, &cell->flags); + afs_queue_cell(cell, afs_cell_trace_queue_dns); + + if (cell->dns_source == DNS_RECORD_UNAVAILABLE) { + if (wait_var_event_interruptible( + &cell->dns_lookup_count, + smp_load_acquire(&cell->dns_lookup_count) + != dns_lookup_count) < 0) { + vc->cumul_error.error = -ERESTARTSYS; + return false; + } + } - if (wait_on_bit(&cell->flags, AFS_CELL_FL_NO_LOOKUP_YET, - TASK_INTERRUPTIBLE)) { - vc->error = -ERESTARTSYS; - return false; + /* Status load is ordered after lookup counter load */ + if (cell->dns_status == DNS_LOOKUP_GOT_NOT_FOUND) { + pr_warn("No record of cell %s\n", cell->name); + vc->cumul_error.error = -ENOENT; + return false; + } + + if (cell->dns_source == DNS_RECORD_UNAVAILABLE) { + vc->cumul_error.error = -EDESTADDRREQ; + return false; + } } read_lock(&cell->vl_servers_lock); @@ -55,11 +78,11 @@ static bool afs_start_vl_iteration(struct afs_vl_cursor *vc) rcu_dereference_protected(cell->vl_servers, lockdep_is_held(&cell->vl_servers_lock))); read_unlock(&cell->vl_servers_lock); - if (!vc->server_list || !vc->server_list->nr_servers) + if (!vc->server_list->nr_servers) return false; - vc->untried = (1UL << vc->server_list->nr_servers) - 1; - vc->index = -1; + vc->untried_servers = (1UL << vc->server_list->nr_servers) - 1; + vc->server_index = -1; return true; } @@ -69,54 +92,57 @@ static bool afs_start_vl_iteration(struct afs_vl_cursor *vc) */ bool afs_select_vlserver(struct afs_vl_cursor *vc) { - struct afs_addr_list *alist; + struct afs_addr_list *alist = vc->alist; struct afs_vlserver *vlserver; - struct afs_error e; - u32 rtt; - int error = vc->ac.error, i; + unsigned long set, failed; + unsigned int rtt; + s32 abort_code = vc->call_abort_code; + int error = vc->call_error, i; + + vc->nr_iterations++; - _enter("%lx[%d],%lx[%d],%d,%d", - vc->untried, vc->index, - vc->ac.tried, vc->ac.index, - error, vc->ac.abort_code); + _enter("VC=%x+%x,%d{%lx},%d{%lx},%d,%d", + vc->debug_id, vc->nr_iterations, vc->server_index, vc->untried_servers, + vc->addr_index, vc->addr_tried, + error, abort_code); if (vc->flags & AFS_VL_CURSOR_STOP) { _leave(" = f [stopped]"); return false; } - vc->nr_iterations++; + if (vc->nr_iterations == 0) + goto start; + + WRITE_ONCE(alist->addrs[vc->addr_index].last_error, error); /* Evaluate the result of the previous operation, if there was one. */ switch (error) { - case SHRT_MAX: - goto start; - default: case 0: /* Success or local failure. Stop. */ - vc->error = error; + vc->cumul_error.error = error; vc->flags |= AFS_VL_CURSOR_STOP; - _leave(" = f [okay/local %d]", vc->ac.error); + _leave(" = f [okay/local %d]", vc->cumul_error.error); return false; case -ECONNABORTED: /* The far side rejected the operation on some grounds. This * might involve the server being busy or the volume having been moved. */ - switch (vc->ac.abort_code) { + switch (abort_code) { case AFSVL_IO: case AFSVL_BADVOLOPER: case AFSVL_NOMEM: /* The server went weird. */ - vc->error = -EREMOTEIO; + afs_prioritise_error(&vc->cumul_error, -EREMOTEIO, abort_code); //write_lock(&vc->cell->vl_servers_lock); - //vc->server_list->weird_mask |= 1 << vc->index; + //vc->server_list->weird_mask |= 1 << vc->server_index; //write_unlock(&vc->cell->vl_servers_lock); goto next_server; default: - vc->error = afs_abort_to_error(vc->ac.abort_code); + afs_prioritise_error(&vc->cumul_error, error, abort_code); goto failed; } @@ -129,19 +155,29 @@ bool afs_select_vlserver(struct afs_vl_cursor *vc) case -ETIMEDOUT: case -ETIME: _debug("no conn %d", error); - vc->error = error; + afs_prioritise_error(&vc->cumul_error, error, 0); goto iterate_address; case -ECONNRESET: _debug("call reset"); - vc->error = error; + afs_prioritise_error(&vc->cumul_error, error, 0); vc->flags |= AFS_VL_CURSOR_RETRY; goto next_server; + + case -EOPNOTSUPP: + _debug("notsupp"); + goto next_server; } restart_from_beginning: _debug("restart"); - afs_end_cursor(&vc->ac); + if (vc->call_responded && + vc->addr_index != vc->alist->preferred && + test_bit(alist->preferred, &vc->addr_tried)) + WRITE_ONCE(alist->preferred, vc->addr_index); + afs_put_addrlist(alist, afs_alist_trace_put_vlrotate_restart); + alist = vc->alist = NULL; + afs_put_vlserverlist(vc->cell->net, vc->server_list); vc->server_list = NULL; if (vc->flags & AFS_VL_CURSOR_RETRIED) @@ -149,52 +185,58 @@ restart_from_beginning: vc->flags |= AFS_VL_CURSOR_RETRIED; start: _debug("start"); + ASSERTCMP(alist, ==, NULL); if (!afs_start_vl_iteration(vc)) goto failed; error = afs_send_vl_probes(vc->cell->net, vc->key, vc->server_list); - if (error < 0) - goto failed_set_error; + if (error < 0) { + afs_prioritise_error(&vc->cumul_error, error, 0); + goto failed; + } pick_server: - _debug("pick [%lx]", vc->untried); + _debug("pick [%lx]", vc->untried_servers); + ASSERTCMP(alist, ==, NULL); - error = afs_wait_for_vl_probes(vc->server_list, vc->untried); - if (error < 0) - goto failed_set_error; + error = afs_wait_for_vl_probes(vc->server_list, vc->untried_servers); + if (error < 0) { + afs_prioritise_error(&vc->cumul_error, error, 0); + goto failed; + } /* Pick the untried server with the lowest RTT. */ - vc->index = vc->server_list->preferred; - if (test_bit(vc->index, &vc->untried)) + vc->server_index = vc->server_list->preferred; + if (test_bit(vc->server_index, &vc->untried_servers)) goto selected_server; - vc->index = -1; - rtt = U32_MAX; + vc->server_index = -1; + rtt = UINT_MAX; for (i = 0; i < vc->server_list->nr_servers; i++) { struct afs_vlserver *s = vc->server_list->servers[i].server; - if (!test_bit(i, &vc->untried) || !s->probe.responded) + if (!test_bit(i, &vc->untried_servers) || + !test_bit(AFS_VLSERVER_FL_RESPONDING, &s->flags)) continue; - if (s->probe.rtt < rtt) { - vc->index = i; + if (s->probe.rtt <= rtt) { + vc->server_index = i; rtt = s->probe.rtt; } } - if (vc->index == -1) + if (vc->server_index == -1) goto no_more_servers; selected_server: - _debug("use %d", vc->index); - __clear_bit(vc->index, &vc->untried); + _debug("use %d", vc->server_index); + __clear_bit(vc->server_index, &vc->untried_servers); /* We're starting on a different vlserver from the list. We need to * check it, find its address list and probe its capabilities before we * use it. */ - ASSERTCMP(vc->ac.alist, ==, NULL); - vlserver = vc->server_list->servers[vc->index].server; + vlserver = vc->server_list->servers[vc->server_index].server; vc->server = vlserver; _debug("USING VLSERVER: %s", vlserver->name); @@ -202,34 +244,48 @@ selected_server: read_lock(&vlserver->lock); alist = rcu_dereference_protected(vlserver->addresses, lockdep_is_held(&vlserver->lock)); - afs_get_addrlist(alist); + vc->alist = afs_get_addrlist(alist, afs_alist_trace_get_vlrotate_set); read_unlock(&vlserver->lock); - memset(&vc->ac, 0, sizeof(vc->ac)); - - if (!vc->ac.alist) - vc->ac.alist = alist; - else - afs_put_addrlist(alist); - - vc->ac.index = -1; + vc->addr_tried = 0; + vc->addr_index = -1; iterate_address: - ASSERT(vc->ac.alist); /* Iterate over the current server's address list to try and find an * address on which it will respond to us. */ - if (!afs_iterate_addresses(&vc->ac)) + set = READ_ONCE(alist->responded); + failed = READ_ONCE(alist->probe_failed); + vc->addr_index = READ_ONCE(alist->preferred); + + _debug("%lx-%lx-%lx,%d", set, failed, vc->addr_tried, vc->addr_index); + + set &= ~(failed | vc->addr_tried); + + if (!set) goto next_server; - _debug("VL address %d/%d", vc->ac.index, vc->ac.alist->nr_addrs); + if (!test_bit(vc->addr_index, &set)) + vc->addr_index = __ffs(set); + + set_bit(vc->addr_index, &vc->addr_tried); + vc->alist = alist; - _leave(" = t %pISpc", &vc->ac.alist->addrs[vc->ac.index].transport); + _debug("VL address %d/%d", vc->addr_index, alist->nr_addrs); + + vc->call_responded = false; + _leave(" = t %pISpc", rxrpc_kernel_remote_addr(alist->addrs[vc->addr_index].peer)); return true; next_server: _debug("next"); - afs_end_cursor(&vc->ac); + ASSERT(alist); + if (vc->call_responded && + vc->addr_index != alist->preferred && + test_bit(alist->preferred, &vc->addr_tried)) + WRITE_ONCE(alist->preferred, vc->addr_index); + afs_put_addrlist(alist, afs_alist_trace_put_vlrotate_next); + alist = vc->alist = NULL; goto pick_server; no_more_servers: @@ -239,21 +295,26 @@ no_more_servers: if (vc->flags & AFS_VL_CURSOR_RETRY) goto restart_from_beginning; - e.error = -EDESTADDRREQ; - e.responded = false; for (i = 0; i < vc->server_list->nr_servers; i++) { struct afs_vlserver *s = vc->server_list->servers[i].server; - afs_prioritise_error(&e, READ_ONCE(s->probe.error), + if (test_bit(AFS_VLSERVER_FL_RESPONDING, &s->flags)) + vc->cumul_error.responded = true; + afs_prioritise_error(&vc->cumul_error, READ_ONCE(s->probe.error), s->probe.abort_code); } -failed_set_error: - vc->error = error; failed: + if (alist) { + if (vc->call_responded && + vc->addr_index != alist->preferred && + test_bit(alist->preferred, &vc->addr_tried)) + WRITE_ONCE(alist->preferred, vc->addr_index); + afs_put_addrlist(alist, afs_alist_trace_put_vlrotate_fail); + alist = vc->alist = NULL; + } vc->flags |= AFS_VL_CURSOR_STOP; - afs_end_cursor(&vc->ac); - _leave(" = f [failed %d]", vc->error); + _leave(" = f [failed %d]", vc->cumul_error.error); return false; } @@ -262,6 +323,7 @@ failed: */ static void afs_vl_dump_edestaddrreq(const struct afs_vl_cursor *vc) { + struct afs_cell *cell = vc->cell; static int count; int i; @@ -271,8 +333,14 @@ static void afs_vl_dump_edestaddrreq(const struct afs_vl_cursor *vc) rcu_read_lock(); pr_notice("EDESTADDR occurred\n"); + pr_notice("CELL: %s err=%d\n", cell->name, cell->error); + pr_notice("DNS: src=%u st=%u lc=%x\n", + cell->dns_source, cell->dns_status, cell->dns_lookup_count); pr_notice("VC: ut=%lx ix=%u ni=%hu fl=%hx err=%hd\n", - vc->untried, vc->index, vc->nr_iterations, vc->flags, vc->error); + vc->untried_servers, vc->server_index, vc->nr_iterations, + vc->flags, vc->cumul_error.error); + pr_notice("VC: call er=%d ac=%d r=%u\n", + vc->call_error, vc->call_abort_code, vc->call_responded); if (vc->server_list) { const struct afs_vlserver_list *sl = vc->server_list; @@ -288,17 +356,15 @@ static void afs_vl_dump_edestaddrreq(const struct afs_vl_cursor *vc) pr_notice("VC: - nr=%u/%u/%u pf=%u\n", a->nr_ipv4, a->nr_addrs, a->max_addrs, a->preferred); - pr_notice("VC: - pr=%lx R=%lx F=%lx\n", - a->probed, a->responded, a->failed); - if (a == vc->ac.alist) + pr_notice("VC: - R=%lx F=%lx\n", + a->responded, a->probe_failed); + if (a == vc->alist) pr_notice("VC: - current\n"); } } } - pr_notice("AC: t=%lx ax=%u ac=%d er=%d r=%u ni=%u\n", - vc->ac.tried, vc->ac.index, vc->ac.abort_code, vc->ac.error, - vc->ac.responded, vc->ac.nr_iterations); + pr_notice("AC: t=%lx ax=%u\n", vc->addr_tried, vc->addr_index); rcu_read_unlock(); } @@ -309,17 +375,25 @@ int afs_end_vlserver_operation(struct afs_vl_cursor *vc) { struct afs_net *net = vc->cell->net; - if (vc->error == -EDESTADDRREQ || - vc->error == -EADDRNOTAVAIL || - vc->error == -ENETUNREACH || - vc->error == -EHOSTUNREACH) + _enter("VC=%x+%x", vc->debug_id, vc->nr_iterations); + + switch (vc->cumul_error.error) { + case -EDESTADDRREQ: + case -EADDRNOTAVAIL: + case -ENETUNREACH: + case -EHOSTUNREACH: afs_vl_dump_edestaddrreq(vc); + break; + } - afs_end_cursor(&vc->ac); + if (vc->alist) { + if (vc->call_responded && + vc->addr_index != vc->alist->preferred && + test_bit(vc->alist->preferred, &vc->addr_tried)) + WRITE_ONCE(vc->alist->preferred, vc->addr_index); + afs_put_addrlist(vc->alist, afs_alist_trace_put_vlrotate_end); + vc->alist = NULL; + } afs_put_vlserverlist(net, vc->server_list); - - if (vc->error == -ECONNABORTED) - vc->error = afs_abort_to_error(vc->ac.abort_code); - - return vc->error; + return vc->cumul_error.error; } diff --git a/fs/afs/vlclient.c b/fs/afs/vlclient.c index c3d9e5a5f67e..3a23c0b08eb6 100644 --- a/fs/afs/vlclient.c +++ b/fs/afs/vlclient.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS Volume Location Service client * * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/gfp.h> @@ -22,8 +18,7 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) { struct afs_uvldbentry__xdr *uvldb; struct afs_vldb_entry *entry; - bool new_only = false; - u32 tmp, nr_servers, vlflags; + u32 nr_servers, vlflags; int i, ret; _enter(""); @@ -34,7 +29,7 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) /* unmarshall the reply once we've received all of it */ uvldb = call->buffer; - entry = call->reply[0]; + entry = call->ret_vldb; nr_servers = ntohl(uvldb->nServers); if (nr_servers > AFS_NMAXNSERVERS) @@ -45,38 +40,26 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) entry->name[i] = 0; entry->name_len = strlen(entry->name); - /* If there is a new replication site that we can use, ignore all the - * sites that aren't marked as new. - */ - for (i = 0; i < nr_servers; i++) { - tmp = ntohl(uvldb->serverFlags[i]); - if (!(tmp & AFS_VLSF_DONTUSE) && - (tmp & AFS_VLSF_NEWREPSITE)) - new_only = true; - } - vlflags = ntohl(uvldb->flags); for (i = 0; i < nr_servers; i++) { struct afs_uuid__xdr *xdr; struct afs_uuid *uuid; + u32 tmp = ntohl(uvldb->serverFlags[i]); int j; + int n = entry->nr_servers; - tmp = ntohl(uvldb->serverFlags[i]); - if (tmp & AFS_VLSF_DONTUSE || - (new_only && !(tmp & AFS_VLSF_NEWREPSITE))) - continue; if (tmp & AFS_VLSF_RWVOL) { - entry->fs_mask[i] |= AFS_VOL_VTM_RW; + entry->fs_mask[n] |= AFS_VOL_VTM_RW; if (vlflags & AFS_VLF_BACKEXISTS) - entry->fs_mask[i] |= AFS_VOL_VTM_BAK; + entry->fs_mask[n] |= AFS_VOL_VTM_BAK; } if (tmp & AFS_VLSF_ROVOL) - entry->fs_mask[i] |= AFS_VOL_VTM_RO; - if (!entry->fs_mask[i]) + entry->fs_mask[n] |= AFS_VOL_VTM_RO; + if (!entry->fs_mask[n]) continue; xdr = &uvldb->serverNumber[i]; - uuid = (struct afs_uuid *)&entry->fs_server[i]; + uuid = (struct afs_uuid *)&entry->fs_server[n]; uuid->time_low = xdr->time_low; uuid->time_mid = htons(ntohl(xdr->time_mid)); uuid->time_hi_and_version = htons(ntohl(xdr->time_hi_and_version)); @@ -85,6 +68,8 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) for (j = 0; j < 6; j++) uuid->node[j] = (u8)ntohl(xdr->node[j]); + entry->vlsf_flags[n] = tmp; + entry->addr_version[n] = ntohl(uvldb->serverUnique[i]); entry->nr_servers++; } @@ -108,12 +93,6 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) return 0; } -static void afs_destroy_vl_get_entry_by_name_u(struct afs_call *call) -{ - kfree(call->reply[0]); - afs_flat_call_destructor(call); -} - /* * VL.GetEntryByNameU operation type. */ @@ -121,7 +100,7 @@ static const struct afs_call_type afs_RXVLGetEntryByNameU = { .name = "VL.GetEntryByNameU", .op = afs_VL_GetEntryByNameU, .deliver = afs_deliver_vl_get_entry_by_name_u, - .destructor = afs_destroy_vl_get_entry_by_name_u, + .destructor = afs_flat_call_destructor, }; /* @@ -155,8 +134,10 @@ struct afs_vldb_entry *afs_vl_get_entry_by_name_u(struct afs_vl_cursor *vc, } call->key = vc->key; - call->reply[0] = entry; - call->ret_reply0 = true; + call->ret_vldb = entry; + call->max_lifespan = AFS_VL_MAX_LIFESPAN; + call->peer = rxrpc_kernel_get_peer(vc->alist->addrs[vc->addr_index].peer); + call->service_id = vc->server->service_id; /* Marshall the parameters */ bp = call->request; @@ -167,7 +148,17 @@ struct afs_vldb_entry *afs_vl_get_entry_by_name_u(struct afs_vl_cursor *vc, memset((void *)bp + volnamesz, 0, padsz); trace_afs_make_vl_call(call); - return (struct afs_vldb_entry *)afs_make_call(&vc->ac, call, GFP_KERNEL, false); + afs_make_call(call, GFP_KERNEL); + afs_wait_for_call_to_complete(call); + vc->call_abort_code = call->abort_code; + vc->call_error = call->error; + vc->call_responded = call->responded; + afs_put_call(call); + if (vc->call_error) { + kfree(entry); + return ERR_PTR(vc->call_error); + } + return entry; } /* @@ -187,7 +178,7 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call) int i, ret; _enter("{%u,%zu/%u}", - call->unmarshall, iov_iter_count(call->_iter), call->count); + call->unmarshall, iov_iter_count(call->iter), call->count); switch (call->unmarshall) { case 0: @@ -195,7 +186,9 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call) sizeof(struct afs_uuid__xdr) + 3 * sizeof(__be32)); call->unmarshall++; - /* Extract the returned uuid, uniquifier, nentries and blkaddrs size */ + /* Extract the returned uuid, uniquifier, nentries and + * blkaddrs size */ + fallthrough; case 1: ret = afs_extract_data(call, true); if (ret < 0) @@ -207,11 +200,11 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call) count = ntohl(*bp); nentries = min(nentries, count); - alist = afs_alloc_addrlist(nentries, FS_SERVICE, AFS_FS_PORT); + alist = afs_alloc_addrlist(nentries); if (!alist) return -ENOMEM; alist->version = uniquifier; - call->reply[0] = alist; + call->ret_alist = alist; call->count = count; call->count2 = nentries; call->unmarshall++; @@ -220,18 +213,22 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call) count = min(call->count, 4U); afs_extract_to_buf(call, count * sizeof(__be32)); - /* Extract entries */ + fallthrough; /* and extract entries */ case 2: ret = afs_extract_data(call, call->count > 4); if (ret < 0) return ret; - alist = call->reply[0]; + alist = call->ret_alist; bp = call->buffer; count = min(call->count, 4U); - for (i = 0; i < count; i++) - if (alist->nr_addrs < call->count2) - afs_merge_fs_addr4(alist, *bp++, AFS_FS_PORT); + for (i = 0; i < count; i++) { + if (alist->nr_addrs < call->count2) { + ret = afs_merge_fs_addr4(call->net, alist, *bp++, AFS_FS_PORT); + if (ret < 0) + return ret; + } + } call->count -= count; if (call->count > 0) @@ -244,13 +241,6 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call) return 0; } -static void afs_vl_get_addrs_u_destructor(struct afs_call *call) -{ - afs_put_server(call->net, (struct afs_server *)call->reply[0]); - kfree(call->reply[1]); - return afs_flat_call_destructor(call); -} - /* * VL.GetAddrsU operation type. */ @@ -258,7 +248,7 @@ static const struct afs_call_type afs_RXVLGetAddrsU = { .name = "VL.GetAddrsU", .op = afs_VL_GetAddrsU, .deliver = afs_deliver_vl_get_addrs_u, - .destructor = afs_vl_get_addrs_u_destructor, + .destructor = afs_flat_call_destructor, }; /* @@ -269,6 +259,7 @@ struct afs_addr_list *afs_vl_get_addrs_u(struct afs_vl_cursor *vc, const uuid_t *uuid) { struct afs_ListAddrByAttributes__xdr *r; + struct afs_addr_list *alist; const struct afs_uuid *u = (const struct afs_uuid *)uuid; struct afs_call *call; struct afs_net *net = vc->cell->net; @@ -284,8 +275,10 @@ struct afs_addr_list *afs_vl_get_addrs_u(struct afs_vl_cursor *vc, return ERR_PTR(-ENOMEM); call->key = vc->key; - call->reply[0] = NULL; - call->ret_reply0 = true; + call->ret_alist = NULL; + call->max_lifespan = AFS_VL_MAX_LIFESPAN; + call->peer = rxrpc_kernel_get_peer(vc->alist->addrs[vc->addr_index].peer); + call->service_id = vc->server->service_id; /* Marshall the parameters */ bp = call->request; @@ -304,7 +297,18 @@ struct afs_addr_list *afs_vl_get_addrs_u(struct afs_vl_cursor *vc, r->uuid.node[i] = htonl(u->node[i]); trace_afs_make_vl_call(call); - return (struct afs_addr_list *)afs_make_call(&vc->ac, call, GFP_KERNEL, false); + afs_make_call(call, GFP_KERNEL); + afs_wait_for_call_to_complete(call); + vc->call_abort_code = call->abort_code; + vc->call_error = call->error; + vc->call_responded = call->responded; + alist = call->ret_alist; + afs_put_call(call); + if (vc->call_error) { + afs_put_addrlist(alist, afs_alist_trace_put_getaddru); + return ERR_PTR(vc->call_error); + } + return alist; } /* @@ -316,14 +320,14 @@ static int afs_deliver_vl_get_capabilities(struct afs_call *call) int ret; _enter("{%u,%zu/%u}", - call->unmarshall, iov_iter_count(call->_iter), call->count); + call->unmarshall, iov_iter_count(call->iter), call->count); switch (call->unmarshall) { case 0: afs_extract_to_tmp(call); call->unmarshall++; - /* Extract the capabilities word count */ + fallthrough; /* and extract the capabilities word count */ case 1: ret = afs_extract_data(call, true); if (ret < 0) @@ -336,7 +340,7 @@ static int afs_deliver_vl_get_capabilities(struct afs_call *call) call->unmarshall++; afs_extract_discard(call, count * sizeof(__be32)); - /* Extract capabilities words */ + fallthrough; /* and extract capabilities words */ case 2: ret = afs_extract_data(call, false); if (ret < 0) @@ -354,9 +358,8 @@ static int afs_deliver_vl_get_capabilities(struct afs_call *call) static void afs_destroy_vl_get_capabilities(struct afs_call *call) { - struct afs_vlserver *server = call->reply[0]; - - afs_put_vlserver(call->net, server); + afs_put_addrlist(call->vl_probe, afs_alist_trace_put_vlgetcaps); + afs_put_vlserver(call->net, call->vlserver); afs_flat_call_destructor(call); } @@ -367,6 +370,7 @@ static const struct afs_call_type afs_RXVLGetCapabilities = { .name = "VL.GetCapabilities", .op = afs_VL_GetCapabilities, .deliver = afs_deliver_vl_get_capabilities, + .immediate_cancel = afs_vlserver_probe_result, .done = afs_vlserver_probe_result, .destructor = afs_destroy_vl_get_capabilities, }; @@ -378,12 +382,12 @@ static const struct afs_call_type afs_RXVLGetCapabilities = { * We use this to probe for service upgrade to determine what the server at the * other end supports. */ -int afs_vl_get_capabilities(struct afs_net *net, - struct afs_addr_cursor *ac, - struct key *key, - struct afs_vlserver *server, - unsigned int server_index, - bool async) +struct afs_call *afs_vl_get_capabilities(struct afs_net *net, + struct afs_addr_list *alist, + unsigned int addr_index, + struct key *key, + struct afs_vlserver *server, + unsigned int server_index) { struct afs_call *call; __be32 *bp; @@ -392,13 +396,18 @@ int afs_vl_get_capabilities(struct afs_net *net, call = afs_alloc_flat_call(net, &afs_RXVLGetCapabilities, 1 * 4, 16 * 4); if (!call) - return -ENOMEM; + return ERR_PTR(-ENOMEM); call->key = key; - call->reply[0] = afs_get_vlserver(server); - call->reply[1] = (void *)(long)server_index; + call->vlserver = afs_get_vlserver(server); + call->server_index = server_index; + call->peer = rxrpc_kernel_get_peer(alist->addrs[addr_index].peer); + call->vl_probe = afs_get_addrlist(alist, afs_alist_trace_get_vlgetcaps); + call->probe_index = addr_index; + call->service_id = server->service_id; call->upgrade = true; - call->want_reply_time = true; + call->async = true; + call->max_lifespan = AFS_PROBE_MAX_LIFESPAN; /* marshall the parameters */ bp = call->request; @@ -406,7 +415,8 @@ int afs_vl_get_capabilities(struct afs_net *net, /* Can't take a ref on server */ trace_afs_make_vl_call(call); - return afs_make_call(ac, call, GFP_KERNEL, async); + afs_make_call(call, GFP_KERNEL); + return call; } /* @@ -426,7 +436,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) int ret; _enter("{%u,%zu,%u}", - call->unmarshall, iov_iter_count(call->_iter), call->count2); + call->unmarshall, iov_iter_count(call->iter), call->count2); switch (call->unmarshall) { case 0: @@ -436,6 +446,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) /* Extract the returned uuid, uniquifier, fsEndpoints count and * either the first fsEndpoint type or the volEndpoints * count if there are no fsEndpoints. */ + fallthrough; case 1: ret = afs_extract_data(call, true); if (ret < 0) @@ -447,14 +458,13 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) call->count2 = ntohl(*bp); /* Type or next count */ if (call->count > YFS_MAXENDPOINTS) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_fsendpt_num); + return afs_protocol_error(call, afs_eproto_yvl_fsendpt_num); - alist = afs_alloc_addrlist(call->count, FS_SERVICE, AFS_FS_PORT); + alist = afs_alloc_addrlist(call->count); if (!alist) return -ENOMEM; alist->version = uniquifier; - call->reply[0] = alist; + call->ret_alist = alist; if (call->count == 0) goto extract_volendpoints; @@ -468,40 +478,42 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) size = sizeof(__be32) * (1 + 4 + 1); break; default: - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_fsendpt_type); + return afs_protocol_error(call, afs_eproto_yvl_fsendpt_type); } size += sizeof(__be32); afs_extract_to_buf(call, size); call->unmarshall = 2; - /* Extract fsEndpoints[] entries */ + fallthrough; /* and extract fsEndpoints[] entries */ case 2: ret = afs_extract_data(call, true); if (ret < 0) return ret; - alist = call->reply[0]; + alist = call->ret_alist; bp = call->buffer; switch (call->count2) { case YFS_ENDPOINT_IPV4: if (ntohl(bp[0]) != sizeof(__be32) * 2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_fsendpt4_len); - afs_merge_fs_addr4(alist, bp[1], ntohl(bp[2])); + return afs_protocol_error( + call, afs_eproto_yvl_fsendpt4_len); + ret = afs_merge_fs_addr4(call->net, alist, bp[1], ntohl(bp[2])); + if (ret < 0) + return ret; bp += 3; break; case YFS_ENDPOINT_IPV6: if (ntohl(bp[0]) != sizeof(__be32) * 5) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_fsendpt6_len); - afs_merge_fs_addr6(alist, bp + 1, ntohl(bp[5])); + return afs_protocol_error( + call, afs_eproto_yvl_fsendpt6_len); + ret = afs_merge_fs_addr6(call->net, alist, bp + 1, ntohl(bp[5])); + if (ret < 0) + return ret; bp += 6; break; default: - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_fsendpt_type); + return afs_protocol_error(call, afs_eproto_yvl_fsendpt_type); } /* Got either the type of the next entry or the count of @@ -519,8 +531,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) if (!call->count) goto end; if (call->count > YFS_MAXENDPOINTS) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_vlendpt_type); + return afs_protocol_error(call, afs_eproto_yvl_vlendpt_type); afs_extract_to_buf(call, 1 * sizeof(__be32)); call->unmarshall = 3; @@ -529,6 +540,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) * extract the type of the next endpoint when we extract the * data of the current one, but this is the first... */ + fallthrough; case 3: ret = afs_extract_data(call, true); if (ret < 0) @@ -546,8 +558,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) size = sizeof(__be32) * (1 + 4 + 1); break; default: - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_vlendpt_type); + return afs_protocol_error(call, afs_eproto_yvl_vlendpt_type); } if (call->count > 1) @@ -555,7 +566,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) afs_extract_to_buf(call, size); call->unmarshall = 4; - /* Extract volEndpoints[] entries */ + fallthrough; /* and extract volEndpoints[] entries */ case 4: ret = afs_extract_data(call, true); if (ret < 0) @@ -565,19 +576,18 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) switch (call->count2) { case YFS_ENDPOINT_IPV4: if (ntohl(bp[0]) != sizeof(__be32) * 2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_vlendpt4_len); + return afs_protocol_error( + call, afs_eproto_yvl_vlendpt4_len); bp += 3; break; case YFS_ENDPOINT_IPV6: if (ntohl(bp[0]) != sizeof(__be32) * 5) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_vlendpt6_len); + return afs_protocol_error( + call, afs_eproto_yvl_vlendpt6_len); bp += 6; break; default: - return afs_protocol_error(call, -EBADMSG, - afs_eproto_yvl_vlendpt_type); + return afs_protocol_error(call, afs_eproto_yvl_vlendpt_type); } /* Got either the type of the next entry or the count of @@ -591,18 +601,18 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call) afs_extract_discard(call, 0); call->unmarshall = 5; - /* Done */ + fallthrough; /* Done */ case 5: ret = afs_extract_data(call, false); if (ret < 0) return ret; call->unmarshall = 6; + fallthrough; case 6: break; } - alist = call->reply[0]; _leave(" = 0 [done]"); return 0; } @@ -614,7 +624,7 @@ static const struct afs_call_type afs_YFSVLGetEndpoints = { .name = "YFSVL.GetEndpoints", .op = afs_YFSVL_GetEndpoints, .deliver = afs_deliver_yfsvl_get_endpoints, - .destructor = afs_vl_get_addrs_u_destructor, + .destructor = afs_flat_call_destructor, }; /* @@ -624,6 +634,7 @@ static const struct afs_call_type afs_YFSVLGetEndpoints = { struct afs_addr_list *afs_yfsvl_get_endpoints(struct afs_vl_cursor *vc, const uuid_t *uuid) { + struct afs_addr_list *alist; struct afs_call *call; struct afs_net *net = vc->cell->net; __be32 *bp; @@ -637,8 +648,10 @@ struct afs_addr_list *afs_yfsvl_get_endpoints(struct afs_vl_cursor *vc, return ERR_PTR(-ENOMEM); call->key = vc->key; - call->reply[0] = NULL; - call->ret_reply0 = true; + call->ret_alist = NULL; + call->max_lifespan = AFS_VL_MAX_LIFESPAN; + call->peer = rxrpc_kernel_get_peer(vc->alist->addrs[vc->addr_index].peer); + call->service_id = vc->server->service_id; /* Marshall the parameters */ bp = call->request; @@ -647,5 +660,134 @@ struct afs_addr_list *afs_yfsvl_get_endpoints(struct afs_vl_cursor *vc, memcpy(bp, uuid, sizeof(*uuid)); /* Type opr_uuid */ trace_afs_make_vl_call(call); - return (struct afs_addr_list *)afs_make_call(&vc->ac, call, GFP_KERNEL, false); + afs_make_call(call, GFP_KERNEL); + afs_wait_for_call_to_complete(call); + vc->call_abort_code = call->abort_code; + vc->call_error = call->error; + vc->call_responded = call->responded; + alist = call->ret_alist; + afs_put_call(call); + if (vc->call_error) { + afs_put_addrlist(alist, afs_alist_trace_put_getaddru); + return ERR_PTR(vc->call_error); + } + return alist; +} + +/* + * Deliver reply data to a YFSVL.GetCellName operation. + */ +static int afs_deliver_yfsvl_get_cell_name(struct afs_call *call) +{ + char *cell_name; + u32 namesz, paddedsz; + int ret; + + _enter("{%u,%zu/%u}", + call->unmarshall, iov_iter_count(call->iter), call->count); + + switch (call->unmarshall) { + case 0: + afs_extract_to_tmp(call); + call->unmarshall++; + + fallthrough; /* and extract the cell name length */ + case 1: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + namesz = ntohl(call->tmp); + if (namesz > YFS_VL_MAXCELLNAME) + return afs_protocol_error(call, afs_eproto_cellname_len); + paddedsz = (namesz + 3) & ~3; + call->count = namesz; + call->count2 = paddedsz - namesz; + + cell_name = kmalloc(namesz + 1, GFP_KERNEL); + if (!cell_name) + return -ENOMEM; + cell_name[namesz] = 0; + call->ret_str = cell_name; + + afs_extract_begin(call, cell_name, namesz); + call->unmarshall++; + + fallthrough; /* and extract cell name */ + case 2: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + afs_extract_discard(call, call->count2); + call->unmarshall++; + + fallthrough; /* and extract padding */ + case 3: + ret = afs_extract_data(call, false); + if (ret < 0) + return ret; + + call->unmarshall++; + break; + } + + _leave(" = 0 [done]"); + return 0; +} + +/* + * VL.GetCapabilities operation type + */ +static const struct afs_call_type afs_YFSVLGetCellName = { + .name = "YFSVL.GetCellName", + .op = afs_YFSVL_GetCellName, + .deliver = afs_deliver_yfsvl_get_cell_name, + .destructor = afs_flat_call_destructor, +}; + +/* + * Probe a volume server for the capabilities that it supports. This can + * return up to 196 words. + * + * We use this to probe for service upgrade to determine what the server at the + * other end supports. + */ +char *afs_yfsvl_get_cell_name(struct afs_vl_cursor *vc) +{ + struct afs_call *call; + struct afs_net *net = vc->cell->net; + __be32 *bp; + char *cellname; + + _enter(""); + + call = afs_alloc_flat_call(net, &afs_YFSVLGetCellName, 1 * 4, 0); + if (!call) + return ERR_PTR(-ENOMEM); + + call->key = vc->key; + call->ret_str = NULL; + call->max_lifespan = AFS_VL_MAX_LIFESPAN; + call->peer = rxrpc_kernel_get_peer(vc->alist->addrs[vc->addr_index].peer); + call->service_id = vc->server->service_id; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(YVLGETCELLNAME); + + /* Can't take a ref on server */ + trace_afs_make_vl_call(call); + afs_make_call(call, GFP_KERNEL); + afs_wait_for_call_to_complete(call); + vc->call_abort_code = call->abort_code; + vc->call_error = call->error; + vc->call_responded = call->responded; + cellname = call->ret_str; + afs_put_call(call); + if (vc->call_error) { + kfree(cellname); + return ERR_PTR(vc->call_error); + } + return cellname; } diff --git a/fs/afs/volume.c b/fs/afs/volume.c index 00975ed3640f..0efff3d25133 100644 --- a/fs/afs/volume.c +++ b/fs/afs/volume.c @@ -1,72 +1,154 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* AFS volume management * * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> #include <linux/slab.h> #include "internal.h" -unsigned __read_mostly afs_volume_gc_delay = 10; -unsigned __read_mostly afs_volume_record_life = 60 * 60; +static unsigned __read_mostly afs_volume_record_life = 60 * 60; +static atomic_t afs_volume_debug_id; + +static void afs_destroy_volume(struct work_struct *work); + +/* + * Insert a volume into a cell. If there's an existing volume record, that is + * returned instead with a ref held. + */ +static struct afs_volume *afs_insert_volume_into_cell(struct afs_cell *cell, + struct afs_volume *volume) +{ + struct afs_volume *p; + struct rb_node *parent = NULL, **pp; + + write_seqlock(&cell->volume_lock); + + pp = &cell->volumes.rb_node; + while (*pp) { + parent = *pp; + p = rb_entry(parent, struct afs_volume, cell_node); + if (p->vid < volume->vid) { + pp = &(*pp)->rb_left; + } else if (p->vid > volume->vid) { + pp = &(*pp)->rb_right; + } else { + if (afs_try_get_volume(p, afs_volume_trace_get_cell_insert)) { + volume = p; + goto found; + } + + set_bit(AFS_VOLUME_RM_TREE, &volume->flags); + rb_replace_node_rcu(&p->cell_node, &volume->cell_node, &cell->volumes); + } + } + + rb_link_node_rcu(&volume->cell_node, parent, pp); + rb_insert_color(&volume->cell_node, &cell->volumes); + hlist_add_head_rcu(&volume->proc_link, &cell->proc_volumes); + +found: + write_sequnlock(&cell->volume_lock); + return volume; -static const char *const afs_voltypes[] = { "R/W", "R/O", "BAK" }; +} + +static void afs_remove_volume_from_cell(struct afs_volume *volume) +{ + struct afs_cell *cell = volume->cell; + + if (!hlist_unhashed(&volume->proc_link)) { + trace_afs_volume(volume->debug_id, volume->vid, refcount_read(&volume->ref), + afs_volume_trace_remove); + write_seqlock(&cell->volume_lock); + hlist_del_rcu(&volume->proc_link); + if (!test_and_set_bit(AFS_VOLUME_RM_TREE, &volume->flags)) + rb_erase(&volume->cell_node, &cell->volumes); + write_sequnlock(&cell->volume_lock); + } +} /* * Allocate a volume record and load it up from a vldb record. */ -static struct afs_volume *afs_alloc_volume(struct afs_mount_params *params, +static struct afs_volume *afs_alloc_volume(struct afs_fs_context *params, struct afs_vldb_entry *vldb, - unsigned long type_mask) + struct afs_server_list **_slist) { struct afs_server_list *slist; struct afs_volume *volume; - int ret = -ENOMEM, nr_servers = 0, i; - - for (i = 0; i < vldb->nr_servers; i++) - if (vldb->fs_mask[i] & type_mask) - nr_servers++; + int ret = -ENOMEM, i; volume = kzalloc(sizeof(struct afs_volume), GFP_KERNEL); if (!volume) goto error_0; + volume->debug_id = atomic_inc_return(&afs_volume_debug_id); volume->vid = vldb->vid[params->type]; volume->update_at = ktime_get_real_seconds() + afs_volume_record_life; - volume->cell = afs_get_cell(params->cell); + volume->cell = afs_get_cell(params->cell, afs_cell_trace_get_vol); volume->type = params->type; volume->type_force = params->force; volume->name_len = vldb->name_len; + volume->creation_time = TIME64_MIN; + volume->update_time = TIME64_MIN; - atomic_set(&volume->usage, 1); - INIT_LIST_HEAD(&volume->proc_link); + refcount_set(&volume->ref, 1); + INIT_HLIST_NODE(&volume->proc_link); + INIT_WORK(&volume->destructor, afs_destroy_volume); rwlock_init(&volume->servers_lock); + mutex_init(&volume->volsync_lock); + mutex_init(&volume->cb_check_lock); + rwlock_init(&volume->cb_v_break_lock); + INIT_LIST_HEAD(&volume->open_mmaps); + init_rwsem(&volume->open_mmaps_lock); memcpy(volume->name, vldb->name, vldb->name_len + 1); - slist = afs_alloc_server_list(params->cell, params->key, vldb, type_mask); + for (i = 0; i < AFS_MAXTYPES; i++) + volume->vids[i] = vldb->vid[i]; + + slist = afs_alloc_server_list(volume, params->key, vldb); if (IS_ERR(slist)) { ret = PTR_ERR(slist); goto error_1; } - refcount_set(&slist->usage, 1); - volume->servers = slist; + *_slist = slist; + rcu_assign_pointer(volume->servers, slist); + trace_afs_volume(volume->debug_id, volume->vid, 1, afs_volume_trace_alloc); return volume; error_1: - afs_put_cell(params->net, volume->cell); + afs_put_cell(volume->cell, afs_cell_trace_put_vol); kfree(volume); error_0: return ERR_PTR(ret); } /* + * Look up or allocate a volume record. + */ +static struct afs_volume *afs_lookup_volume(struct afs_fs_context *params, + struct afs_vldb_entry *vldb) +{ + struct afs_server_list *slist; + struct afs_volume *candidate, *volume; + + candidate = afs_alloc_volume(params, vldb, &slist); + if (IS_ERR(candidate)) + return candidate; + + volume = afs_insert_volume_into_cell(params->cell, candidate); + if (volume == candidate) + afs_attach_volume_to_servers(volume, slist); + else + afs_put_volume(candidate, afs_volume_trace_put_cell_dup); + return volume; +} + +/* * Look up a VLDB record for a volume. */ static struct afs_vldb_entry *afs_vl_lookup_vldb(struct afs_cell *cell, @@ -113,7 +195,7 @@ static struct afs_vldb_entry *afs_vl_lookup_vldb(struct afs_cell *cell, * - Rule 3: If parent volume is R/W, then only mount R/W volume unless * explicitly told otherwise */ -struct afs_volume *afs_create_volume(struct afs_mount_params *params) +struct afs_volume *afs_create_volume(struct afs_fs_context *params) { struct afs_vldb_entry *vldb; struct afs_volume *volume; @@ -142,8 +224,7 @@ struct afs_volume *afs_create_volume(struct afs_mount_params *params) goto error; } - type_mask = 1UL << params->type; - volume = afs_alloc_volume(params, vldb, type_mask); + volume = afs_lookup_volume(params, vldb); error: kfree(vldb); @@ -153,50 +234,103 @@ error: /* * Destroy a volume record */ -static void afs_destroy_volume(struct afs_net *net, struct afs_volume *volume) +static void afs_destroy_volume(struct work_struct *work) { + struct afs_volume *volume = container_of(work, struct afs_volume, destructor); + struct afs_server_list *slist = rcu_access_pointer(volume->servers); + _enter("%p", volume); #ifdef CONFIG_AFS_FSCACHE ASSERTCMP(volume->cache, ==, NULL); #endif - afs_put_serverlist(net, volume->servers); - afs_put_cell(net, volume->cell); - kfree(volume); + afs_detach_volume_from_servers(volume, slist); + afs_remove_volume_from_cell(volume); + afs_put_serverlist(volume->cell->net, slist); + afs_put_cell(volume->cell, afs_cell_trace_put_vol); + trace_afs_volume(volume->debug_id, volume->vid, refcount_read(&volume->ref), + afs_volume_trace_free); + kfree_rcu(volume, rcu); _leave(" [destroyed]"); } /* - * Drop a reference on a volume record. + * Try to get a reference on a volume record. + */ +bool afs_try_get_volume(struct afs_volume *volume, enum afs_volume_trace reason) +{ + int r; + + if (__refcount_inc_not_zero(&volume->ref, &r)) { + trace_afs_volume(volume->debug_id, volume->vid, r + 1, reason); + return true; + } + return false; +} + +/* + * Get a reference on a volume record. */ -void afs_put_volume(struct afs_cell *cell, struct afs_volume *volume) +struct afs_volume *afs_get_volume(struct afs_volume *volume, + enum afs_volume_trace reason) { if (volume) { - _enter("%s", volume->name); + int r; - if (atomic_dec_and_test(&volume->usage)) - afs_destroy_volume(cell->net, volume); + __refcount_inc(&volume->ref, &r); + trace_afs_volume(volume->debug_id, volume->vid, r + 1, reason); + } + return volume; +} + + +/* + * Drop a reference on a volume record. + */ +void afs_put_volume(struct afs_volume *volume, enum afs_volume_trace reason) +{ + if (volume) { + unsigned int debug_id = volume->debug_id; + afs_volid_t vid = volume->vid; + bool zero; + int r; + + zero = __refcount_dec_and_test(&volume->ref, &r); + trace_afs_volume(debug_id, vid, r - 1, reason); + if (zero) + schedule_work(&volume->destructor); } } /* * Activate a volume. */ -void afs_activate_volume(struct afs_volume *volume) +int afs_activate_volume(struct afs_volume *volume) { #ifdef CONFIG_AFS_FSCACHE - volume->cache = fscache_acquire_cookie(volume->cell->cache, - &afs_volume_cache_index_def, - &volume->vid, sizeof(volume->vid), - NULL, 0, - volume, 0, true); + struct fscache_volume *vcookie; + char *name; + + name = kasprintf(GFP_KERNEL, "afs,%s,%llx", + volume->cell->name, volume->vid); + if (!name) + return -ENOMEM; + + vcookie = fscache_acquire_volume(name, NULL, NULL, 0); + if (IS_ERR(vcookie)) { + if (vcookie != ERR_PTR(-EBUSY)) { + kfree(name); + return PTR_ERR(vcookie); + } + pr_err("AFS: Cache volume key already in use (%s)\n", name); + vcookie = NULL; + } + volume->cache = vcookie; + kfree(name); #endif - - write_lock(&volume->cell->proc_lock); - list_add_tail(&volume->proc_link, &volume->cell->proc_volumes); - write_unlock(&volume->cell->proc_lock); + return 0; } /* @@ -206,12 +340,8 @@ void afs_deactivate_volume(struct afs_volume *volume) { _enter("%s", volume->name); - write_lock(&volume->cell->proc_lock); - list_del_init(&volume->proc_link); - write_unlock(&volume->cell->proc_lock); - #ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(volume->cache, NULL, + fscache_relinquish_volume(volume->cache, NULL, test_bit(AFS_VOLUME_DELETED, &volume->flags)); volume->cache = NULL; #endif @@ -226,7 +356,7 @@ static int afs_update_volume_status(struct afs_volume *volume, struct key *key) { struct afs_server_list *new, *old, *discard; struct afs_vldb_entry *vldb; - char idbuf[16]; + char idbuf[24]; int ret, idsz; _enter(""); @@ -234,7 +364,7 @@ static int afs_update_volume_status(struct afs_volume *volume, struct key *key) /* We look up an ID by passing it as a decimal string in the * operation's name parameter. */ - idsz = sprintf(idbuf, "%llu", volume->vid); + idsz = snprintf(idbuf, sizeof(idbuf), "%llu", volume->vid); vldb = afs_vl_lookup_vldb(volume->cell, key, idbuf, idsz); if (IS_ERR(vldb)) { @@ -251,8 +381,7 @@ static int afs_update_volume_status(struct afs_volume *volume, struct key *key) } /* See if the volume's server list got updated. */ - new = afs_alloc_server_list(volume->cell, key, - vldb, (1 << volume->type)); + new = afs_alloc_server_list(volume, key, vldb); if (IS_ERR(new)) { ret = PTR_ERR(new); goto error_vldb; @@ -261,21 +390,27 @@ static int afs_update_volume_status(struct afs_volume *volume, struct key *key) write_lock(&volume->servers_lock); discard = new; - old = volume->servers; + old = rcu_dereference_protected(volume->servers, + lockdep_is_held(&volume->servers_lock)); if (afs_annotate_server_list(new, old)) { new->seq = volume->servers_seq + 1; - volume->servers = new; + rcu_assign_pointer(volume->servers, new); smp_wmb(); volume->servers_seq++; discard = old; } - volume->update_at = ktime_get_real_seconds() + afs_volume_record_life; - clear_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags); + /* Check more often if replication is ongoing. */ + if (new->ro_replicating) + volume->update_at = ktime_get_real_seconds() + 10 * 60; + else + volume->update_at = ktime_get_real_seconds() + afs_volume_record_life; write_unlock(&volume->servers_lock); - ret = 0; + if (discard == old) + afs_reattach_volume_to_servers(volume, new, old); afs_put_serverlist(volume->cell->net, discard); + ret = 0; error_vldb: kfree(vldb); error: @@ -286,25 +421,27 @@ error: /* * Make sure the volume record is up to date. */ -int afs_check_volume_status(struct afs_volume *volume, struct key *key) +int afs_check_volume_status(struct afs_volume *volume, struct afs_operation *op) { - time64_t now = ktime_get_real_seconds(); int ret, retries = 0; _enter(""); - if (volume->update_at <= now) - set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags); - retry: - if (!test_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags) && - !test_bit(AFS_VOLUME_WAIT, &volume->flags)) { - _leave(" = 0"); - return 0; - } - + if (test_bit(AFS_VOLUME_WAIT, &volume->flags)) + goto wait; + if (volume->update_at <= ktime_get_real_seconds() || + test_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags)) + goto update; + _leave(" = 0"); + return 0; + +update: if (!test_and_set_bit_lock(AFS_VOLUME_UPDATING, &volume->flags)) { - ret = afs_update_volume_status(volume, key); + clear_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags); + ret = afs_update_volume_status(volume, op->key); + if (ret < 0) + set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags); clear_bit_unlock(AFS_VOLUME_WAIT, &volume->flags); clear_bit_unlock(AFS_VOLUME_UPDATING, &volume->flags); wake_up_bit(&volume->flags, AFS_VOLUME_WAIT); @@ -312,12 +449,15 @@ retry: return ret; } +wait: if (!test_bit(AFS_VOLUME_WAIT, &volume->flags)) { _leave(" = 0 [no wait]"); return 0; } - ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT, TASK_INTERRUPTIBLE); + ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT, + (op->flags & AFS_OPERATION_UNINTR) ? + TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE); if (ret == -ERESTARTSYS) { _leave(" = %d", ret); return ret; diff --git a/fs/afs/write.c b/fs/afs/write.c index 72efcfcf9f95..93ad86ff3345 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* handling of writes to regular files and writing back to the server * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/backing-dev.h> @@ -15,368 +11,146 @@ #include <linux/pagemap.h> #include <linux/writeback.h> #include <linux/pagevec.h> +#include <linux/netfs.h> +#include <trace/events/netfs.h> #include "internal.h" /* - * mark a page as having been made dirty and thus needing writeback - */ -int afs_set_page_dirty(struct page *page) -{ - _enter(""); - return __set_page_dirty_nobuffers(page); -} - -/* - * partly or wholly fill a page that's under preparation for writing + * completion of write to server */ -static int afs_fill_page(struct afs_vnode *vnode, struct key *key, - loff_t pos, unsigned int len, struct page *page) +static void afs_pages_written_back(struct afs_vnode *vnode, loff_t start, unsigned int len) { - struct afs_read *req; - size_t p; - void *data; - int ret; - - _enter(",,%llu", (unsigned long long)pos); - - if (pos >= vnode->vfs_inode.i_size) { - p = pos & ~PAGE_MASK; - ASSERTCMP(p + len, <=, PAGE_SIZE); - data = kmap(page); - memset(data + p, 0, len); - kunmap(page); - return 0; - } - - req = kzalloc(sizeof(struct afs_read) + sizeof(struct page *), - GFP_KERNEL); - if (!req) - return -ENOMEM; - - refcount_set(&req->usage, 1); - req->pos = pos; - req->len = len; - req->nr_pages = 1; - req->pages = req->array; - req->pages[0] = page; - get_page(page); - - ret = afs_fetch_data(vnode, key, req); - afs_put_read(req); - if (ret < 0) { - if (ret == -ENOENT) { - _debug("got NOENT from server" - " - marking file deleted and stale"); - set_bit(AFS_VNODE_DELETED, &vnode->flags); - ret = -ESTALE; - } - } + _enter("{%llx:%llu},{%x @%llx}", + vnode->fid.vid, vnode->fid.vnode, len, start); - _leave(" = %d", ret); - return ret; + afs_prune_wb_keys(vnode); + _leave(""); } /* - * prepare to perform part of a write to a page + * Find a key to use for the writeback. We cached the keys used to author the + * writes on the vnode. wreq->netfs_priv2 will contain the last writeback key + * record used or NULL and we need to start from there if it's set. + * wreq->netfs_priv will be set to the key itself or NULL. */ -int afs_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata) +static void afs_get_writeback_key(struct netfs_io_request *wreq) { - struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); - struct page *page; - struct key *key = afs_file_key(file); - unsigned long priv; - unsigned f, from = pos & (PAGE_SIZE - 1); - unsigned t, to = from + len; - pgoff_t index = pos >> PAGE_SHIFT; - int ret; + struct afs_wb_key *wbk, *old = wreq->netfs_priv2; + struct afs_vnode *vnode = AFS_FS_I(wreq->inode); - _enter("{%llx:%llu},{%lx},%u,%u", - vnode->fid.vid, vnode->fid.vnode, index, from, to); + key_put(wreq->netfs_priv); + wreq->netfs_priv = NULL; + wreq->netfs_priv2 = NULL; - /* We want to store information about how much of a page is altered in - * page->private. - */ - BUILD_BUG_ON(PAGE_SIZE > 32768 && sizeof(page->private) < 8); - - page = grab_cache_page_write_begin(mapping, index, flags); - if (!page) - return -ENOMEM; - - if (!PageUptodate(page) && len != PAGE_SIZE) { - ret = afs_fill_page(vnode, key, pos & PAGE_MASK, PAGE_SIZE, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - _leave(" = %d [prep]", ret); - return ret; - } - SetPageUptodate(page); - } - - /* page won't leak in error case: it eventually gets cleaned off LRU */ - *pagep = page; - -try_again: - /* See if this page is already partially written in a way that we can - * merge the new write with. - */ - t = f = 0; - if (PagePrivate(page)) { - priv = page_private(page); - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; - ASSERTCMP(f, <=, t); - } + spin_lock(&vnode->wb_lock); + if (old) + wbk = list_next_entry(old, vnode_link); + else + wbk = list_first_entry(&vnode->wb_keys, struct afs_wb_key, vnode_link); - if (f != t) { - if (PageWriteback(page)) { - trace_afs_page_dirty(vnode, tracepoint_string("alrdy"), - page->index, priv); - goto flush_conflicting_write; + list_for_each_entry_from(wbk, &vnode->wb_keys, vnode_link) { + _debug("wbk %u", key_serial(wbk->key)); + if (key_validate(wbk->key) == 0) { + refcount_inc(&wbk->usage); + wreq->netfs_priv = key_get(wbk->key); + wreq->netfs_priv2 = wbk; + _debug("USE WB KEY %u", key_serial(wbk->key)); + break; } - /* If the file is being filled locally, allow inter-write - * spaces to be merged into writes. If it's not, only write - * back what the user gives us. - */ - if (!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags) && - (to < f || from > t)) - goto flush_conflicting_write; - if (from < f) - f = from; - if (to > t) - t = to; - } else { - f = from; - t = to; } - priv = (unsigned long)t << AFS_PRIV_SHIFT; - priv |= f; - trace_afs_page_dirty(vnode, tracepoint_string("begin"), - page->index, priv); - SetPagePrivate(page); - set_page_private(page, priv); - _leave(" = 0"); - return 0; - - /* The previous write and this write aren't adjacent or overlapping, so - * flush the page out. - */ -flush_conflicting_write: - _debug("flush conflict"); - ret = write_one_page(page); - if (ret < 0) { - _leave(" = %d", ret); - return ret; - } + spin_unlock(&vnode->wb_lock); - ret = lock_page_killable(page); - if (ret < 0) { - _leave(" = %d", ret); - return ret; - } - goto try_again; + afs_put_wb_key(old); } -/* - * finalise part of a write to a page - */ -int afs_write_end(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) +static void afs_store_data_success(struct afs_operation *op) { - struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); - struct key *key = afs_file_key(file); - loff_t i_size, maybe_i_size; - int ret; - - _enter("{%llx:%llu},{%lx}", - vnode->fid.vid, vnode->fid.vnode, page->index); - - maybe_i_size = pos + copied; - - i_size = i_size_read(&vnode->vfs_inode); - if (maybe_i_size > i_size) { - spin_lock(&vnode->wb_lock); - i_size = i_size_read(&vnode->vfs_inode); - if (maybe_i_size > i_size) - i_size_write(&vnode->vfs_inode, maybe_i_size); - spin_unlock(&vnode->wb_lock); - } + struct afs_vnode *vnode = op->file[0].vnode; - if (!PageUptodate(page)) { - if (copied < len) { - /* Try and load any missing data from the server. The - * unmarshalling routine will take care of clearing any - * bits that are beyond the EOF. - */ - ret = afs_fill_page(vnode, key, pos + copied, - len - copied, page); - if (ret < 0) - goto out; - } - SetPageUptodate(page); + op->ctime = op->file[0].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[0]); + if (!afs_op_error(op)) { + afs_pages_written_back(vnode, op->store.pos, op->store.size); + afs_stat_v(vnode, n_stores); + atomic_long_add(op->store.size, &afs_v2net(vnode)->n_store_bytes); } - - set_page_dirty(page); - if (PageDirty(page)) - _debug("dirtied"); - ret = copied; - -out: - unlock_page(page); - put_page(page); - return ret; } -/* - * kill all the pages in the given range - */ -static void afs_kill_pages(struct address_space *mapping, - pgoff_t first, pgoff_t last) -{ - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - struct pagevec pv; - unsigned count, loop; - - _enter("{%llx:%llu},%lx-%lx", - vnode->fid.vid, vnode->fid.vnode, first, last); - - pagevec_init(&pv); - - do { - _debug("kill %lx-%lx", first, last); - - count = last - first + 1; - if (count > PAGEVEC_SIZE) - count = PAGEVEC_SIZE; - pv.nr = find_get_pages_contig(mapping, first, count, pv.pages); - ASSERTCMP(pv.nr, ==, count); - - for (loop = 0; loop < count; loop++) { - struct page *page = pv.pages[loop]; - ClearPageUptodate(page); - SetPageError(page); - end_page_writeback(page); - if (page->index >= first) - first = page->index + 1; - lock_page(page); - generic_error_remove_page(mapping, page); - } - - __pagevec_release(&pv); - } while (first <= last); - - _leave(""); -} +static const struct afs_operation_ops afs_store_data_operation = { + .issue_afs_rpc = afs_fs_store_data, + .issue_yfs_rpc = yfs_fs_store_data, + .success = afs_store_data_success, +}; /* - * Redirty all the pages in a given range. + * Prepare a subrequest to write to the server. This sets the max_len + * parameter. */ -static void afs_redirty_pages(struct writeback_control *wbc, - struct address_space *mapping, - pgoff_t first, pgoff_t last) +void afs_prepare_write(struct netfs_io_subrequest *subreq) { - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - struct pagevec pv; - unsigned count, loop; - - _enter("{%llx:%llu},%lx-%lx", - vnode->fid.vid, vnode->fid.vnode, first, last); - - pagevec_init(&pv); - - do { - _debug("redirty %lx-%lx", first, last); - - count = last - first + 1; - if (count > PAGEVEC_SIZE) - count = PAGEVEC_SIZE; - pv.nr = find_get_pages_contig(mapping, first, count, pv.pages); - ASSERTCMP(pv.nr, ==, count); + struct netfs_io_stream *stream = &subreq->rreq->io_streams[subreq->stream_nr]; - for (loop = 0; loop < count; loop++) { - struct page *page = pv.pages[loop]; - - redirty_page_for_writepage(wbc, page); - end_page_writeback(page); - if (page->index >= first) - first = page->index + 1; - } - - __pagevec_release(&pv); - } while (first <= last); - - _leave(""); + //if (test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) + // subreq->max_len = 512 * 1024; + //else + stream->sreq_max_len = 256 * 1024 * 1024; } /* - * write to a file + * Issue a subrequest to write to the server. */ -static int afs_store_data(struct address_space *mapping, - pgoff_t first, pgoff_t last, - unsigned offset, unsigned to) +static void afs_issue_write_worker(struct work_struct *work) { - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - struct afs_fs_cursor fc; - struct afs_wb_key *wbk = NULL; - struct list_head *p; - int ret = -ENOKEY, ret2; - - _enter("%s{%llx:%llu.%u},%lx,%lx,%x,%x", + struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work); + struct netfs_io_request *wreq = subreq->rreq; + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(wreq->inode); + unsigned long long pos = subreq->start + subreq->transferred; + size_t len = subreq->len - subreq->transferred; + int ret = -ENOKEY; + + _enter("R=%x[%x],%s{%llx:%llu.%u},%llx,%zx", + wreq->debug_id, subreq->debug_index, vnode->volume->name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, - first, last, offset, to); + pos, len); - spin_lock(&vnode->wb_lock); - p = vnode->wb_keys.next; +#if 0 // Error injection + if (subreq->debug_index == 3) + return netfs_write_subrequest_terminated(subreq, -ENOANO); - /* Iterate through the list looking for a valid key to use. */ -try_next_key: - while (p != &vnode->wb_keys) { - wbk = list_entry(p, struct afs_wb_key, vnode_link); - _debug("wbk %u", key_serial(wbk->key)); - ret2 = key_validate(wbk->key); - if (ret2 == 0) - goto found_key; - if (ret == -ENOKEY) - ret = ret2; - p = p->next; + if (!subreq->retry_count) { + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + return netfs_write_subrequest_terminated(subreq, -EAGAIN); } +#endif - spin_unlock(&vnode->wb_lock); - afs_put_wb_key(wbk); - _leave(" = %d [no keys]", ret); - return ret; - -found_key: - refcount_inc(&wbk->usage); - spin_unlock(&vnode->wb_lock); + op = afs_alloc_operation(wreq->netfs_priv, vnode->volume); + if (IS_ERR(op)) + return netfs_write_subrequest_terminated(subreq, -EAGAIN); - _debug("USE WB KEY %u", key_serial(wbk->key)); + afs_op_set_vnode(op, 0, vnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->store.pos = pos; + op->store.size = len; + op->flags |= AFS_OPERATION_UNINTR; + op->ops = &afs_store_data_operation; - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, vnode, wbk->key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(vnode); - afs_fs_store_data(&fc, mapping, first, last, offset, to); - } + afs_begin_vnode_operation(op); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, vnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - } + op->store.write_iter = &subreq->io_iter; + op->store.i_size = umax(pos + len, vnode->netfs.remote_i_size); + op->mtime = inode_get_mtime(&vnode->netfs.inode); + afs_wait_for_operation(op); + ret = afs_put_operation(op); switch (ret) { case 0: - afs_stat_v(vnode, n_stores); - atomic_long_add((last * PAGE_SIZE + to) - - (first * PAGE_SIZE + offset), - &afs_v2net(vnode)->n_store_bytes); + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); break; case -EACCES: case -EPERM: @@ -384,364 +158,91 @@ found_key: case -EKEYEXPIRED: case -EKEYREJECTED: case -EKEYREVOKED: - _debug("next"); - spin_lock(&vnode->wb_lock); - p = wbk->vnode_link.next; - afs_put_wb_key(wbk); - goto try_next_key; + /* If there are more keys we can try, use the retry algorithm + * to rotate the keys. + */ + if (wreq->netfs_priv2) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + break; } - afs_put_wb_key(wbk); - _leave(" = %d", ret); - return ret; + netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len); +} + +void afs_issue_write(struct netfs_io_subrequest *subreq) +{ + subreq->work.func = afs_issue_write_worker; + if (!queue_work(system_dfl_wq, &subreq->work)) + WARN_ON_ONCE(1); } /* - * Synchronously write back the locked page and any subsequent non-locked dirty - * pages. + * Writeback calls this when it finds a folio that needs uploading. This isn't + * called if writeback only has copy-to-cache to deal with. */ -static int afs_write_back_from_locked_page(struct address_space *mapping, - struct writeback_control *wbc, - struct page *primary_page, - pgoff_t final_page) +void afs_begin_writeback(struct netfs_io_request *wreq) { - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - struct page *pages[8], *page; - unsigned long count, priv; - unsigned n, offset, to, f, t; - pgoff_t start, first, last; - int loop, ret; - - _enter(",%lx", primary_page->index); - - count = 1; - if (test_set_page_writeback(primary_page)) - BUG(); - - /* Find all consecutive lockable dirty pages that have contiguous - * written regions, stopping when we find a page that is not - * immediately lockable, is not dirty or is missing, or we reach the - * end of the range. - */ - start = primary_page->index; - priv = page_private(primary_page); - offset = priv & AFS_PRIV_MAX; - to = priv >> AFS_PRIV_SHIFT; - trace_afs_page_dirty(vnode, tracepoint_string("store"), - primary_page->index, priv); - - WARN_ON(offset == to); - if (offset == to) - trace_afs_page_dirty(vnode, tracepoint_string("WARN"), - primary_page->index, priv); - - if (start >= final_page || - (to < PAGE_SIZE && !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags))) - goto no_more; - - start++; - do { - _debug("more %lx [%lx]", start, count); - n = final_page - start + 1; - if (n > ARRAY_SIZE(pages)) - n = ARRAY_SIZE(pages); - n = find_get_pages_contig(mapping, start, ARRAY_SIZE(pages), pages); - _debug("fgpc %u", n); - if (n == 0) - goto no_more; - if (pages[0]->index != start) { - do { - put_page(pages[--n]); - } while (n > 0); - goto no_more; - } - - for (loop = 0; loop < n; loop++) { - page = pages[loop]; - if (to != PAGE_SIZE && - !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) - break; - if (page->index > final_page) - break; - if (!trylock_page(page)) - break; - if (!PageDirty(page) || PageWriteback(page)) { - unlock_page(page); - break; - } - - priv = page_private(page); - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; - if (f != 0 && - !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) { - unlock_page(page); - break; - } - to = t; - - trace_afs_page_dirty(vnode, tracepoint_string("store+"), - page->index, priv); - - if (!clear_page_dirty_for_io(page)) - BUG(); - if (test_set_page_writeback(page)) - BUG(); - unlock_page(page); - put_page(page); - } - count += loop; - if (loop < n) { - for (; loop < n; loop++) - put_page(pages[loop]); - goto no_more; - } - - start += loop; - } while (start <= final_page && count < 65536); - -no_more: - /* We now have a contiguous set of dirty pages, each with writeback - * set; the first page is still locked at this point, but all the rest - * have been unlocked. - */ - unlock_page(primary_page); - - first = primary_page->index; - last = first + count - 1; - - _debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to); + if (S_ISREG(wreq->inode->i_mode)) + afs_get_writeback_key(wreq); +} - ret = afs_store_data(mapping, first, last, offset, to); - switch (ret) { - case 0: - ret = count; +/* + * Prepare to retry the writes in request. Use this to try rotating the + * available writeback keys. + */ +void afs_retry_request(struct netfs_io_request *wreq, struct netfs_io_stream *stream) +{ + struct netfs_io_subrequest *subreq = + list_first_entry(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + + switch (wreq->origin) { + case NETFS_READAHEAD: + case NETFS_READPAGE: + case NETFS_READ_GAPS: + case NETFS_READ_SINGLE: + case NETFS_READ_FOR_WRITE: + case NETFS_UNBUFFERED_READ: + case NETFS_DIO_READ: + return; + default: break; + } - default: - pr_notice("kAFS: Unexpected error from FS.StoreData %d\n", ret); - /* Fall through */ + switch (subreq->error) { case -EACCES: case -EPERM: case -ENOKEY: case -EKEYEXPIRED: case -EKEYREJECTED: case -EKEYREVOKED: - afs_redirty_pages(wbc, mapping, first, last); - mapping_set_error(mapping, ret); - break; - - case -EDQUOT: - case -ENOSPC: - afs_redirty_pages(wbc, mapping, first, last); - mapping_set_error(mapping, -ENOSPC); - break; - - case -EROFS: - case -EIO: - case -EREMOTEIO: - case -EFBIG: - case -ENOENT: - case -ENOMEDIUM: - case -ENXIO: - trace_afs_file_error(vnode, ret, afs_file_error_writeback_fail); - afs_kill_pages(mapping, first, last); - mapping_set_error(mapping, ret); + afs_get_writeback_key(wreq); + if (!wreq->netfs_priv) + stream->failed = true; break; } - - _leave(" = %d", ret); - return ret; -} - -/* - * write a page back to the server - * - the caller locked the page for us - */ -int afs_writepage(struct page *page, struct writeback_control *wbc) -{ - int ret; - - _enter("{%lx},", page->index); - - ret = afs_write_back_from_locked_page(page->mapping, wbc, page, - wbc->range_end >> PAGE_SHIFT); - if (ret < 0) { - _leave(" = %d", ret); - return 0; - } - - wbc->nr_to_write -= ret; - - _leave(" = 0"); - return 0; -} - -/* - * write a region of pages back to the server - */ -static int afs_writepages_region(struct address_space *mapping, - struct writeback_control *wbc, - pgoff_t index, pgoff_t end, pgoff_t *_next) -{ - struct page *page; - int ret, n; - - _enter(",,%lx,%lx,", index, end); - - do { - n = find_get_pages_range_tag(mapping, &index, end, - PAGECACHE_TAG_DIRTY, 1, &page); - if (!n) - break; - - _debug("wback %lx", page->index); - - /* - * at this point we hold neither the i_pages lock nor the - * page lock: the page may be truncated or invalidated - * (changing page->mapping to NULL), or even swizzled - * back from swapper_space to tmpfs file mapping - */ - ret = lock_page_killable(page); - if (ret < 0) { - put_page(page); - _leave(" = %d", ret); - return ret; - } - - if (page->mapping != mapping || !PageDirty(page)) { - unlock_page(page); - put_page(page); - continue; - } - - if (PageWriteback(page)) { - unlock_page(page); - if (wbc->sync_mode != WB_SYNC_NONE) - wait_on_page_writeback(page); - put_page(page); - continue; - } - - if (!clear_page_dirty_for_io(page)) - BUG(); - ret = afs_write_back_from_locked_page(mapping, wbc, page, end); - put_page(page); - if (ret < 0) { - _leave(" = %d", ret); - return ret; - } - - wbc->nr_to_write -= ret; - - cond_resched(); - } while (index < end && wbc->nr_to_write > 0); - - *_next = index; - _leave(" = 0 [%lx]", *_next); - return 0; } /* * write some of the pending data back to the server */ -int afs_writepages(struct address_space *mapping, - struct writeback_control *wbc) +int afs_writepages(struct address_space *mapping, struct writeback_control *wbc) { - pgoff_t start, end, next; + struct afs_vnode *vnode = AFS_FS_I(mapping->host); int ret; - _enter(""); - - if (wbc->range_cyclic) { - start = mapping->writeback_index; - end = -1; - ret = afs_writepages_region(mapping, wbc, start, end, &next); - if (start > 0 && wbc->nr_to_write > 0 && ret == 0) - ret = afs_writepages_region(mapping, wbc, 0, start, - &next); - mapping->writeback_index = next; - } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { - end = (pgoff_t)(LLONG_MAX >> PAGE_SHIFT); - ret = afs_writepages_region(mapping, wbc, 0, end, &next); - if (wbc->nr_to_write > 0) - mapping->writeback_index = next; - } else { - start = wbc->range_start >> PAGE_SHIFT; - end = wbc->range_end >> PAGE_SHIFT; - ret = afs_writepages_region(mapping, wbc, start, end, &next); - } - - _leave(" = %d", ret); - return ret; -} - -/* - * completion of write to server - */ -void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call) -{ - struct pagevec pv; - unsigned long priv; - unsigned count, loop; - pgoff_t first = call->first, last = call->last; - - _enter("{%llx:%llu},{%lx-%lx}", - vnode->fid.vid, vnode->fid.vnode, first, last); - - pagevec_init(&pv); - - do { - _debug("done %lx-%lx", first, last); - - count = last - first + 1; - if (count > PAGEVEC_SIZE) - count = PAGEVEC_SIZE; - pv.nr = find_get_pages_contig(vnode->vfs_inode.i_mapping, - first, count, pv.pages); - ASSERTCMP(pv.nr, ==, count); - - for (loop = 0; loop < count; loop++) { - priv = page_private(pv.pages[loop]); - trace_afs_page_dirty(vnode, tracepoint_string("clear"), - pv.pages[loop]->index, priv); - set_page_private(pv.pages[loop], 0); - end_page_writeback(pv.pages[loop]); - } - first += count; - __pagevec_release(&pv); - } while (first <= last); - - afs_prune_wb_keys(vnode); - _leave(""); -} - -/* - * write to an AFS file - */ -ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from) -{ - struct afs_vnode *vnode = AFS_FS_I(file_inode(iocb->ki_filp)); - ssize_t result; - size_t count = iov_iter_count(from); - - _enter("{%llx:%llu},{%zu},", - vnode->fid.vid, vnode->fid.vnode, count); - - if (IS_SWAPFILE(&vnode->vfs_inode)) { - printk(KERN_INFO - "AFS: Attempt to write to active swap file!\n"); - return -EBUSY; - } - - if (!count) + /* We have to be careful as we can end up racing with setattr() + * truncating the pagecache since the caller doesn't take a lock here + * to prevent it. + */ + if (wbc->sync_mode == WB_SYNC_ALL) + down_read(&vnode->validate_lock); + else if (!down_read_trylock(&vnode->validate_lock)) return 0; - result = generic_file_write_iter(iocb, from); - - _leave(" = %zd", result); - return result; + ret = netfs_writepages(mapping, wbc); + up_read(&vnode->validate_lock); + return ret; } /* @@ -751,13 +252,18 @@ ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from) */ int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync) { - struct inode *inode = file_inode(file); - struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + struct afs_file *af = file->private_data; + int ret; _enter("{%llx:%llu},{n=%pD},%d", vnode->fid.vid, vnode->fid.vnode, file, datasync); + ret = afs_validate(vnode, af->key); + if (ret < 0) + return ret; + return file_write_and_wait_range(file, start, end); } @@ -768,44 +274,10 @@ int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync) vm_fault_t afs_page_mkwrite(struct vm_fault *vmf) { struct file *file = vmf->vma->vm_file; - struct inode *inode = file_inode(file); - struct afs_vnode *vnode = AFS_FS_I(inode); - unsigned long priv; - - _enter("{{%llx:%llu}},{%lx}", - vnode->fid.vid, vnode->fid.vnode, vmf->page->index); - - sb_start_pagefault(inode->i_sb); - - /* Wait for the page to be written to the cache before we allow it to - * be modified. We then assume the entire page will need writing back. - */ -#ifdef CONFIG_AFS_FSCACHE - fscache_wait_on_page_write(vnode->cache, vmf->page); -#endif - - if (PageWriteback(vmf->page) && - wait_on_page_bit_killable(vmf->page, PG_writeback) < 0) - return VM_FAULT_RETRY; - if (lock_page_killable(vmf->page) < 0) - return VM_FAULT_RETRY; - - /* We mustn't change page->private until writeback is complete as that - * details the portion of the page we need to write back and we might - * need to redirty the page if there's a problem. - */ - wait_on_page_writeback(vmf->page); - - priv = (unsigned long)PAGE_SIZE << AFS_PRIV_SHIFT; /* To */ - priv |= 0; /* From */ - trace_afs_page_dirty(vnode, tracepoint_string("mkwrite"), - vmf->page->index, priv); - SetPagePrivate(vmf->page); - set_page_private(vmf->page, priv); - - sb_end_pagefault(inode->i_sb); - return VM_FAULT_LOCKED; + if (afs_validate(AFS_FS_I(file_inode(file)), afs_file_key(file)) < 0) + return VM_FAULT_SIGBUS; + return netfs_page_mkwrite(vmf, NULL); } /* @@ -819,8 +291,8 @@ void afs_prune_wb_keys(struct afs_vnode *vnode) /* Discard unused keys */ spin_lock(&vnode->wb_lock); - if (!mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_WRITEBACK) && - !mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_DIRTY)) { + if (!mapping_tagged(&vnode->netfs.inode.i_data, PAGECACHE_TAG_WRITEBACK) && + !mapping_tagged(&vnode->netfs.inode.i_data, PAGECACHE_TAG_DIRTY)) { list_for_each_entry_safe(wbk, tmp, &vnode->wb_keys, vnode_link) { if (refcount_read(&wbk->usage) == 1) list_move(&wbk->vnode_link, &graveyard); @@ -835,44 +307,3 @@ void afs_prune_wb_keys(struct afs_vnode *vnode) afs_put_wb_key(wbk); } } - -/* - * Clean up a page during invalidation. - */ -int afs_launder_page(struct page *page) -{ - struct address_space *mapping = page->mapping; - struct afs_vnode *vnode = AFS_FS_I(mapping->host); - unsigned long priv; - unsigned int f, t; - int ret = 0; - - _enter("{%lx}", page->index); - - priv = page_private(page); - if (clear_page_dirty_for_io(page)) { - f = 0; - t = PAGE_SIZE; - if (PagePrivate(page)) { - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; - } - - trace_afs_page_dirty(vnode, tracepoint_string("launder"), - page->index, priv); - ret = afs_store_data(mapping, page->index, page->index, t, f); - } - - trace_afs_page_dirty(vnode, tracepoint_string("laundered"), - page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); - -#ifdef CONFIG_AFS_FSCACHE - if (PageFsCache(page)) { - fscache_wait_on_page_write(vnode->cache, page); - fscache_uncache_page(vnode->cache, page); - } -#endif - return ret; -} diff --git a/fs/afs/xattr.c b/fs/afs/xattr.c index a2cdf25573e2..e19f396aa370 100644 --- a/fs/afs/xattr.c +++ b/fs/afs/xattr.c @@ -1,13 +1,9 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Extended attribute handling for AFS. We use xattrs to get and set metadata * instead of providing pioctl(). * * Copyright (C) 2017 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/slab.h> @@ -15,24 +11,257 @@ #include <linux/xattr.h> #include "internal.h" -static const char afs_xattr_list[] = - "afs.cell\0" - "afs.fid\0" - "afs.volume"; +/* + * Deal with the result of a successful fetch ACL operation. + */ +static void afs_acl_success(struct afs_operation *op) +{ + afs_vnode_commit_status(op, &op->file[0]); +} + +static void afs_acl_put(struct afs_operation *op) +{ + kfree(op->acl); +} + +static const struct afs_operation_ops afs_fetch_acl_operation = { + .issue_afs_rpc = afs_fs_fetch_acl, + .success = afs_acl_success, + .put = afs_acl_put, +}; /* - * Retrieve a list of the supported xattrs. + * Get a file's ACL. */ -ssize_t afs_listxattr(struct dentry *dentry, char *buffer, size_t size) +static int afs_xattr_get_acl(const struct xattr_handler *handler, + struct dentry *dentry, + struct inode *inode, const char *name, + void *buffer, size_t size) { - if (size == 0) - return sizeof(afs_xattr_list); - if (size < sizeof(afs_xattr_list)) - return -ERANGE; - memcpy(buffer, afs_xattr_list, sizeof(afs_xattr_list)); - return sizeof(afs_xattr_list); + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_acl *acl = NULL; + int ret; + + op = afs_alloc_operation(NULL, vnode->volume); + if (IS_ERR(op)) + return -ENOMEM; + + afs_op_set_vnode(op, 0, vnode); + op->ops = &afs_fetch_acl_operation; + + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + acl = op->acl; + op->acl = NULL; + ret = afs_put_operation(op); + + if (ret == 0) { + ret = acl->size; + if (size > 0) { + if (acl->size <= size) + memcpy(buffer, acl->data, acl->size); + else + ret = -ERANGE; + } + } + + kfree(acl); + return ret; +} + +static bool afs_make_acl(struct afs_operation *op, + const void *buffer, size_t size) +{ + struct afs_acl *acl; + + acl = kmalloc(struct_size(acl, data, size), GFP_KERNEL); + if (!acl) { + afs_op_nomem(op); + return false; + } + + acl->size = size; + memcpy(acl->data, buffer, size); + op->acl = acl; + return true; } +static const struct afs_operation_ops afs_store_acl_operation = { + .issue_afs_rpc = afs_fs_store_acl, + .success = afs_acl_success, + .put = afs_acl_put, +}; + +/* + * Set a file's AFS3 ACL. + */ +static int afs_xattr_set_acl(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *dentry, + struct inode *inode, const char *name, + const void *buffer, size_t size, int flags) +{ + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(inode); + + if (flags == XATTR_CREATE) + return -EINVAL; + + op = afs_alloc_operation(NULL, vnode->volume); + if (IS_ERR(op)) + return -ENOMEM; + + afs_op_set_vnode(op, 0, vnode); + if (!afs_make_acl(op, buffer, size)) + return afs_put_operation(op); + + op->ops = &afs_store_acl_operation; + return afs_do_sync_operation(op); +} + +static const struct xattr_handler afs_xattr_afs_acl_handler = { + .name = "afs.acl", + .get = afs_xattr_get_acl, + .set = afs_xattr_set_acl, +}; + +static const struct afs_operation_ops yfs_fetch_opaque_acl_operation = { + .issue_yfs_rpc = yfs_fs_fetch_opaque_acl, + .success = afs_acl_success, + /* Don't free op->yacl in .put here */ +}; + +/* + * Get a file's YFS ACL. + */ +static int afs_xattr_get_yfs(const struct xattr_handler *handler, + struct dentry *dentry, + struct inode *inode, const char *name, + void *buffer, size_t size) +{ + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(inode); + struct yfs_acl *yacl = NULL; + char buf[16], *data; + int which = 0, dsize, ret = -ENOMEM; + + if (strcmp(name, "acl") == 0) + which = 0; + else if (strcmp(name, "acl_inherited") == 0) + which = 1; + else if (strcmp(name, "acl_num_cleaned") == 0) + which = 2; + else if (strcmp(name, "vol_acl") == 0) + which = 3; + else + return -EOPNOTSUPP; + + yacl = kzalloc(sizeof(struct yfs_acl), GFP_KERNEL); + if (!yacl) + goto error; + + if (which == 0) + yacl->flags |= YFS_ACL_WANT_ACL; + else if (which == 3) + yacl->flags |= YFS_ACL_WANT_VOL_ACL; + + op = afs_alloc_operation(NULL, vnode->volume); + if (IS_ERR(op)) + goto error_yacl; + + afs_op_set_vnode(op, 0, vnode); + op->yacl = yacl; + op->ops = &yfs_fetch_opaque_acl_operation; + + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); + ret = afs_put_operation(op); + + if (ret == 0) { + switch (which) { + case 0: + data = yacl->acl->data; + dsize = yacl->acl->size; + break; + case 1: + data = buf; + dsize = scnprintf(buf, sizeof(buf), "%u", yacl->inherit_flag); + break; + case 2: + data = buf; + dsize = scnprintf(buf, sizeof(buf), "%u", yacl->num_cleaned); + break; + case 3: + data = yacl->vol_acl->data; + dsize = yacl->vol_acl->size; + break; + default: + ret = -EOPNOTSUPP; + goto error_yacl; + } + + ret = dsize; + if (size > 0) { + if (dsize <= size) + memcpy(buffer, data, dsize); + else + ret = -ERANGE; + } + } else if (ret == -ENOTSUPP) { + ret = -ENODATA; + } + +error_yacl: + yfs_free_opaque_acl(yacl); +error: + return ret; +} + +static const struct afs_operation_ops yfs_store_opaque_acl2_operation = { + .issue_yfs_rpc = yfs_fs_store_opaque_acl2, + .success = afs_acl_success, + .put = afs_acl_put, +}; + +/* + * Set a file's YFS ACL. + */ +static int afs_xattr_set_yfs(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *dentry, + struct inode *inode, const char *name, + const void *buffer, size_t size, int flags) +{ + struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(inode); + int ret; + + if (flags == XATTR_CREATE || + strcmp(name, "acl") != 0) + return -EINVAL; + + op = afs_alloc_operation(NULL, vnode->volume); + if (IS_ERR(op)) + return -ENOMEM; + + afs_op_set_vnode(op, 0, vnode); + if (!afs_make_acl(op, buffer, size)) + return afs_put_operation(op); + + op->ops = &yfs_store_opaque_acl2_operation; + ret = afs_do_sync_operation(op); + if (ret == -ENOTSUPP) + ret = -ENODATA; + return ret; +} + +static const struct xattr_handler afs_xattr_yfs_handler = { + .prefix = "afs.yfs.", + .get = afs_xattr_get_yfs, + .set = afs_xattr_set_yfs, +}; + /* * Get the name of the cell on which a file resides. */ @@ -50,7 +279,7 @@ static int afs_xattr_get_cell(const struct xattr_handler *handler, return namelen; if (namelen > size) return -ERANGE; - memcpy(buffer, cell->name, size); + memcpy(buffer, cell->name, namelen); return namelen; } @@ -69,11 +298,22 @@ static int afs_xattr_get_fid(const struct xattr_handler *handler, void *buffer, size_t size) { struct afs_vnode *vnode = AFS_FS_I(inode); - char text[8 + 1 + 8 + 1 + 8 + 1]; + char text[16 + 1 + 24 + 1 + 8 + 1]; size_t len; - len = sprintf(text, "%llx:%llx:%x", - vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); + /* The volume ID is 64-bit, the vnode ID is 96-bit and the + * uniquifier is 32-bit. + */ + len = scnprintf(text, sizeof(text), "%llx:", vnode->fid.vid); + if (vnode->fid.vnode_hi) + len += scnprintf(text + len, sizeof(text) - len, "%x%016llx", + vnode->fid.vnode_hi, vnode->fid.vnode); + else + len += scnprintf(text + len, sizeof(text) - len, "%llx", + vnode->fid.vnode); + len += scnprintf(text + len, sizeof(text) - len, ":%x", + vnode->fid.unique); + if (size == 0) return len; if (len > size) @@ -104,7 +344,7 @@ static int afs_xattr_get_volume(const struct xattr_handler *handler, return namelen; if (namelen > size) return -ERANGE; - memcpy(buffer, volname, size); + memcpy(buffer, volname, namelen); return namelen; } @@ -113,9 +353,11 @@ static const struct xattr_handler afs_xattr_afs_volume_handler = { .get = afs_xattr_get_volume, }; -const struct xattr_handler *afs_xattr_handlers[] = { +const struct xattr_handler * const afs_xattr_handlers[] = { + &afs_xattr_afs_acl_handler, &afs_xattr_afs_cell_handler, &afs_xattr_afs_fid_handler, &afs_xattr_afs_volume_handler, + &afs_xattr_yfs_handler, /* afs.yfs. prefix */ NULL }; diff --git a/fs/afs/xdr_fs.h b/fs/afs/xdr_fs.h index aa21f3068d52..cc5f143d21a3 100644 --- a/fs/afs/xdr_fs.h +++ b/fs/afs/xdr_fs.h @@ -1,12 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ /* AFS fileserver XDR types * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #ifndef XDR_FS_H @@ -58,10 +54,16 @@ union afs_xdr_dirent { __be16 hash_next; __be32 vnode; __be32 unique; - u8 name[16]; - u8 overflow[4]; /* if any char of the name (inc - * NUL) reaches here, consume - * the next dirent too */ + u8 name[]; + /* When determining the number of dirent slots needed to + * represent a directory entry, name should be assumed to be 16 + * bytes, due to a now-standardised (mis)calculation, but it is + * in fact 20 bytes in size. afs_dir_calc_slots() should be + * used for this. + * + * For names longer than (16 or) 20 bytes, extra slots should + * be annexed to this one using the extended_name format. + */ } u; u8 extended_name[32]; } __packed; @@ -86,7 +88,7 @@ union afs_xdr_dir_block { struct { struct afs_xdr_dir_hdr hdr; - u8 alloc_ctrs[AFS_DIR_MAX_BLOCKS]; + u8 alloc_ctrs[AFS_DIR_BLOCKS_WITH_CTR]; __be16 hashtable[AFS_DIR_HASHTBL_SIZE]; } meta; @@ -100,4 +102,15 @@ struct afs_xdr_dir_page { union afs_xdr_dir_block blocks[AFS_DIR_BLOCKS_PER_PAGE]; }; +/* + * Calculate the number of dirent slots required for any given name length. + * The calculation is made assuming the part of the name in the first slot is + * 16 bytes, rather than 20, but this miscalculation is now standardised. + */ +static inline unsigned int afs_dir_calc_slots(size_t name_len) +{ + name_len++; /* NUL-terminated */ + return 1 + ((name_len + 15) / AFS_DIR_DIRENT_SIZE); +} + #endif /* XDR_FS_H */ diff --git a/fs/afs/yfsclient.c b/fs/afs/yfsclient.c index 5aa57929e8c2..febf13a49f0b 100644 --- a/fs/afs/yfsclient.c +++ b/fs/afs/yfsclient.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* YFS File Server client stubs * * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. */ #include <linux/init.h> @@ -19,13 +15,6 @@ #include "xdr_fs.h" #include "protocol_yfs.h" -static const struct afs_fid afs_zero_fid; - -static inline void afs_use_fs_server(struct afs_call *call, struct afs_cb_interest *cbi) -{ - call->cbi = afs_get_cb_interest(cbi); -} - #define xdr_size(x) (sizeof(*x) / sizeof(__be32)) static void xdr_decode_YFSFid(const __be32 **_bp, struct afs_fid *fid) @@ -83,31 +72,29 @@ static __be32 *xdr_encode_string(__be32 *bp, const char *p, unsigned int len) return bp + len / sizeof(__be32); } -static s64 linux_to_yfs_time(const struct timespec64 *t) +static __be32 *xdr_encode_name(__be32 *bp, const struct qstr *p) { - /* Convert to 100ns intervals. */ - return (u64)t->tv_sec * 10000000 + t->tv_nsec/100; + return xdr_encode_string(bp, p->name, p->len); } -static __be32 *xdr_encode_YFSStoreStatus_mode(__be32 *bp, mode_t mode) +static s64 linux_to_yfs_time(const struct timespec64 *t) { - struct yfs_xdr_YFSStoreStatus *x = (void *)bp; - - x->mask = htonl(AFS_SET_MODE); - x->mode = htonl(mode & S_IALLUGO); - x->mtime_client = u64_to_xdr(0); - x->owner = u64_to_xdr(0); - x->group = u64_to_xdr(0); - return bp + xdr_size(x); + /* Convert to 100ns intervals. */ + return (u64)t->tv_sec * 10000000 + t->tv_nsec/100; } -static __be32 *xdr_encode_YFSStoreStatus_mtime(__be32 *bp, const struct timespec64 *t) +static __be32 *xdr_encode_YFSStoreStatus(__be32 *bp, mode_t *mode, + const struct timespec64 *t) { struct yfs_xdr_YFSStoreStatus *x = (void *)bp; + mode_t masked_mode = mode ? *mode & S_IALLUGO : 0; s64 mtime = linux_to_yfs_time(t); + u32 mask = AFS_SET_MTIME; - x->mask = htonl(AFS_SET_MTIME); - x->mode = htonl(0); + mask |= mode ? AFS_SET_MODE : 0; + + x->mask = htonl(mask); + x->mode = htonl(masked_mode); x->mtime_client = u64_to_xdr(mtime); x->owner = u64_to_xdr(0); x->group = u64_to_xdr(0); @@ -156,8 +143,8 @@ static void yfs_check_req(struct afs_call *call, __be32 *bp) pr_err("kAFS: %s: Request buffer overflow (%zu>%u)\n", call->type->name, len, call->request_size); else if (len < call->request_size) - pr_warning("kAFS: %s: Request buffer underflow (%zu<%u)\n", - call->type->name, len, call->request_size); + pr_warn("kAFS: %s: Request buffer underflow (%zu<%u)\n", + call->type->name, len, call->request_size); } /* @@ -169,39 +156,34 @@ static void xdr_dump_bad(const __be32 *bp) int i; pr_notice("YFS XDR: Bad status record\n"); - for (i = 0; i < 5 * 4 * 4; i += 16) { + for (i = 0; i < 6 * 4 * 4; i += 16) { memcpy(x, bp, 16); bp += 4; pr_notice("%03x: %08x %08x %08x %08x\n", i, ntohl(x[0]), ntohl(x[1]), ntohl(x[2]), ntohl(x[3])); } - memcpy(x, bp, 4); - pr_notice("0x50: %08x\n", ntohl(x[0])); + memcpy(x, bp, 8); + pr_notice("0x60: %08x %08x\n", ntohl(x[0]), ntohl(x[1])); } /* * Decode a YFSFetchStatus block */ -static int xdr_decode_YFSFetchStatus(struct afs_call *call, - const __be32 **_bp, - struct afs_file_status *status, - struct afs_vnode *vnode, - const afs_dataversion_t *expected_version, - struct afs_read *read_req) +static void xdr_decode_YFSFetchStatus(const __be32 **_bp, + struct afs_call *call, + struct afs_status_cb *scb) { const struct yfs_xdr_YFSFetchStatus *xdr = (const void *)*_bp; + struct afs_file_status *status = &scb->status; u32 type; - u8 flags = 0; status->abort_code = ntohl(xdr->abort_code); if (status->abort_code != 0) { - if (vnode && status->abort_code == VNOVNODE) { - set_bit(AFS_VNODE_DELETED, &vnode->flags); + if (status->abort_code == VNOVNODE) status->nlink = 0; - __afs_break_callback(vnode); - } - return 0; + scb->have_error = true; + goto advance; } type = ntohl(xdr->type); @@ -209,152 +191,50 @@ static int xdr_decode_YFSFetchStatus(struct afs_call *call, case AFS_FTYPE_FILE: case AFS_FTYPE_DIR: case AFS_FTYPE_SYMLINK: - if (type != status->type && - vnode && - !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { - pr_warning("Vnode %llx:%llx:%x changed type %u to %u\n", - vnode->fid.vid, - vnode->fid.vnode, - vnode->fid.unique, - status->type, type); - goto bad; - } status->type = type; break; default: goto bad; } -#define EXTRACT_M4(FIELD) \ - do { \ - u32 x = ntohl(xdr->FIELD); \ - if (status->FIELD != x) { \ - flags |= AFS_VNODE_META_CHANGED; \ - status->FIELD = x; \ - } \ - } while (0) - -#define EXTRACT_M8(FIELD) \ - do { \ - u64 x = xdr_to_u64(xdr->FIELD); \ - if (status->FIELD != x) { \ - flags |= AFS_VNODE_META_CHANGED; \ - status->FIELD = x; \ - } \ - } while (0) - -#define EXTRACT_D8(FIELD) \ - do { \ - u64 x = xdr_to_u64(xdr->FIELD); \ - if (status->FIELD != x) { \ - flags |= AFS_VNODE_DATA_CHANGED; \ - status->FIELD = x; \ - } \ - } while (0) - - EXTRACT_M4(nlink); - EXTRACT_D8(size); - EXTRACT_D8(data_version); - EXTRACT_M8(author); - EXTRACT_M8(owner); - EXTRACT_M8(group); - EXTRACT_M4(mode); - EXTRACT_M4(caller_access); /* call ticket dependent */ - EXTRACT_M4(anon_access); - - status->mtime_client = xdr_to_time(xdr->mtime_client); - status->mtime_server = xdr_to_time(xdr->mtime_server); - status->lock_count = ntohl(xdr->lock_count); - - if (read_req) { - read_req->data_version = status->data_version; - read_req->file_size = status->size; - } - + status->nlink = ntohl(xdr->nlink); + status->author = xdr_to_u64(xdr->author); + status->owner = xdr_to_u64(xdr->owner); + status->caller_access = ntohl(xdr->caller_access); /* Ticket dependent */ + status->anon_access = ntohl(xdr->anon_access); + status->mode = ntohl(xdr->mode) & S_IALLUGO; + status->group = xdr_to_u64(xdr->group); + status->lock_count = ntohl(xdr->lock_count); + + status->mtime_client = xdr_to_time(xdr->mtime_client); + status->mtime_server = xdr_to_time(xdr->mtime_server); + status->size = xdr_to_u64(xdr->size); + status->data_version = xdr_to_u64(xdr->data_version); + scb->have_status = true; +advance: *_bp += xdr_size(xdr); - - if (vnode) { - if (test_bit(AFS_VNODE_UNSET, &vnode->flags)) - flags |= AFS_VNODE_NOT_YET_SET; - afs_update_inode_from_status(vnode, status, expected_version, - flags); - } - - return 0; + return; bad: xdr_dump_bad(*_bp); - return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status); -} - -/* - * Decode the file status. We need to lock the target vnode if we're going to - * update its status so that stat() sees the attributes update atomically. - */ -static int yfs_decode_status(struct afs_call *call, - const __be32 **_bp, - struct afs_file_status *status, - struct afs_vnode *vnode, - const afs_dataversion_t *expected_version, - struct afs_read *read_req) -{ - int ret; - - if (!vnode) - return xdr_decode_YFSFetchStatus(call, _bp, status, vnode, - expected_version, read_req); - - write_seqlock(&vnode->cb_lock); - ret = xdr_decode_YFSFetchStatus(call, _bp, status, vnode, - expected_version, read_req); - write_sequnlock(&vnode->cb_lock); - return ret; + afs_protocol_error(call, afs_eproto_bad_status); + goto advance; } /* * Decode a YFSCallBack block */ -static void xdr_decode_YFSCallBack(struct afs_call *call, - struct afs_vnode *vnode, - const __be32 **_bp) -{ - struct yfs_xdr_YFSCallBack *xdr = (void *)*_bp; - struct afs_cb_interest *old, *cbi = call->cbi; - u64 cb_expiry; - - write_seqlock(&vnode->cb_lock); - - if (!afs_cb_is_broken(call->cb_break, vnode, cbi)) { - cb_expiry = xdr_to_u64(xdr->expiration_time); - do_div(cb_expiry, 10 * 1000 * 1000); - vnode->cb_version = ntohl(xdr->version); - vnode->cb_type = ntohl(xdr->type); - vnode->cb_expires_at = cb_expiry + ktime_get_real_seconds(); - old = vnode->cb_interest; - if (old != call->cbi) { - vnode->cb_interest = cbi; - cbi = old; - } - set_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - } - - write_sequnlock(&vnode->cb_lock); - call->cbi = cbi; - *_bp += xdr_size(xdr); -} - -static void xdr_decode_YFSCallBack_raw(const __be32 **_bp, - struct afs_callback *cb) +static void xdr_decode_YFSCallBack(const __be32 **_bp, + struct afs_call *call, + struct afs_status_cb *scb) { struct yfs_xdr_YFSCallBack *x = (void *)*_bp; - u64 cb_expiry; - - cb_expiry = xdr_to_u64(x->expiration_time); - do_div(cb_expiry, 10 * 1000 * 1000); - cb->version = ntohl(x->version); - cb->type = ntohl(x->type); - cb->expires_at = cb_expiry + ktime_get_real_seconds(); + struct afs_callback *cb = &scb->callback; + ktime_t cb_expiry; + cb_expiry = ktime_add(call->issue_time, xdr_to_u64(x->expiration_time) * 100); + cb->expires_at = ktime_divns(cb_expiry, NSEC_PER_SEC); + scb->have_cb = true; *_bp += xdr_size(x); } @@ -365,12 +245,15 @@ static void xdr_decode_YFSVolSync(const __be32 **_bp, struct afs_volsync *volsync) { struct yfs_xdr_YFSVolSync *x = (void *)*_bp; - u64 creation; + u64 creation, update; if (volsync) { creation = xdr_to_u64(x->vol_creation_date); do_div(creation, 10 * 1000 * 1000); volsync->creation = creation; + update = xdr_to_u64(x->vol_update_date); + do_div(update, 10 * 1000 * 1000); + volsync->update = update; } *_bp += xdr_size(x); @@ -442,11 +325,12 @@ static void xdr_decode_YFSFetchVolumeStatus(const __be32 **_bp, } /* - * deliver reply data to an FS.FetchStatus + * Deliver reply data to operations that just return a file status and a volume + * sync record. */ -static int yfs_deliver_fs_fetch_status_vnode(struct afs_call *call) +static int yfs_deliver_status_and_volsync(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; const __be32 *bp; int ret; @@ -454,153 +338,80 @@ static int yfs_deliver_fs_fetch_status_vnode(struct afs_call *call) if (ret < 0) return ret; - _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); - - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSCallBack(call, vnode, &bp); - xdr_decode_YFSVolSync(&bp, call->reply[1]); + xdr_decode_YFSFetchStatus(&bp, call, &op->file[0].scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; } /* - * YFS.FetchStatus operation type - */ -static const struct afs_call_type yfs_RXYFSFetchStatus_vnode = { - .name = "YFS.FetchStatus(vnode)", - .op = yfs_FS_FetchStatus, - .deliver = yfs_deliver_fs_fetch_status_vnode, - .destructor = afs_flat_call_destructor, -}; - -/* - * Fetch the status information for a file. - */ -int yfs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsync, - bool new_inode) -{ - struct afs_vnode *vnode = fc->vnode; - struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - __be32 *bp; - - _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); - - call = afs_alloc_flat_call(net, &yfs_RXYFSFetchStatus_vnode, - sizeof(__be32) * 2 + - sizeof(struct yfs_xdr_YFSFid), - sizeof(struct yfs_xdr_YFSFetchStatus) + - sizeof(struct yfs_xdr_YFSCallBack) + - sizeof(struct yfs_xdr_YFSVolSync)); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = volsync; - call->expected_version = new_inode ? 1 : vnode->status.data_version; - - /* marshall the parameters */ - bp = call->request; - bp = xdr_encode_u32(bp, YFSFETCHSTATUS); - bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - yfs_check_req(call, bp); - - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); -} - -/* * Deliver reply data to an YFS.FetchData64. */ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; - struct afs_read *req = call->reply[2]; + struct afs_operation *op = call->op; + struct netfs_io_subrequest *subreq = op->fetch.subreq; + struct afs_vnode_param *vp = &op->file[0]; const __be32 *bp; - unsigned int size; + size_t count_before; int ret; - _enter("{%u,%zu/%llu}", - call->unmarshall, iov_iter_count(&call->iter), req->actual_len); + _enter("{%u,%zu, %zu/%llu}", + call->unmarshall, call->iov_len, iov_iter_count(call->iter), + call->remaining); switch (call->unmarshall) { case 0: - req->actual_len = 0; - req->index = 0; - req->offset = req->pos & (PAGE_SIZE - 1); + call->remaining = 0; afs_extract_to_tmp64(call); call->unmarshall++; + fallthrough; - /* extract the returned data length */ + /* Extract the returned data length into ->actual_len. This + * may indicate more or less data than was requested will be + * returned. + */ case 1: _debug("extract data length"); ret = afs_extract_data(call, true); if (ret < 0) return ret; - req->actual_len = be64_to_cpu(call->tmp64); - _debug("DATA length: %llu", req->actual_len); - req->remain = min(req->len, req->actual_len); - if (req->remain == 0) + call->remaining = be64_to_cpu(call->tmp64); + _debug("DATA length: %llu", call->remaining); + + if (call->remaining == 0) goto no_more_data; + call->iter = &subreq->io_iter; + call->iov_len = min(call->remaining, subreq->len - subreq->transferred); call->unmarshall++; - - begin_page: - ASSERTCMP(req->index, <, req->nr_pages); - if (req->remain > PAGE_SIZE - req->offset) - size = PAGE_SIZE - req->offset; - else - size = req->remain; - call->bvec[0].bv_len = size; - call->bvec[0].bv_offset = req->offset; - call->bvec[0].bv_page = req->pages[req->index]; - iov_iter_bvec(&call->iter, READ, call->bvec, 1, size); - ASSERTCMP(size, <=, PAGE_SIZE); + fallthrough; /* extract the returned data */ case 2: - _debug("extract data %zu/%llu", - iov_iter_count(&call->iter), req->remain); + count_before = call->iov_len; + _debug("extract data %zu/%llu", count_before, call->remaining); ret = afs_extract_data(call, true); + subreq->transferred += count_before - call->iov_len; if (ret < 0) return ret; - req->remain -= call->bvec[0].bv_len; - req->offset += call->bvec[0].bv_len; - ASSERTCMP(req->offset, <=, PAGE_SIZE); - if (req->offset == PAGE_SIZE) { - req->offset = 0; - if (req->page_done) - req->page_done(call, req); - req->index++; - if (req->remain > 0) - goto begin_page; - } - ASSERTCMP(req->remain, ==, 0); - if (req->actual_len <= req->len) + call->iter = &call->def_iter; + if (call->remaining) goto no_more_data; /* Discard any excess data the server gave us */ - iov_iter_discard(&call->iter, READ, req->actual_len - req->len); + afs_extract_discard(call, call->remaining); call->unmarshall = 3; + fallthrough; + case 3: _debug("extract discard %zu/%llu", - iov_iter_count(&call->iter), req->actual_len - req->len); + iov_iter_count(call->iter), call->remaining); ret = afs_extract_data(call, true); if (ret < 0) @@ -612,6 +423,7 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSCallBack) + sizeof(struct yfs_xdr_YFSVolSync)); + fallthrough; /* extract the metadata */ case 4: @@ -620,65 +432,52 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) return ret; bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &vnode->status.data_version, req); - if (ret < 0) - return ret; - xdr_decode_YFSCallBack(call, vnode, &bp); - xdr_decode_YFSVolSync(&bp, call->reply[1]); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSCallBack(&bp, call, &vp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); + + if (subreq->start + subreq->transferred >= vp->scb.status.size) + __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); call->unmarshall++; + fallthrough; case 5: break; } - for (; req->index < req->nr_pages; req->index++) { - if (req->offset < PAGE_SIZE) - zero_user_segment(req->pages[req->index], - req->offset, PAGE_SIZE); - if (req->page_done) - req->page_done(call, req); - req->offset = 0; - } - _leave(" = 0 [done]"); return 0; } -static void yfs_fetch_data_destructor(struct afs_call *call) -{ - struct afs_read *req = call->reply[2]; - - afs_put_read(req); - afs_flat_call_destructor(call); -} - /* * YFS.FetchData64 operation type */ static const struct afs_call_type yfs_RXYFSFetchData64 = { .name = "YFS.FetchData64", .op = yfs_FS_FetchData64, + .async_rx = afs_fetch_data_async_rx, .deliver = yfs_deliver_fs_fetch_data64, - .destructor = yfs_fetch_data_destructor, + .immediate_cancel = afs_fetch_data_immediate_cancel, + .destructor = afs_flat_call_destructor, }; /* * Fetch data from a file. */ -int yfs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) +void yfs_fs_fetch_data(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct netfs_io_subrequest *subreq = op->fetch.subreq; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - _enter(",%x,{%llx:%llu},%llx,%llx", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode, - req->pos, req->len); + _enter(",%x,{%llx:%llu},%llx,%zx", + key_serial(op->key), vp->fid.vid, vp->fid.vnode, + subreq->start + subreq->transferred, + subreq->len - subreq->transferred); - call = afs_alloc_flat_call(net, &yfs_RXYFSFetchData64, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSFetchData64, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid) + sizeof(struct yfs_xdr_u64) * 2, @@ -686,29 +485,23 @@ int yfs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) sizeof(struct yfs_xdr_YFSCallBack) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; + return afs_op_nomem(op); - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = NULL; /* volsync */ - call->reply[2] = req; - call->expected_version = vnode->status.data_version; - call->want_reply_time = true; + if (op->flags & AFS_OPERATION_ASYNC) + call->async = true; /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSFETCHDATA64); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - bp = xdr_encode_u64(bp, req->pos); - bp = xdr_encode_u64(bp, req->len); + bp = xdr_encode_YFSFid(bp, &vp->fid); + bp = xdr_encode_u64(bp, subreq->start + subreq->transferred); + bp = xdr_encode_u64(bp, subreq->len - subreq->transferred); yfs_check_req(call, bp); - refcount_inc(&req->usage); - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -716,7 +509,9 @@ int yfs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) */ static int yfs_deliver_fs_create_vnode(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; @@ -728,16 +523,11 @@ static int yfs_deliver_fs_create_vnode(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_YFSFid(&bp, call->reply[1]); - ret = yfs_decode_status(call, &bp, call->reply[2], NULL, NULL, NULL); - if (ret < 0) - return ret; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSCallBack_raw(&bp, call->reply[3]); - xdr_decode_YFSVolSync(&bp, NULL); + xdr_decode_YFSFid(&bp, &op->file[1].fid); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_YFSCallBack(&bp, call, &vp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -756,27 +546,20 @@ static const struct afs_call_type afs_RXFSCreateFile = { /* * Create a file. */ -int yfs_fs_create_file(struct afs_fs_cursor *fc, - const char *name, - umode_t mode, - u64 current_data_version, - struct afs_fid *newfid, - struct afs_file_status *newstatus, - struct afs_callback *newcb) +void yfs_fs_create_file(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - size_t namesz, reqsz, rplsz; + size_t reqsz, rplsz; __be32 *bp; _enter(""); - namesz = strlen(name); reqsz = (sizeof(__be32) + sizeof(__be32) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz) + + xdr_strlen(name->len) + sizeof(struct yfs_xdr_YFSStoreStatus) + sizeof(__be32)); rplsz = (sizeof(struct yfs_xdr_YFSFid) + @@ -785,30 +568,23 @@ int yfs_fs_create_file(struct afs_fs_cursor *fc, sizeof(struct yfs_xdr_YFSCallBack) + sizeof(struct yfs_xdr_YFSVolSync)); - call = afs_alloc_flat_call(net, &afs_RXFSCreateFile, reqsz, rplsz); + call = afs_alloc_flat_call(op->net, &afs_RXFSCreateFile, reqsz, rplsz); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = newfid; - call->reply[2] = newstatus; - call->reply[3] = newcb; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSCREATEFILE); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - bp = xdr_encode_string(bp, name, namesz); - bp = xdr_encode_YFSStoreStatus_mode(bp, mode); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); + bp = xdr_encode_YFSStoreStatus(bp, &op->create.mode, &op->mtime); bp = xdr_encode_u32(bp, yfs_LockNone); /* ViceLockType */ yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } static const struct afs_call_type yfs_RXFSMakeDir = { @@ -821,27 +597,20 @@ static const struct afs_call_type yfs_RXFSMakeDir = { /* * Make a directory. */ -int yfs_fs_make_dir(struct afs_fs_cursor *fc, - const char *name, - umode_t mode, - u64 current_data_version, - struct afs_fid *newfid, - struct afs_file_status *newstatus, - struct afs_callback *newcb) +void yfs_fs_make_dir(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - size_t namesz, reqsz, rplsz; + size_t reqsz, rplsz; __be32 *bp; _enter(""); - namesz = strlen(name); reqsz = (sizeof(__be32) + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz) + + xdr_strlen(name->len) + sizeof(struct yfs_xdr_YFSStoreStatus)); rplsz = (sizeof(struct yfs_xdr_YFSFid) + sizeof(struct yfs_xdr_YFSFetchStatus) + @@ -849,29 +618,22 @@ int yfs_fs_make_dir(struct afs_fs_cursor *fc, sizeof(struct yfs_xdr_YFSCallBack) + sizeof(struct yfs_xdr_YFSVolSync)); - call = afs_alloc_flat_call(net, &yfs_RXFSMakeDir, reqsz, rplsz); + call = afs_alloc_flat_call(op->net, &yfs_RXFSMakeDir, reqsz, rplsz); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = newfid; - call->reply[2] = newstatus; - call->reply[3] = newcb; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSMAKEDIR); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - bp = xdr_encode_string(bp, name, namesz); - bp = xdr_encode_YFSStoreStatus_mode(bp, mode); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); + bp = xdr_encode_YFSStoreStatus(bp, &op->create.mode, &op->mtime); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -879,8 +641,9 @@ int yfs_fs_make_dir(struct afs_fs_cursor *fc, */ static int yfs_deliver_fs_remove_file2(struct afs_call *call) { - struct afs_vnode *dvnode = call->reply[0]; - struct afs_vnode *vnode = call->reply[1]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_fid fid; const __be32 *bp; int ret; @@ -891,23 +654,26 @@ static int yfs_deliver_fs_remove_file2(struct afs_call *call) if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &dvnode->status, dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - + xdr_decode_YFSFetchStatus(&bp, call, &dvp->scb); xdr_decode_YFSFid(&bp, &fid); - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, NULL, NULL); - if (ret < 0) - return ret; + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); /* Was deleted if vnode->status.abort_code == VNOVNODE. */ - xdr_decode_YFSVolSync(&bp, NULL); + xdr_decode_YFSVolSync(&bp, &op->volsync); return 0; } +static void yfs_done_fs_remove_file2(struct afs_call *call) +{ + if (call->error == -ECONNABORTED && + (call->abort_code == RX_INVALID_OPERATION || + call->abort_code == RXGEN_OPCODE)) { + set_bit(AFS_SERVER_FL_NO_RM2, &call->op->server->flags); + call->op->flags |= AFS_OPERATION_DOWNGRADE; + } +} + /* * YFS.RemoveFile2 operation type. */ @@ -915,53 +681,45 @@ static const struct afs_call_type yfs_RXYFSRemoveFile2 = { .name = "YFS.RemoveFile2", .op = yfs_FS_RemoveFile2, .deliver = yfs_deliver_fs_remove_file2, + .done = yfs_done_fs_remove_file2, .destructor = afs_flat_call_destructor, }; /* * Remove a file and retrieve new file status. */ -int yfs_fs_remove_file2(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name, u64 current_data_version) +void yfs_fs_remove_file2(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + struct afs_vnode_param *dvp = &op->file[0]; + const struct qstr *name = &op->dentry->d_name; struct afs_call *call; - struct afs_net *net = afs_v2net(dvnode); - size_t namesz; __be32 *bp; _enter(""); - namesz = strlen(name); - - call = afs_alloc_flat_call(net, &yfs_RXYFSRemoveFile2, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRemoveFile2, sizeof(__be32) + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz), + xdr_strlen(name->len), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSFid) + sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = vnode; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSREMOVEFILE2); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &dvnode->fid); - bp = xdr_encode_string(bp, name, namesz); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -969,7 +727,8 @@ int yfs_fs_remove_file2(struct afs_fs_cursor *fc, struct afs_vnode *vnode, */ static int yfs_deliver_fs_remove(struct afs_call *call) { - struct afs_vnode *dvnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; const __be32 *bp; int ret; @@ -979,14 +738,9 @@ static int yfs_deliver_fs_remove(struct afs_call *call) if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &dvnode->status, dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - - xdr_decode_YFSVolSync(&bp, NULL); + xdr_decode_YFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); return 0; } @@ -1000,6 +754,44 @@ static const struct afs_call_type yfs_RXYFSRemoveFile = { .destructor = afs_flat_call_destructor, }; +/* + * Remove a file. + */ +void yfs_fs_remove_file(struct afs_operation *op) +{ + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_call *call; + __be32 *bp; + + _enter(""); + + if (!test_bit(AFS_SERVER_FL_NO_RM2, &op->server->flags)) + return yfs_fs_remove_file2(op); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRemoveFile, + sizeof(__be32) + + sizeof(struct yfs_xdr_RPCFlags) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(name->len), + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSREMOVEFILE); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); + yfs_check_req(call, bp); + + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); +} + static const struct afs_call_type yfs_RXYFSRemoveDir = { .name = "YFS.RemoveDir", .op = yfs_FS_RemoveDir, @@ -1008,47 +800,38 @@ static const struct afs_call_type yfs_RXYFSRemoveDir = { }; /* - * remove a file or directory + * Remove a directory. */ -int yfs_fs_remove(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name, bool isdir, u64 current_data_version) +void yfs_fs_remove_dir(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(dvnode); - size_t namesz; __be32 *bp; _enter(""); - namesz = strlen(name); - call = afs_alloc_flat_call( - net, isdir ? &yfs_RXYFSRemoveDir : &yfs_RXYFSRemoveFile, - sizeof(__be32) + - sizeof(struct yfs_xdr_RPCFlags) + - sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz), - sizeof(struct yfs_xdr_YFSFetchStatus) + - sizeof(struct yfs_xdr_YFSVolSync)); + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRemoveDir, + sizeof(__be32) + + sizeof(struct yfs_xdr_RPCFlags) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(name->len), + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = vnode; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; - bp = xdr_encode_u32(bp, isdir ? YFSREMOVEDIR : YFSREMOVEFILE); + bp = xdr_encode_u32(bp, YFSREMOVEDIR); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &dvnode->fid); - bp = xdr_encode_string(bp, name, namesz); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1056,7 +839,9 @@ int yfs_fs_remove(struct afs_fs_cursor *fc, struct afs_vnode *vnode, */ static int yfs_deliver_fs_link(struct afs_call *call) { - struct afs_vnode *dvnode = call->reply[0], *vnode = call->reply[1]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; @@ -1066,16 +851,10 @@ static int yfs_deliver_fs_link(struct afs_call *call) if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, NULL, NULL); - if (ret < 0) - return ret; - ret = yfs_decode_status(call, &bp, &dvnode->status, dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSVolSync(&bp, NULL); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; } @@ -1093,47 +872,40 @@ static const struct afs_call_type yfs_RXYFSLink = { /* * Make a hard link. */ -int yfs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name, u64 current_data_version) +void yfs_fs_link(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - size_t namesz; __be32 *bp; _enter(""); - namesz = strlen(name); - call = afs_alloc_flat_call(net, &yfs_RXYFSLink, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSLink, sizeof(__be32) + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz) + + xdr_strlen(name->len) + sizeof(struct yfs_xdr_YFSFid), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = vnode; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSLINK); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &dvnode->fid); - bp = xdr_encode_string(bp, name, namesz); - bp = xdr_encode_YFSFid(bp, &vnode->fid); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); + bp = xdr_encode_YFSFid(bp, &vp->fid); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call1(call, &vp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1141,7 +913,9 @@ int yfs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, */ static int yfs_deliver_fs_symlink(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; const __be32 *bp; int ret; @@ -1153,15 +927,10 @@ static int yfs_deliver_fs_symlink(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_YFSFid(&bp, call->reply[1]); - ret = yfs_decode_status(call, &bp, call->reply[2], NULL, NULL, NULL); - if (ret < 0) - return ret; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSVolSync(&bp, NULL); + xdr_decode_YFSFid(&bp, &vp->fid); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &dvp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1180,28 +949,23 @@ static const struct afs_call_type yfs_RXYFSSymlink = { /* * Create a symbolic link. */ -int yfs_fs_symlink(struct afs_fs_cursor *fc, - const char *name, - const char *contents, - u64 current_data_version, - struct afs_fid *newfid, - struct afs_file_status *newstatus) +void yfs_fs_symlink(struct afs_operation *op) { - struct afs_vnode *dvnode = fc->vnode; + const struct qstr *name = &op->dentry->d_name; + struct afs_vnode_param *dvp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(dvnode); - size_t namesz, contents_sz; + size_t contents_sz; + mode_t mode = 0777; __be32 *bp; _enter(""); - namesz = strlen(name); - contents_sz = strlen(contents); - call = afs_alloc_flat_call(net, &yfs_RXYFSSymlink, + contents_sz = strlen(op->create.symlink); + call = afs_alloc_flat_call(op->net, &yfs_RXYFSSymlink, sizeof(__be32) + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(namesz) + + xdr_strlen(name->len) + xdr_strlen(contents_sz) + sizeof(struct yfs_xdr_YFSStoreStatus), sizeof(struct yfs_xdr_YFSFid) + @@ -1209,27 +973,21 @@ int yfs_fs_symlink(struct afs_fs_cursor *fc, sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = dvnode; - call->reply[1] = newfid; - call->reply[2] = newstatus; - call->expected_version = current_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSSYMLINK); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &dvnode->fid); - bp = xdr_encode_string(bp, name, namesz); - bp = xdr_encode_string(bp, contents, contents_sz); - bp = xdr_encode_YFSStoreStatus_mode(bp, S_IRWXUGO); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_name(bp, name); + bp = xdr_encode_string(bp, op->create.symlink, contents_sz); + bp = xdr_encode_YFSStoreStatus(bp, &mode, &op->mtime); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = dvp->fid; + trace_afs_make_fs_call1(call, &dvp->fid, name); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1237,8 +995,9 @@ int yfs_fs_symlink(struct afs_fs_cursor *fc, */ static int yfs_deliver_fs_rename(struct afs_call *call) { - struct afs_vnode *orig_dvnode = call->reply[0]; - struct afs_vnode *new_dvnode = call->reply[1]; + struct afs_operation *op = call->op; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; const __be32 *bp; int ret; @@ -1248,20 +1007,13 @@ static int yfs_deliver_fs_rename(struct afs_call *call) if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &orig_dvnode->status, orig_dvnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - if (new_dvnode != orig_dvnode) { - ret = yfs_decode_status(call, &bp, &new_dvnode->status, new_dvnode, - &call->expected_version_2, NULL); - if (ret < 0) - return ret; - } - - xdr_decode_YFSVolSync(&bp, NULL); + /* If the two dirs are the same, we have two copies of the same status + * report, so we just decode it twice. + */ + xdr_decode_YFSFetchStatus(&bp, call, &orig_dvp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &new_dvp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; } @@ -1279,188 +1031,348 @@ static const struct afs_call_type yfs_RXYFSRename = { /* * Rename a file or directory. */ -int yfs_fs_rename(struct afs_fs_cursor *fc, - const char *orig_name, - struct afs_vnode *new_dvnode, - const char *new_name, - u64 current_orig_data_version, - u64 current_new_data_version) +void yfs_fs_rename(struct afs_operation *op) { - struct afs_vnode *orig_dvnode = fc->vnode; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + const struct qstr *orig_name = &op->dentry->d_name; + const struct qstr *new_name = &op->dentry_2->d_name; struct afs_call *call; - struct afs_net *net = afs_v2net(orig_dvnode); - size_t o_namesz, n_namesz; __be32 *bp; _enter(""); - o_namesz = strlen(orig_name); - n_namesz = strlen(new_name); - call = afs_alloc_flat_call(net, &yfs_RXYFSRename, + if (!test_bit(AFS_SERVER_FL_NO_RENAME2, &op->server->flags)) + return yfs_fs_rename_replace(op); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRename, sizeof(__be32) + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(o_namesz) + + xdr_strlen(orig_name->len) + sizeof(struct yfs_xdr_YFSFid) + - xdr_strlen(n_namesz), + xdr_strlen(new_name->len), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = orig_dvnode; - call->reply[1] = new_dvnode; - call->expected_version = current_orig_data_version + 1; - call->expected_version_2 = current_new_data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSRENAME); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &orig_dvnode->fid); - bp = xdr_encode_string(bp, orig_name, o_namesz); - bp = xdr_encode_YFSFid(bp, &new_dvnode->fid); - bp = xdr_encode_string(bp, new_name, n_namesz); + bp = xdr_encode_YFSFid(bp, &orig_dvp->fid); + bp = xdr_encode_name(bp, orig_name); + bp = xdr_encode_YFSFid(bp, &new_dvp->fid); + bp = xdr_encode_name(bp, new_name); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &orig_dvnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = orig_dvp->fid; + trace_afs_make_fs_call2(call, &orig_dvp->fid, orig_name, new_name); + afs_make_op_call(op, call, GFP_NOFS); } /* - * Deliver reply data to a YFS.StoreData64 operation. + * Deliver reply data to a YFS.Rename_NoReplace operation. This does not + * return the status of a displaced target inode as there cannot be one. */ -static int yfs_deliver_fs_store_data(struct afs_call *call) +static int yfs_deliver_fs_rename_1(struct afs_call *call) { - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode_param *old_vp = &op->more_files[0]; const __be32 *bp; int ret; - _enter(""); + _enter("{%u}", call->unmarshall); ret = afs_transfer_reply(call); if (ret < 0) return ret; - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); + /* If the two dirs are the same, we have two copies of the same status + * report, so we just decode it twice. + */ + xdr_decode_YFSFetchStatus(&bp, call, &orig_dvp->scb); + xdr_decode_YFSFid(&bp, &old_vp->fid); + xdr_decode_YFSFetchStatus(&bp, call, &old_vp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &new_dvp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); + _leave(" = 0 [done]"); + return 0; +} + +/* + * Deliver reply data to a YFS.Rename_Replace or a YFS.Rename_Exchange + * operation. These return the status of the displaced target inode if there + * was one. + */ +static int yfs_deliver_fs_rename_2(struct afs_call *call) +{ + struct afs_operation *op = call->op; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode_param *old_vp = &op->more_files[0]; + struct afs_vnode_param *new_vp = &op->more_files[1]; + const __be32 *bp; + int ret; + + _enter("{%u}", call->unmarshall); + + ret = afs_transfer_reply(call); if (ret < 0) return ret; - xdr_decode_YFSVolSync(&bp, NULL); - - afs_pages_written_back(vnode, call); + bp = call->buffer; + /* If the two dirs are the same, we have two copies of the same status + * report, so we just decode it twice. + */ + xdr_decode_YFSFetchStatus(&bp, call, &orig_dvp->scb); + xdr_decode_YFSFid(&bp, &old_vp->fid); + xdr_decode_YFSFetchStatus(&bp, call, &old_vp->scb); + xdr_decode_YFSFetchStatus(&bp, call, &new_dvp->scb); + xdr_decode_YFSFid(&bp, &new_vp->fid); + xdr_decode_YFSFetchStatus(&bp, call, &new_vp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; } +static void yfs_done_fs_rename_replace(struct afs_call *call) +{ + if (call->error == -ECONNABORTED && + (call->abort_code == RX_INVALID_OPERATION || + call->abort_code == RXGEN_OPCODE)) { + set_bit(AFS_SERVER_FL_NO_RENAME2, &call->op->server->flags); + call->op->flags |= AFS_OPERATION_DOWNGRADE; + } +} + /* - * YFS.StoreData64 operation type. + * YFS.Rename_Replace operation type */ -static const struct afs_call_type yfs_RXYFSStoreData64 = { - .name = "YFS.StoreData64", - .op = yfs_FS_StoreData64, - .deliver = yfs_deliver_fs_store_data, +static const struct afs_call_type yfs_RXYFSRename_Replace = { + .name = "FS.Rename_Replace", + .op = yfs_FS_Rename_Replace, + .deliver = yfs_deliver_fs_rename_2, + .done = yfs_done_fs_rename_replace, .destructor = afs_flat_call_destructor, }; /* - * Store a set of pages to a large file. + * YFS.Rename_NoReplace operation type + */ +static const struct afs_call_type yfs_RXYFSRename_NoReplace = { + .name = "FS.Rename_NoReplace", + .op = yfs_FS_Rename_NoReplace, + .deliver = yfs_deliver_fs_rename_1, + .destructor = afs_flat_call_destructor, +}; + +/* + * YFS.Rename_Exchange operation type + */ +static const struct afs_call_type yfs_RXYFSRename_Exchange = { + .name = "FS.Rename_Exchange", + .op = yfs_FS_Rename_Exchange, + .deliver = yfs_deliver_fs_rename_2, + .destructor = afs_flat_call_destructor, +}; + +/* + * Rename a file or directory, replacing the target if it exists. The status + * of a displaced target is returned. */ -int yfs_fs_store_data(struct afs_fs_cursor *fc, struct address_space *mapping, - pgoff_t first, pgoff_t last, - unsigned offset, unsigned to) +void yfs_fs_rename_replace(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + const struct qstr *orig_name = &op->dentry->d_name; + const struct qstr *new_name = &op->dentry_2->d_name; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); - loff_t size, pos, i_size; __be32 *bp; - _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + _enter(""); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRename_Replace, + sizeof(__be32) + + sizeof(struct yfs_xdr_RPCFlags) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(orig_name->len) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(new_name->len), + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); - size = (loff_t)to - (loff_t)offset; - if (first != last) - size += (loff_t)(last - first) << PAGE_SHIFT; - pos = (loff_t)first << PAGE_SHIFT; - pos += offset; + /* Marshall the parameters. */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSRENAME_REPLACE); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &orig_dvp->fid); + bp = xdr_encode_name(bp, orig_name); + bp = xdr_encode_YFSFid(bp, &new_dvp->fid); + bp = xdr_encode_name(bp, new_name); + yfs_check_req(call, bp); - i_size = i_size_read(&vnode->vfs_inode); - if (pos + size > i_size) - i_size = size + pos; + call->fid = orig_dvp->fid; + trace_afs_make_fs_call2(call, &orig_dvp->fid, orig_name, new_name); + afs_make_op_call(op, call, GFP_NOFS); +} - _debug("size %llx, at %llx, i_size %llx", - (unsigned long long)size, (unsigned long long)pos, - (unsigned long long)i_size); +/* + * Rename a file or directory, failing if the target dirent exists. + */ +void yfs_fs_rename_noreplace(struct afs_operation *op) +{ + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + const struct qstr *orig_name = &op->dentry->d_name; + const struct qstr *new_name = &op->dentry_2->d_name; + struct afs_call *call; + __be32 *bp; - call = afs_alloc_flat_call(net, &yfs_RXYFSStoreData64, - sizeof(__be32) + + _enter(""); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRename_NoReplace, sizeof(__be32) + + sizeof(struct yfs_xdr_RPCFlags) + sizeof(struct yfs_xdr_YFSFid) + - sizeof(struct yfs_xdr_YFSStoreStatus) + - sizeof(struct yfs_xdr_u64) * 3, + xdr_strlen(orig_name->len) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(new_name->len), + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->mapping = mapping; - call->reply[0] = vnode; - call->first = first; - call->last = last; - call->first_offset = offset; - call->last_to = to; - call->send_pages = true; - call->expected_version = vnode->status.data_version + 1; + return afs_op_nomem(op); - /* marshall the parameters */ + /* Marshall the parameters. */ bp = call->request; - bp = xdr_encode_u32(bp, YFSSTOREDATA64); + bp = xdr_encode_u32(bp, YFSRENAME_NOREPLACE); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - bp = xdr_encode_YFSStoreStatus_mtime(bp, &vnode->vfs_inode.i_mtime); - bp = xdr_encode_u64(bp, pos); - bp = xdr_encode_u64(bp, size); - bp = xdr_encode_u64(bp, i_size); + bp = xdr_encode_YFSFid(bp, &orig_dvp->fid); + bp = xdr_encode_name(bp, orig_name); + bp = xdr_encode_YFSFid(bp, &new_dvp->fid); + bp = xdr_encode_name(bp, new_name); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = orig_dvp->fid; + trace_afs_make_fs_call2(call, &orig_dvp->fid, orig_name, new_name); + afs_make_op_call(op, call, GFP_NOFS); } /* - * deliver reply data to an FS.StoreStatus + * Exchange a pair of files directories. */ -static int yfs_deliver_fs_store_status(struct afs_call *call) +void yfs_fs_rename_exchange(struct afs_operation *op) { - struct afs_vnode *vnode = call->reply[0]; - const __be32 *bp; - int ret; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + const struct qstr *orig_name = &op->dentry->d_name; + const struct qstr *new_name = &op->dentry_2->d_name; + struct afs_call *call; + __be32 *bp; _enter(""); - ret = afs_transfer_reply(call); - if (ret < 0) - return ret; + call = afs_alloc_flat_call(op->net, &yfs_RXYFSRename_Exchange, + sizeof(__be32) + + sizeof(struct yfs_xdr_RPCFlags) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(orig_name->len) + + sizeof(struct yfs_xdr_YFSFid) + + xdr_strlen(new_name->len), + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); - /* unmarshall the reply once we've received all of it */ - bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSVolSync(&bp, NULL); + /* Marshall the parameters. */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSRENAME_EXCHANGE); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &orig_dvp->fid); + bp = xdr_encode_name(bp, orig_name); + bp = xdr_encode_YFSFid(bp, &new_dvp->fid); + bp = xdr_encode_name(bp, new_name); + yfs_check_req(call, bp); - _leave(" = 0 [done]"); - return 0; + call->fid = orig_dvp->fid; + trace_afs_make_fs_call2(call, &orig_dvp->fid, orig_name, new_name); + afs_make_op_call(op, call, GFP_NOFS); +} + +/* + * YFS.StoreData64 operation type. + */ +static const struct afs_call_type yfs_RXYFSStoreData64 = { + .name = "YFS.StoreData64", + .op = yfs_FS_StoreData64, + .deliver = yfs_deliver_status_and_volsync, + .destructor = afs_flat_call_destructor, +}; + +/* + * Store a set of pages to a large file. + */ +void yfs_fs_store_data(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%llx:%llu},,", + key_serial(op->key), vp->fid.vid, vp->fid.vnode); + + _debug("size %llx, at %llx, i_size %llx", + (unsigned long long)op->store.size, + (unsigned long long)op->store.pos, + (unsigned long long)op->store.i_size); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSStoreData64, + sizeof(__be32) + + sizeof(__be32) + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(struct yfs_xdr_YFSStoreStatus) + + sizeof(struct yfs_xdr_u64) * 3, + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); + + call->write_iter = op->store.write_iter; + + /* marshall the parameters */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSSTOREDATA64); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &vp->fid); + bp = xdr_encode_YFSStoreStatus(bp, NULL, &op->mtime); + bp = xdr_encode_u64(bp, op->store.pos); + bp = xdr_encode_u64(bp, op->store.size); + bp = xdr_encode_u64(bp, op->store.i_size); + yfs_check_req(call, bp); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1469,14 +1381,14 @@ static int yfs_deliver_fs_store_status(struct afs_call *call) static const struct afs_call_type yfs_RXYFSStoreStatus = { .name = "YFS.StoreStatus", .op = yfs_FS_StoreStatus, - .deliver = yfs_deliver_fs_store_status, + .deliver = yfs_deliver_status_and_volsync, .destructor = afs_flat_call_destructor, }; static const struct afs_call_type yfs_RXYFSStoreData64_as_Status = { .name = "YFS.StoreData64", .op = yfs_FS_StoreData64, - .deliver = yfs_deliver_fs_store_status, + .deliver = yfs_deliver_status_and_volsync, .destructor = afs_flat_call_destructor, }; @@ -1484,17 +1396,17 @@ static const struct afs_call_type yfs_RXYFSStoreData64_as_Status = { * Set the attributes on a file, using YFS.StoreData64 rather than * YFS.StoreStatus so as to alter the file size also. */ -static int yfs_fs_setattr_size(struct afs_fs_cursor *fc, struct iattr *attr) +static void yfs_fs_setattr_size(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); + struct iattr *attr = op->setattr.attr; __be32 *bp; _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &yfs_RXYFSStoreData64_as_Status, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSStoreData64_as_Status, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid) + sizeof(struct yfs_xdr_YFSStoreStatus) + @@ -1502,69 +1414,61 @@ static int yfs_fs_setattr_size(struct afs_fs_cursor *fc, struct iattr *attr) sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->expected_version = vnode->status.data_version + 1; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSSTOREDATA64); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); bp = xdr_encode_YFS_StoreStatus(bp, attr); - bp = xdr_encode_u64(bp, 0); /* position of start of write */ + bp = xdr_encode_u64(bp, attr->ia_size); /* position of start of write */ bp = xdr_encode_u64(bp, 0); /* size of write */ bp = xdr_encode_u64(bp, attr->ia_size); /* new file length */ yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * Set the attributes on a file, using YFS.StoreData64 if there's a change in * file size, and YFS.StoreStatus otherwise. */ -int yfs_fs_setattr(struct afs_fs_cursor *fc, struct iattr *attr) +void yfs_fs_setattr(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); + struct iattr *attr = op->setattr.attr; __be32 *bp; if (attr->ia_valid & ATTR_SIZE) - return yfs_fs_setattr_size(fc, attr); + return yfs_fs_setattr_size(op); _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &yfs_RXYFSStoreStatus, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSStoreStatus, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid) + sizeof(struct yfs_xdr_YFSStoreStatus), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; - call->expected_version = vnode->status.data_version; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSSTORESTATUS); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); bp = xdr_encode_YFS_StoreStatus(bp, attr); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1572,6 +1476,7 @@ int yfs_fs_setattr(struct afs_fs_cursor *fc, struct iattr *attr) */ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) { + struct afs_operation *op = call->op; const __be32 *bp; char *p; u32 size; @@ -1583,6 +1488,7 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) case 0: call->unmarshall++; afs_extract_to_buf(call, sizeof(struct yfs_xdr_YFSFetchVolumeStatus)); + fallthrough; /* extract the returned status record */ case 1: @@ -1592,9 +1498,10 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) return ret; bp = call->buffer; - xdr_decode_YFSFetchVolumeStatus(&bp, call->reply[1]); + xdr_decode_YFSFetchVolumeStatus(&bp, &op->volstatus.vs); call->unmarshall++; afs_extract_to_tmp(call); + fallthrough; /* extract the volume name length */ case 2: @@ -1605,11 +1512,11 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("volname length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_volname_len); + return afs_protocol_error(call, afs_eproto_volname_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the volume name */ case 3: @@ -1618,11 +1525,12 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("volname '%s'", p); afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* extract the offline message length */ case 4: @@ -1633,11 +1541,11 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("offline msg length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_offline_msg_len); + return afs_protocol_error(call, afs_eproto_offline_msg_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the offline message */ case 5: @@ -1646,12 +1554,13 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("offline '%s'", p); afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* extract the message of the day length */ case 6: @@ -1662,11 +1571,11 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("motd length: %u", call->count); if (call->count >= AFSNAMEMAX) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_motd_len); + return afs_protocol_error(call, afs_eproto_motd_len); size = (call->count + 3) & ~3; /* It's padded */ - afs_extract_begin(call, call->reply[2], size); + afs_extract_to_buf(call, size); call->unmarshall++; + fallthrough; /* extract the message of the day */ case 7: @@ -1675,11 +1584,12 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) if (ret < 0) return ret; - p = call->reply[2]; + p = call->buffer; p[call->count] = 0; _debug("motd '%s'", p); call->unmarshall++; + fallthrough; case 8: break; @@ -1690,95 +1600,46 @@ static int yfs_deliver_fs_get_volume_status(struct afs_call *call) } /* - * Destroy a YFS.GetVolumeStatus call. - */ -static void yfs_get_volume_status_call_destructor(struct afs_call *call) -{ - kfree(call->reply[2]); - call->reply[2] = NULL; - afs_flat_call_destructor(call); -} - -/* * YFS.GetVolumeStatus operation type */ static const struct afs_call_type yfs_RXYFSGetVolumeStatus = { .name = "YFS.GetVolumeStatus", .op = yfs_FS_GetVolumeStatus, .deliver = yfs_deliver_fs_get_volume_status, - .destructor = yfs_get_volume_status_call_destructor, + .destructor = afs_flat_call_destructor, }; /* * fetch the status of a volume */ -int yfs_fs_get_volume_status(struct afs_fs_cursor *fc, - struct afs_volume_status *vs) +void yfs_fs_get_volume_status(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; - void *tmpbuf; _enter(""); - tmpbuf = kmalloc(AFSOPAQUEMAX, GFP_KERNEL); - if (!tmpbuf) - return -ENOMEM; - - call = afs_alloc_flat_call(net, &yfs_RXYFSGetVolumeStatus, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSGetVolumeStatus, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_u64), - sizeof(struct yfs_xdr_YFSFetchVolumeStatus) + - sizeof(__be32)); - if (!call) { - kfree(tmpbuf); - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = vnode; - call->reply[1] = vs; - call->reply[2] = tmpbuf; + max_t(size_t, + sizeof(struct yfs_xdr_YFSFetchVolumeStatus) + + sizeof(__be32), + AFSOPAQUEMAX + 1)); + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSGETVOLUMESTATUS); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_u64(bp, vnode->fid.vid); + bp = xdr_encode_u64(bp, vp->fid.vid); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); -} - -/* - * Deliver reply data to an YFS.SetLock, YFS.ExtendLock or YFS.ReleaseLock - */ -static int yfs_deliver_fs_xxxx_lock(struct afs_call *call) -{ - struct afs_vnode *vnode = call->reply[0]; - const __be32 *bp; - int ret; - - _enter("{%u}", call->unmarshall); - - ret = afs_transfer_reply(call); - if (ret < 0) - return ret; - - /* unmarshall the reply once we've received all of it */ - bp = call->buffer; - ret = yfs_decode_status(call, &bp, &vnode->status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSVolSync(&bp, NULL); - - _leave(" = 0 [done]"); - return 0; + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -1787,7 +1648,8 @@ static int yfs_deliver_fs_xxxx_lock(struct afs_call *call) static const struct afs_call_type yfs_RXYFSSetLock = { .name = "YFS.SetLock", .op = yfs_FS_SetLock, - .deliver = yfs_deliver_fs_xxxx_lock, + .deliver = yfs_deliver_status_and_volsync, + .done = afs_lock_op_done, .destructor = afs_flat_call_destructor, }; @@ -1797,7 +1659,8 @@ static const struct afs_call_type yfs_RXYFSSetLock = { static const struct afs_call_type yfs_RXYFSExtendLock = { .name = "YFS.ExtendLock", .op = yfs_FS_ExtendLock, - .deliver = yfs_deliver_fs_xxxx_lock, + .deliver = yfs_deliver_status_and_volsync, + .done = afs_lock_op_done, .destructor = afs_flat_call_destructor, }; @@ -1807,126 +1670,112 @@ static const struct afs_call_type yfs_RXYFSExtendLock = { static const struct afs_call_type yfs_RXYFSReleaseLock = { .name = "YFS.ReleaseLock", .op = yfs_FS_ReleaseLock, - .deliver = yfs_deliver_fs_xxxx_lock, + .deliver = yfs_deliver_status_and_volsync, .destructor = afs_flat_call_destructor, }; /* * Set a lock on a file */ -int yfs_fs_set_lock(struct afs_fs_cursor *fc, afs_lock_type_t type) +void yfs_fs_set_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; _enter(""); - call = afs_alloc_flat_call(net, &yfs_RXYFSSetLock, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSSetLock, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid) + sizeof(__be32), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSSETLOCK); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); - bp = xdr_encode_u32(bp, type); + bp = xdr_encode_YFSFid(bp, &vp->fid); + bp = xdr_encode_u32(bp, op->lock.type); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_calli(call, &vp->fid, op->lock.type); + afs_make_op_call(op, call, GFP_NOFS); } /* * extend a lock on a file */ -int yfs_fs_extend_lock(struct afs_fs_cursor *fc) +void yfs_fs_extend_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; _enter(""); - call = afs_alloc_flat_call(net, &yfs_RXYFSExtendLock, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSExtendLock, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSEXTENDLOCK); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* * release a lock on a file */ -int yfs_fs_release_lock(struct afs_fs_cursor *fc) +void yfs_fs_release_lock(struct afs_operation *op) { - struct afs_vnode *vnode = fc->vnode; + struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_net *net = afs_v2net(vnode); __be32 *bp; _enter(""); - call = afs_alloc_flat_call(net, &yfs_RXYFSReleaseLock, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSReleaseLock, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSVolSync)); if (!call) - return -ENOMEM; - - call->key = fc->key; - call->reply[0] = vnode; + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSRELEASELOCK); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, &vnode->fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); yfs_check_req(call, bp); - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &vnode->fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* - * Deliver reply data to an FS.FetchStatus with no vnode. + * Deliver a reply to YFS.FetchStatus */ static int yfs_deliver_fs_fetch_status(struct afs_call *call) { - struct afs_file_status *status = call->reply[1]; - struct afs_callback *callback = call->reply[2]; - struct afs_volsync *volsync = call->reply[3]; - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[op->fetch_status.which]; const __be32 *bp; int ret; @@ -1934,16 +1783,11 @@ static int yfs_deliver_fs_fetch_status(struct afs_call *call) if (ret < 0) return ret; - _enter("{%llx:%llu}", vnode->fid.vid, vnode->fid.vnode); - /* unmarshall the reply once we've received all of it */ bp = call->buffer; - ret = yfs_decode_status(call, &bp, status, vnode, - &call->expected_version, NULL); - if (ret < 0) - return ret; - xdr_decode_YFSCallBack_raw(&bp, callback); - xdr_decode_YFSVolSync(&bp, volsync); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSCallBack(&bp, call, &vp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); _leave(" = 0 [done]"); return 0; @@ -1962,48 +1806,34 @@ static const struct afs_call_type yfs_RXYFSFetchStatus = { /* * Fetch the status information for a fid without needing a vnode handle. */ -int yfs_fs_fetch_status(struct afs_fs_cursor *fc, - struct afs_net *net, - struct afs_fid *fid, - struct afs_file_status *status, - struct afs_callback *callback, - struct afs_volsync *volsync) +void yfs_fs_fetch_status(struct afs_operation *op) { + struct afs_vnode_param *vp = &op->file[op->fetch_status.which]; struct afs_call *call; __be32 *bp; _enter(",%x,{%llx:%llu},,", - key_serial(fc->key), fid->vid, fid->vnode); + key_serial(op->key), vp->fid.vid, vp->fid.vnode); - call = afs_alloc_flat_call(net, &yfs_RXYFSFetchStatus, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSFetchStatus, sizeof(__be32) * 2 + sizeof(struct yfs_xdr_YFSFid), sizeof(struct yfs_xdr_YFSFetchStatus) + sizeof(struct yfs_xdr_YFSCallBack) + sizeof(struct yfs_xdr_YFSVolSync)); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = NULL; /* vnode for fid[0] */ - call->reply[1] = status; - call->reply[2] = callback; - call->reply[3] = volsync; - call->expected_version = 1; /* vnode->status.data_version */ + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSFETCHSTATUS); bp = xdr_encode_u32(bp, 0); /* RPC flags */ - bp = xdr_encode_YFSFid(bp, fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); yfs_check_req(call, bp); - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, fid); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); } /* @@ -2011,9 +1841,8 @@ int yfs_fs_fetch_status(struct afs_fs_cursor *fc, */ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) { - struct afs_file_status *statuses; - struct afs_callback *callbacks; - struct afs_vnode *vnode = call->reply[0]; + struct afs_operation *op = call->op; + struct afs_status_cb *scb; const __be32 *bp; u32 tmp; int ret; @@ -2024,6 +1853,7 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) case 0: afs_extract_to_tmp(call); call->unmarshall++; + fallthrough; /* Extract the file status count and array in two steps */ case 1: @@ -2033,15 +1863,15 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; tmp = ntohl(call->tmp); - _debug("status count: %u/%u", tmp, call->count2); - if (tmp != call->count2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_ibulkst_count); + _debug("status count: %u/%u", tmp, op->nr_files); + if (tmp != op->nr_files) + return afs_protocol_error(call, afs_eproto_ibulkst_count); call->count = 0; call->unmarshall++; more_counts: afs_extract_to_buf(call, sizeof(struct yfs_xdr_YFSFetchStatus)); + fallthrough; case 2: _debug("extract status array %u", call->count); @@ -2049,21 +1879,29 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) if (ret < 0) return ret; + switch (call->count) { + case 0: + scb = &op->file[0].scb; + break; + case 1: + scb = &op->file[1].scb; + break; + default: + scb = &op->more_files[call->count - 2].scb; + break; + } + bp = call->buffer; - statuses = call->reply[1]; - ret = yfs_decode_status(call, &bp, &statuses[call->count], - call->count == 0 ? vnode : NULL, - NULL, NULL); - if (ret < 0) - return ret; + xdr_decode_YFSFetchStatus(&bp, call, scb); call->count++; - if (call->count < call->count2) + if (call->count < op->nr_files) goto more_counts; call->count = 0; call->unmarshall++; afs_extract_to_tmp(call); + fallthrough; /* Extract the callback count and array in two steps */ case 3: @@ -2074,13 +1912,13 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) tmp = ntohl(call->tmp); _debug("CB count: %u", tmp); - if (tmp != call->count2) - return afs_protocol_error(call, -EBADMSG, - afs_eproto_ibulkst_cb_count); + if (tmp != op->nr_files) + return afs_protocol_error(call, afs_eproto_ibulkst_cb_count); call->count = 0; call->unmarshall++; more_cbs: afs_extract_to_buf(call, sizeof(struct yfs_xdr_YFSCallBack)); + fallthrough; case 4: _debug("extract CB array"); @@ -2089,20 +1927,27 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; _debug("unmarshall CB array"); - bp = call->buffer; - callbacks = call->reply[2]; - xdr_decode_YFSCallBack_raw(&bp, &callbacks[call->count]); - statuses = call->reply[1]; - if (call->count == 0 && vnode && statuses[0].abort_code == 0) { - bp = call->buffer; - xdr_decode_YFSCallBack(call, vnode, &bp); + switch (call->count) { + case 0: + scb = &op->file[0].scb; + break; + case 1: + scb = &op->file[1].scb; + break; + default: + scb = &op->more_files[call->count - 2].scb; + break; } + + bp = call->buffer; + xdr_decode_YFSCallBack(&bp, call, scb); call->count++; - if (call->count < call->count2) + if (call->count < op->nr_files) goto more_cbs; afs_extract_to_buf(call, sizeof(struct yfs_xdr_YFSVolSync)); call->unmarshall++; + fallthrough; case 5: ret = afs_extract_data(call, false); @@ -2110,9 +1955,10 @@ static int yfs_deliver_fs_inline_bulk_status(struct afs_call *call) return ret; bp = call->buffer; - xdr_decode_YFSVolSync(&bp, call->reply[3]); + xdr_decode_YFSVolSync(&bp, &op->volsync); call->unmarshall++; + fallthrough; case 6: break; @@ -2135,50 +1981,252 @@ static const struct afs_call_type yfs_RXYFSInlineBulkStatus = { /* * Fetch the status information for up to 1024 files */ -int yfs_fs_inline_bulk_status(struct afs_fs_cursor *fc, - struct afs_net *net, - struct afs_fid *fids, - struct afs_file_status *statuses, - struct afs_callback *callbacks, - unsigned int nr_fids, - struct afs_volsync *volsync) +void yfs_fs_inline_bulk_status(struct afs_operation *op) { + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_call *call; __be32 *bp; int i; _enter(",%x,{%llx:%llu},%u", - key_serial(fc->key), fids[0].vid, fids[1].vnode, nr_fids); + key_serial(op->key), vp->fid.vid, vp->fid.vnode, op->nr_files); - call = afs_alloc_flat_call(net, &yfs_RXYFSInlineBulkStatus, + call = afs_alloc_flat_call(op->net, &yfs_RXYFSInlineBulkStatus, sizeof(__be32) + sizeof(__be32) + sizeof(__be32) + - sizeof(struct yfs_xdr_YFSFid) * nr_fids, + sizeof(struct yfs_xdr_YFSFid) * op->nr_files, sizeof(struct yfs_xdr_YFSFetchStatus)); - if (!call) { - fc->ac.error = -ENOMEM; - return -ENOMEM; - } - - call->key = fc->key; - call->reply[0] = NULL; /* vnode for fid[0] */ - call->reply[1] = statuses; - call->reply[2] = callbacks; - call->reply[3] = volsync; - call->count2 = nr_fids; + if (!call) + return afs_op_nomem(op); /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSINLINEBULKSTATUS); bp = xdr_encode_u32(bp, 0); /* RPCFlags */ - bp = xdr_encode_u32(bp, nr_fids); - for (i = 0; i < nr_fids; i++) - bp = xdr_encode_YFSFid(bp, &fids[i]); + bp = xdr_encode_u32(bp, op->nr_files); + bp = xdr_encode_YFSFid(bp, &dvp->fid); + bp = xdr_encode_YFSFid(bp, &vp->fid); + for (i = 0; i < op->nr_files - 2; i++) + bp = xdr_encode_YFSFid(bp, &op->more_files[i].fid); + yfs_check_req(call, bp); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_NOFS); +} + +/* + * Deliver reply data to an YFS.FetchOpaqueACL. + */ +static int yfs_deliver_fs_fetch_opaque_acl(struct afs_call *call) +{ + struct afs_operation *op = call->op; + struct afs_vnode_param *vp = &op->file[0]; + struct yfs_acl *yacl = op->yacl; + struct afs_acl *acl; + const __be32 *bp; + unsigned int size; + int ret; + + _enter("{%u}", call->unmarshall); + + switch (call->unmarshall) { + case 0: + afs_extract_to_tmp(call); + call->unmarshall++; + fallthrough; + + /* Extract the file ACL length */ + case 1: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + size = call->count2 = ntohl(call->tmp); + size = round_up(size, 4); + + if (yacl->flags & YFS_ACL_WANT_ACL) { + acl = kmalloc(struct_size(acl, data, size), GFP_KERNEL); + if (!acl) + return -ENOMEM; + yacl->acl = acl; + acl->size = call->count2; + afs_extract_begin(call, acl->data, size); + } else { + afs_extract_discard(call, size); + } + call->unmarshall++; + fallthrough; + + /* Extract the file ACL */ + case 2: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + afs_extract_to_tmp(call); + call->unmarshall++; + fallthrough; + + /* Extract the volume ACL length */ + case 3: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + size = call->count2 = ntohl(call->tmp); + size = round_up(size, 4); + + if (yacl->flags & YFS_ACL_WANT_VOL_ACL) { + acl = kmalloc(struct_size(acl, data, size), GFP_KERNEL); + if (!acl) + return -ENOMEM; + yacl->vol_acl = acl; + acl->size = call->count2; + afs_extract_begin(call, acl->data, size); + } else { + afs_extract_discard(call, size); + } + call->unmarshall++; + fallthrough; + + /* Extract the volume ACL */ + case 4: + ret = afs_extract_data(call, true); + if (ret < 0) + return ret; + + afs_extract_to_buf(call, + sizeof(__be32) * 2 + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + call->unmarshall++; + fallthrough; + + /* extract the metadata */ + case 5: + ret = afs_extract_data(call, false); + if (ret < 0) + return ret; + + bp = call->buffer; + yacl->inherit_flag = ntohl(*bp++); + yacl->num_cleaned = ntohl(*bp++); + xdr_decode_YFSFetchStatus(&bp, call, &vp->scb); + xdr_decode_YFSVolSync(&bp, &op->volsync); + + call->unmarshall++; + fallthrough; + + case 6: + break; + } + + _leave(" = 0 [done]"); + return 0; +} + +void yfs_free_opaque_acl(struct yfs_acl *yacl) +{ + if (yacl) { + kfree(yacl->acl); + kfree(yacl->vol_acl); + kfree(yacl); + } +} + +/* + * YFS.FetchOpaqueACL operation type + */ +static const struct afs_call_type yfs_RXYFSFetchOpaqueACL = { + .name = "YFS.FetchOpaqueACL", + .op = yfs_FS_FetchOpaqueACL, + .deliver = yfs_deliver_fs_fetch_opaque_acl, + .destructor = afs_flat_call_destructor, +}; + +/* + * Fetch the YFS advanced ACLs for a file. + */ +void yfs_fs_fetch_opaque_acl(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%llx:%llu},,", + key_serial(op->key), vp->fid.vid, vp->fid.vnode); + + call = afs_alloc_flat_call(op->net, &yfs_RXYFSFetchOpaqueACL, + sizeof(__be32) * 2 + + sizeof(struct yfs_xdr_YFSFid), + sizeof(__be32) * 2 + + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSFETCHOPAQUEACL); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &vp->fid); + yfs_check_req(call, bp); + + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_KERNEL); +} + +/* + * YFS.StoreOpaqueACL2 operation type + */ +static const struct afs_call_type yfs_RXYFSStoreOpaqueACL2 = { + .name = "YFS.StoreOpaqueACL2", + .op = yfs_FS_StoreOpaqueACL2, + .deliver = yfs_deliver_status_and_volsync, + .destructor = afs_flat_call_destructor, +}; + +/* + * Fetch the YFS ACL for a file. + */ +void yfs_fs_store_opaque_acl2(struct afs_operation *op) +{ + struct afs_vnode_param *vp = &op->file[0]; + struct afs_call *call; + struct afs_acl *acl = op->acl; + size_t size; + __be32 *bp; + + _enter(",%x,{%llx:%llu},,", + key_serial(op->key), vp->fid.vid, vp->fid.vnode); + + size = round_up(acl->size, 4); + call = afs_alloc_flat_call(op->net, &yfs_RXYFSStoreOpaqueACL2, + sizeof(__be32) * 2 + + sizeof(struct yfs_xdr_YFSFid) + + sizeof(__be32) + size, + sizeof(struct yfs_xdr_YFSFetchStatus) + + sizeof(struct yfs_xdr_YFSVolSync)); + if (!call) + return afs_op_nomem(op); + + /* marshall the parameters */ + bp = call->request; + bp = xdr_encode_u32(bp, YFSSTOREOPAQUEACL2); + bp = xdr_encode_u32(bp, 0); /* RPC flags */ + bp = xdr_encode_YFSFid(bp, &vp->fid); + bp = xdr_encode_u32(bp, acl->size); + memcpy(bp, acl->data, acl->size); + if (acl->size != size) + memset((void *)bp + acl->size, 0, size - acl->size); + bp += size / sizeof(__be32); yfs_check_req(call, bp); - call->cb_break = fc->cb_break; - afs_use_fs_server(call, fc->cbi); - trace_afs_make_fs_call(call, &fids[0]); - return afs_make_call(&fc->ac, call, GFP_NOFS, false); + call->fid = vp->fid; + trace_afs_make_fs_call(call, &vp->fid); + afs_make_op_call(op, call, GFP_KERNEL); } |