From 6c2e8ac0953fccdd24dc6c4b9e08e8f1cd68cf07 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Wed, 31 Dec 2008 12:54:11 -0500 Subject: netlabel: Update kernel configuration API Update the NetLabel kernel API to expose the new features added in kernel releases 2.6.25 and 2.6.28: the static/fallback label functionality and network address based selectors. Signed-off-by: Paul Moore --- net/ipv4/cipso_ipv4.c | 86 ++++++--- net/netlabel/netlabel_cipso_v4.c | 61 ++----- net/netlabel/netlabel_domainhash.c | 67 +++++++ net/netlabel/netlabel_domainhash.h | 4 + net/netlabel/netlabel_kapi.c | 347 ++++++++++++++++++++++++++++++------- net/netlabel/netlabel_unlabeled.c | 26 +-- net/netlabel/netlabel_unlabeled.h | 15 ++ 7 files changed, 469 insertions(+), 137 deletions(-) (limited to 'net') diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index e52799047a5f..6bb2635b5ded 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -449,6 +450,7 @@ static struct cipso_v4_doi *cipso_v4_doi_search(u32 doi) /** * cipso_v4_doi_add - Add a new DOI to the CIPSO protocol engine * @doi_def: the DOI structure + * @audit_info: NetLabel audit information * * Description: * The caller defines a new DOI for use by the CIPSO engine and calls this @@ -458,50 +460,78 @@ static struct cipso_v4_doi *cipso_v4_doi_search(u32 doi) * zero on success and non-zero on failure. * */ -int cipso_v4_doi_add(struct cipso_v4_doi *doi_def) +int cipso_v4_doi_add(struct cipso_v4_doi *doi_def, + struct netlbl_audit *audit_info) { + int ret_val = -EINVAL; u32 iter; + u32 doi; + u32 doi_type; + struct audit_buffer *audit_buf; + + doi = doi_def->doi; + doi_type = doi_def->type; if (doi_def == NULL || doi_def->doi == CIPSO_V4_DOI_UNKNOWN) - return -EINVAL; + goto doi_add_return; for (iter = 0; iter < CIPSO_V4_TAG_MAXCNT; iter++) { switch (doi_def->tags[iter]) { case CIPSO_V4_TAG_RBITMAP: break; case CIPSO_V4_TAG_RANGE: - if (doi_def->type != CIPSO_V4_MAP_PASS) - return -EINVAL; - break; - case CIPSO_V4_TAG_INVALID: - if (iter == 0) - return -EINVAL; - break; case CIPSO_V4_TAG_ENUM: if (doi_def->type != CIPSO_V4_MAP_PASS) - return -EINVAL; + goto doi_add_return; break; case CIPSO_V4_TAG_LOCAL: if (doi_def->type != CIPSO_V4_MAP_LOCAL) - return -EINVAL; + goto doi_add_return; + break; + case CIPSO_V4_TAG_INVALID: + if (iter == 0) + goto doi_add_return; break; default: - return -EINVAL; + goto doi_add_return; } } atomic_set(&doi_def->refcount, 1); spin_lock(&cipso_v4_doi_list_lock); - if (cipso_v4_doi_search(doi_def->doi) != NULL) - goto doi_add_failure; + if (cipso_v4_doi_search(doi_def->doi) != NULL) { + spin_unlock(&cipso_v4_doi_list_lock); + ret_val = -EEXIST; + goto doi_add_return; + } list_add_tail_rcu(&doi_def->list, &cipso_v4_doi_list); spin_unlock(&cipso_v4_doi_list_lock); + ret_val = 0; - return 0; +doi_add_return: + audit_buf = netlbl_audit_start(AUDIT_MAC_CIPSOV4_ADD, audit_info); + if (audit_buf != NULL) { + const char *type_str; + switch (doi_type) { + case CIPSO_V4_MAP_TRANS: + type_str = "trans"; + break; + case CIPSO_V4_MAP_PASS: + type_str = "pass"; + break; + case CIPSO_V4_MAP_LOCAL: + type_str = "local"; + break; + default: + type_str = "(unknown)"; + } + audit_log_format(audit_buf, + " cipso_doi=%u cipso_type=%s res=%u", + doi, type_str, ret_val == 0 ? 1 : 0); + audit_log_end(audit_buf); + } -doi_add_failure: - spin_unlock(&cipso_v4_doi_list_lock); - return -EEXIST; + return ret_val; } /** @@ -559,25 +589,39 @@ static void cipso_v4_doi_free_rcu(struct rcu_head *entry) */ int cipso_v4_doi_remove(u32 doi, struct netlbl_audit *audit_info) { + int ret_val; struct cipso_v4_doi *doi_def; + struct audit_buffer *audit_buf; spin_lock(&cipso_v4_doi_list_lock); doi_def = cipso_v4_doi_search(doi); if (doi_def == NULL) { spin_unlock(&cipso_v4_doi_list_lock); - return -ENOENT; + ret_val = -ENOENT; + goto doi_remove_return; } if (!atomic_dec_and_test(&doi_def->refcount)) { spin_unlock(&cipso_v4_doi_list_lock); - return -EBUSY; + ret_val = -EBUSY; + goto doi_remove_return; } list_del_rcu(&doi_def->list); spin_unlock(&cipso_v4_doi_list_lock); cipso_v4_cache_invalidate(); call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu); + ret_val = 0; + +doi_remove_return: + audit_buf = netlbl_audit_start(AUDIT_MAC_CIPSOV4_DEL, audit_info); + if (audit_buf != NULL) { + audit_log_format(audit_buf, + " cipso_doi=%u res=%u", + doi, ret_val == 0 ? 1 : 0); + audit_log_end(audit_buf); + } - return 0; + return ret_val; } /** diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index fff32b70efa9..bf1ab1a6790d 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -130,6 +130,7 @@ static int netlbl_cipsov4_add_common(struct genl_info *info, /** * netlbl_cipsov4_add_std - Adds a CIPSO V4 DOI definition * @info: the Generic NETLINK info block + * @audit_info: NetLabel audit information * * Description: * Create a new CIPSO_V4_MAP_TRANS DOI definition based on the given ADD @@ -137,7 +138,8 @@ static int netlbl_cipsov4_add_common(struct genl_info *info, * non-zero on error. * */ -static int netlbl_cipsov4_add_std(struct genl_info *info) +static int netlbl_cipsov4_add_std(struct genl_info *info, + struct netlbl_audit *audit_info) { int ret_val = -EINVAL; struct cipso_v4_doi *doi_def = NULL; @@ -316,7 +318,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info) } } - ret_val = cipso_v4_doi_add(doi_def); + ret_val = cipso_v4_doi_add(doi_def, audit_info); if (ret_val != 0) goto add_std_failure; return 0; @@ -330,6 +332,7 @@ add_std_failure: /** * netlbl_cipsov4_add_pass - Adds a CIPSO V4 DOI definition * @info: the Generic NETLINK info block + * @audit_info: NetLabel audit information * * Description: * Create a new CIPSO_V4_MAP_PASS DOI definition based on the given ADD message @@ -337,7 +340,8 @@ add_std_failure: * error. * */ -static int netlbl_cipsov4_add_pass(struct genl_info *info) +static int netlbl_cipsov4_add_pass(struct genl_info *info, + struct netlbl_audit *audit_info) { int ret_val; struct cipso_v4_doi *doi_def = NULL; @@ -354,7 +358,7 @@ static int netlbl_cipsov4_add_pass(struct genl_info *info) if (ret_val != 0) goto add_pass_failure; - ret_val = cipso_v4_doi_add(doi_def); + ret_val = cipso_v4_doi_add(doi_def, audit_info); if (ret_val != 0) goto add_pass_failure; return 0; @@ -367,6 +371,7 @@ add_pass_failure: /** * netlbl_cipsov4_add_local - Adds a CIPSO V4 DOI definition * @info: the Generic NETLINK info block + * @audit_info: NetLabel audit information * * Description: * Create a new CIPSO_V4_MAP_LOCAL DOI definition based on the given ADD @@ -374,7 +379,8 @@ add_pass_failure: * non-zero on error. * */ -static int netlbl_cipsov4_add_local(struct genl_info *info) +static int netlbl_cipsov4_add_local(struct genl_info *info, + struct netlbl_audit *audit_info) { int ret_val; struct cipso_v4_doi *doi_def = NULL; @@ -391,7 +397,7 @@ static int netlbl_cipsov4_add_local(struct genl_info *info) if (ret_val != 0) goto add_local_failure; - ret_val = cipso_v4_doi_add(doi_def); + ret_val = cipso_v4_doi_add(doi_def, audit_info); if (ret_val != 0) goto add_local_failure; return 0; @@ -415,48 +421,31 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info) { int ret_val = -EINVAL; - u32 type; - u32 doi; const char *type_str = "(unknown)"; - struct audit_buffer *audit_buf; struct netlbl_audit audit_info; if (!info->attrs[NLBL_CIPSOV4_A_DOI] || !info->attrs[NLBL_CIPSOV4_A_MTYPE]) return -EINVAL; - doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]); netlbl_netlink_auditinfo(skb, &audit_info); - - type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]); - switch (type) { + switch (nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE])) { case CIPSO_V4_MAP_TRANS: type_str = "trans"; - ret_val = netlbl_cipsov4_add_std(info); + ret_val = netlbl_cipsov4_add_std(info, &audit_info); break; case CIPSO_V4_MAP_PASS: type_str = "pass"; - ret_val = netlbl_cipsov4_add_pass(info); + ret_val = netlbl_cipsov4_add_pass(info, &audit_info); break; case CIPSO_V4_MAP_LOCAL: type_str = "local"; - ret_val = netlbl_cipsov4_add_local(info); + ret_val = netlbl_cipsov4_add_local(info, &audit_info); break; } if (ret_val == 0) atomic_inc(&netlabel_mgmt_protocount); - audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD, - &audit_info); - if (audit_buf != NULL) { - audit_log_format(audit_buf, - " cipso_doi=%u cipso_type=%s res=%u", - doi, - type_str, - ret_val == 0 ? 1 : 0); - audit_log_end(audit_buf); - } - return ret_val; } @@ -725,9 +714,7 @@ static int netlbl_cipsov4_remove_cb(struct netlbl_dom_map *entry, void *arg) static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info) { int ret_val = -EINVAL; - u32 doi = 0; struct netlbl_domhsh_walk_arg cb_arg; - struct audit_buffer *audit_buf; struct netlbl_audit audit_info; u32 skip_bkt = 0; u32 skip_chain = 0; @@ -735,29 +722,17 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[NLBL_CIPSOV4_A_DOI]) return -EINVAL; - doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]); netlbl_netlink_auditinfo(skb, &audit_info); - - cb_arg.doi = doi; + cb_arg.doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]); cb_arg.audit_info = &audit_info; ret_val = netlbl_domhsh_walk(&skip_bkt, &skip_chain, netlbl_cipsov4_remove_cb, &cb_arg); if (ret_val == 0 || ret_val == -ENOENT) { - ret_val = cipso_v4_doi_remove(doi, &audit_info); + ret_val = cipso_v4_doi_remove(cb_arg.doi, &audit_info); if (ret_val == 0) atomic_dec(&netlabel_mgmt_protocount); } - audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL, - &audit_info); - if (audit_buf != NULL) { - audit_log_format(audit_buf, - " cipso_doi=%u res=%u", - doi, - ret_val == 0 ? 1 : 0); - audit_log_end(audit_buf); - } - return ret_val; } diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 5fadf10e5ddf..7a10bbe02c13 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -482,6 +482,73 @@ int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry, return ret_val; } +/** + * netlbl_domhsh_remove_af4 - Removes an address selector entry + * @domain: the domain + * @addr: IPv4 address + * @mask: IPv4 address mask + * @audit_info: NetLabel audit information + * + * Description: + * Removes an individual address selector from a domain mapping and potentially + * the entire mapping if it is empty. Returns zero on success, negative values + * on failure. + * + */ +int netlbl_domhsh_remove_af4(const char *domain, + const struct in_addr *addr, + const struct in_addr *mask, + struct netlbl_audit *audit_info) +{ + struct netlbl_dom_map *entry_map; + struct netlbl_af4list *entry_addr; + struct netlbl_af4list *iter4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; +#endif /* IPv6 */ + struct netlbl_domaddr4_map *entry; + + rcu_read_lock(); + + if (domain) + entry_map = netlbl_domhsh_search(domain); + else + entry_map = netlbl_domhsh_search_def(domain); + if (entry_map == NULL || entry_map->type != NETLBL_NLTYPE_ADDRSELECT) + goto remove_af4_failure; + + spin_lock(&netlbl_domhsh_lock); + entry_addr = netlbl_af4list_remove(addr->s_addr, mask->s_addr, + &entry_map->type_def.addrsel->list4); + spin_unlock(&netlbl_domhsh_lock); + + if (entry_addr == NULL) + goto remove_af4_failure; + netlbl_af4list_foreach_rcu(iter4, &entry_map->type_def.addrsel->list4) + goto remove_af4_single_addr; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(iter6, &entry_map->type_def.addrsel->list6) + goto remove_af4_single_addr; +#endif /* IPv6 */ + /* the domain mapping is empty so remove it from the mapping table */ + netlbl_domhsh_remove_entry(entry_map, audit_info); + +remove_af4_single_addr: + rcu_read_unlock(); + /* yick, we can't use call_rcu here because we don't have a rcu head + * pointer but hopefully this should be a rare case so the pause + * shouldn't be a problem */ + synchronize_rcu(); + entry = netlbl_domhsh_addr4_entry(entry_addr); + cipso_v4_doi_putdef(entry->type_def.cipsov4); + kfree(entry); + return 0; + +remove_af4_failure: + rcu_read_unlock(); + return -ENOENT; +} + /** * netlbl_domhsh_remove - Removes an entry from the domain hash table * @domain: the domain to remove diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h index bfcb6763a1a1..0261dda3f2d2 100644 --- a/net/netlabel/netlabel_domainhash.h +++ b/net/netlabel/netlabel_domainhash.h @@ -90,6 +90,10 @@ int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, struct netlbl_audit *audit_info); int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry, struct netlbl_audit *audit_info); +int netlbl_domhsh_remove_af4(const char *domain, + const struct in_addr *addr, + const struct in_addr *mask, + struct netlbl_audit *audit_info); int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info); int netlbl_domhsh_remove_default(struct netlbl_audit *audit_info); struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain); diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index b32eceb3ab0d..fd9229db075c 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -31,7 +31,10 @@ #include #include #include +#include +#include #include +#include #include #include #include @@ -42,6 +45,7 @@ #include "netlabel_cipso_v4.h" #include "netlabel_user.h" #include "netlabel_mgmt.h" +#include "netlabel_addrlist.h" /* * Configuration Functions @@ -50,6 +54,9 @@ /** * netlbl_cfg_map_del - Remove a NetLabel/LSM domain mapping * @domain: the domain mapping to remove + * @family: address family + * @addr: IP address + * @mask: IP address mask * @audit_info: NetLabel audit information * * Description: @@ -58,14 +65,32 @@ * values on failure. * */ -int netlbl_cfg_map_del(const char *domain, struct netlbl_audit *audit_info) +int netlbl_cfg_map_del(const char *domain, + u16 family, + const void *addr, + const void *mask, + struct netlbl_audit *audit_info) { - return netlbl_domhsh_remove(domain, audit_info); + if (addr == NULL && mask == NULL) { + return netlbl_domhsh_remove(domain, audit_info); + } else if (addr != NULL && mask != NULL) { + switch (family) { + case AF_INET: + return netlbl_domhsh_remove_af4(domain, addr, mask, + audit_info); + default: + return -EPFNOSUPPORT; + } + } else + return -EINVAL; } /** - * netlbl_cfg_unlbl_add_map - Add an unlabeled NetLabel/LSM domain mapping + * netlbl_cfg_unlbl_map_add - Add a new unlabeled mapping * @domain: the domain mapping to add + * @family: address family + * @addr: IP address + * @mask: IP address mask * @audit_info: NetLabel audit information * * Description: @@ -74,11 +99,19 @@ int netlbl_cfg_map_del(const char *domain, struct netlbl_audit *audit_info) * negative values on failure. * */ -int netlbl_cfg_unlbl_add_map(const char *domain, +int netlbl_cfg_unlbl_map_add(const char *domain, + u16 family, + const void *addr, + const void *mask, struct netlbl_audit *audit_info) { int ret_val = -ENOMEM; struct netlbl_dom_map *entry; + struct netlbl_domaddr_map *addrmap = NULL; + struct netlbl_domaddr4_map *map4 = NULL; + struct netlbl_domaddr6_map *map6 = NULL; + const struct in_addr *addr4, *mask4; + const struct in6_addr *addr6, *mask6; entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) @@ -86,49 +119,225 @@ int netlbl_cfg_unlbl_add_map(const char *domain, if (domain != NULL) { entry->domain = kstrdup(domain, GFP_ATOMIC); if (entry->domain == NULL) - goto cfg_unlbl_add_map_failure; + goto cfg_unlbl_map_add_failure; + } + + if (addr == NULL && mask == NULL) + entry->type = NETLBL_NLTYPE_UNLABELED; + else if (addr != NULL && mask != NULL) { + addrmap = kzalloc(sizeof(*addrmap), GFP_ATOMIC); + if (addrmap == NULL) + goto cfg_unlbl_map_add_failure; + INIT_LIST_HEAD(&addrmap->list4); + INIT_LIST_HEAD(&addrmap->list6); + + switch (family) { + case AF_INET: + addr4 = addr; + mask4 = mask; + map4 = kzalloc(sizeof(*map4), GFP_ATOMIC); + if (map4 == NULL) + goto cfg_unlbl_map_add_failure; + map4->type = NETLBL_NLTYPE_UNLABELED; + map4->list.addr = addr4->s_addr & mask4->s_addr; + map4->list.mask = mask4->s_addr; + map4->list.valid = 1; + ret_val = netlbl_af4list_add(&map4->list, + &addrmap->list4); + if (ret_val != 0) + goto cfg_unlbl_map_add_failure; + break; + case AF_INET6: + addr6 = addr; + mask6 = mask; + map6 = kzalloc(sizeof(*map6), GFP_ATOMIC); + if (map4 == NULL) + goto cfg_unlbl_map_add_failure; + map6->type = NETLBL_NLTYPE_UNLABELED; + ipv6_addr_copy(&map6->list.addr, addr6); + map6->list.addr.s6_addr32[0] &= mask6->s6_addr32[0]; + map6->list.addr.s6_addr32[1] &= mask6->s6_addr32[1]; + map6->list.addr.s6_addr32[2] &= mask6->s6_addr32[2]; + map6->list.addr.s6_addr32[3] &= mask6->s6_addr32[3]; + ipv6_addr_copy(&map6->list.mask, mask6); + map6->list.valid = 1; + ret_val = netlbl_af4list_add(&map4->list, + &addrmap->list4); + if (ret_val != 0) + goto cfg_unlbl_map_add_failure; + break; + default: + goto cfg_unlbl_map_add_failure; + break; + } + + entry->type_def.addrsel = addrmap; + entry->type = NETLBL_NLTYPE_ADDRSELECT; + } else { + ret_val = -EINVAL; + goto cfg_unlbl_map_add_failure; } - entry->type = NETLBL_NLTYPE_UNLABELED; ret_val = netlbl_domhsh_add(entry, audit_info); if (ret_val != 0) - goto cfg_unlbl_add_map_failure; + goto cfg_unlbl_map_add_failure; return 0; -cfg_unlbl_add_map_failure: +cfg_unlbl_map_add_failure: if (entry != NULL) kfree(entry->domain); kfree(entry); + kfree(addrmap); + kfree(map4); + kfree(map6); return ret_val; } + +/** + * netlbl_cfg_unlbl_static_add - Adds a new static label + * @net: network namespace + * @dev_name: interface name + * @addr: IP address in network byte order (struct in[6]_addr) + * @mask: address mask in network byte order (struct in[6]_addr) + * @family: address family + * @secid: LSM secid value for the entry + * @audit_info: NetLabel audit information + * + * Description: + * Adds a new NetLabel static label to be used when protocol provided labels + * are not present on incoming traffic. If @dev_name is NULL then the default + * interface will be used. Returns zero on success, negative values on failure. + * + */ +int netlbl_cfg_unlbl_static_add(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u16 family, + u32 secid, + struct netlbl_audit *audit_info) +{ + u32 addr_len; + + switch (family) { + case AF_INET: + addr_len = sizeof(struct in_addr); + break; + case AF_INET6: + addr_len = sizeof(struct in6_addr); + break; + default: + return -EPFNOSUPPORT; + } + + return netlbl_unlhsh_add(net, + dev_name, addr, mask, addr_len, + secid, audit_info); +} + +/** + * netlbl_cfg_unlbl_static_del - Removes an existing static label + * @net: network namespace + * @dev_name: interface name + * @addr: IP address in network byte order (struct in[6]_addr) + * @mask: address mask in network byte order (struct in[6]_addr) + * @family: address family + * @secid: LSM secid value for the entry + * @audit_info: NetLabel audit information + * + * Description: + * Removes an existing NetLabel static label used when protocol provided labels + * are not present on incoming traffic. If @dev_name is NULL then the default + * interface will be used. Returns zero on success, negative values on failure. + * + */ +int netlbl_cfg_unlbl_static_del(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u16 family, + struct netlbl_audit *audit_info) +{ + u32 addr_len; + + switch (family) { + case AF_INET: + addr_len = sizeof(struct in_addr); + break; + case AF_INET6: + addr_len = sizeof(struct in6_addr); + break; + default: + return -EPFNOSUPPORT; + } + + return netlbl_unlhsh_remove(net, + dev_name, addr, mask, addr_len, + audit_info); +} + +/** + * netlbl_cfg_cipsov4_add - Add a new CIPSOv4 DOI definition + * @doi_def: CIPSO DOI definition + * @audit_info: NetLabel audit information + * + * Description: + * Add a new CIPSO DOI definition as defined by @doi_def. Returns zero on + * success and negative values on failure. + * + */ +int netlbl_cfg_cipsov4_add(struct cipso_v4_doi *doi_def, + struct netlbl_audit *audit_info) +{ + return cipso_v4_doi_add(doi_def, audit_info); +} + +/** + * netlbl_cfg_cipsov4_del - Remove an existing CIPSOv4 DOI definition + * @doi: CIPSO DOI + * @audit_info: NetLabel audit information + * + * Description: + * Remove an existing CIPSO DOI definition matching @doi. Returns zero on + * success and negative values on failure. + * + */ +void netlbl_cfg_cipsov4_del(u32 doi, struct netlbl_audit *audit_info) +{ + cipso_v4_doi_remove(doi, audit_info); +} + /** - * netlbl_cfg_cipsov4_add_map - Add a new CIPSOv4 DOI definition and mapping - * @doi_def: the DOI definition + * netlbl_cfg_cipsov4_map_add - Add a new CIPSOv4 DOI mapping + * @doi: the CIPSO DOI * @domain: the domain mapping to add + * @addr: IP address + * @mask: IP address mask * @audit_info: NetLabel audit information * * Description: - * Add a new CIPSOv4 DOI definition and NetLabel/LSM domain mapping for this - * new DOI definition to the NetLabel subsystem. A @domain value of NULL adds - * a new default domain mapping. Returns zero on success, negative values on - * failure. + * Add a new NetLabel/LSM domain mapping for the given CIPSO DOI to the NetLabel + * subsystem. A @domain value of NULL adds a new default domain mapping. + * Returns zero on success, negative values on failure. * */ -int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def, +int netlbl_cfg_cipsov4_map_add(u32 doi, const char *domain, + const struct in_addr *addr, + const struct in_addr *mask, struct netlbl_audit *audit_info) { int ret_val = -ENOMEM; - u32 doi; - u32 doi_type; + struct cipso_v4_doi *doi_def; struct netlbl_dom_map *entry; - const char *type_str; - struct audit_buffer *audit_buf; + struct netlbl_domaddr_map *addrmap = NULL; + struct netlbl_domaddr4_map *addrinfo = NULL; - doi = doi_def->doi; - doi_type = doi_def->type; + doi_def = cipso_v4_doi_getdef(doi); + if (doi_def == NULL) + return -ENOENT; entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) @@ -136,56 +345,52 @@ int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def, if (domain != NULL) { entry->domain = kstrdup(domain, GFP_ATOMIC); if (entry->domain == NULL) - goto cfg_cipsov4_add_map_failure; + goto cfg_cipsov4_map_add_failure; } - ret_val = cipso_v4_doi_add(doi_def); - if (ret_val != 0) - goto cfg_cipsov4_add_map_failure_remove_doi; - entry->type = NETLBL_NLTYPE_CIPSOV4; - entry->type_def.cipsov4 = cipso_v4_doi_getdef(doi); - if (entry->type_def.cipsov4 == NULL) { - ret_val = -ENOENT; - goto cfg_cipsov4_add_map_failure_remove_doi; + if (addr == NULL && mask == NULL) { + entry->type_def.cipsov4 = doi_def; + entry->type = NETLBL_NLTYPE_CIPSOV4; + } else if (addr != NULL && mask != NULL) { + addrmap = kzalloc(sizeof(*addrmap), GFP_ATOMIC); + if (addrmap == NULL) + goto cfg_cipsov4_map_add_failure; + INIT_LIST_HEAD(&addrmap->list4); + INIT_LIST_HEAD(&addrmap->list6); + + addrinfo = kzalloc(sizeof(*addrinfo), GFP_ATOMIC); + if (addrinfo == NULL) + goto cfg_cipsov4_map_add_failure; + addrinfo->type_def.cipsov4 = doi_def; + addrinfo->type = NETLBL_NLTYPE_CIPSOV4; + addrinfo->list.addr = addr->s_addr & mask->s_addr; + addrinfo->list.mask = mask->s_addr; + addrinfo->list.valid = 1; + ret_val = netlbl_af4list_add(&addrinfo->list, &addrmap->list4); + if (ret_val != 0) + goto cfg_cipsov4_map_add_failure; + + entry->type_def.addrsel = addrmap; + entry->type = NETLBL_NLTYPE_ADDRSELECT; + } else { + ret_val = -EINVAL; + goto cfg_cipsov4_map_add_failure; } + ret_val = netlbl_domhsh_add(entry, audit_info); if (ret_val != 0) - goto cfg_cipsov4_add_map_failure_release_doi; - -cfg_cipsov4_add_map_return: - audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD, - audit_info); - if (audit_buf != NULL) { - switch (doi_type) { - case CIPSO_V4_MAP_TRANS: - type_str = "trans"; - break; - case CIPSO_V4_MAP_PASS: - type_str = "pass"; - break; - case CIPSO_V4_MAP_LOCAL: - type_str = "local"; - break; - default: - type_str = "(unknown)"; - } - audit_log_format(audit_buf, - " cipso_doi=%u cipso_type=%s res=%u", - doi, type_str, ret_val == 0 ? 1 : 0); - audit_log_end(audit_buf); - } + goto cfg_cipsov4_map_add_failure; - return ret_val; + return 0; -cfg_cipsov4_add_map_failure_release_doi: +cfg_cipsov4_map_add_failure: cipso_v4_doi_putdef(doi_def); -cfg_cipsov4_add_map_failure_remove_doi: - cipso_v4_doi_remove(doi, audit_info); -cfg_cipsov4_add_map_failure: if (entry != NULL) kfree(entry->domain); kfree(entry); - goto cfg_cipsov4_add_map_return; + kfree(addrmap); + kfree(addrinfo); + return ret_val; } /* @@ -690,6 +895,28 @@ int netlbl_cache_add(const struct sk_buff *skb, return -ENOMSG; } +/* + * Protocol Engine Functions + */ + +/** + * netlbl_audit_start - Start an audit message + * @type: audit message type + * @audit_info: NetLabel audit information + * + * Description: + * Start an audit message using the type specified in @type and fill the audit + * message with some fields common to all NetLabel audit messages. This + * function should only be used by protocol engines, not LSMs. Returns a + * pointer to the audit buffer on success, NULL on failure. + * + */ +struct audit_buffer *netlbl_audit_start(int type, + struct netlbl_audit *audit_info) +{ + return netlbl_audit_start_common(type, audit_info); +} + /* * Setup Functions */ diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 8c0308032178..f3c5c68c6848 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -450,13 +450,13 @@ add_iface_failure: * success, negative values on failure. * */ -static int netlbl_unlhsh_add(struct net *net, - const char *dev_name, - const void *addr, - const void *mask, - u32 addr_len, - u32 secid, - struct netlbl_audit *audit_info) +int netlbl_unlhsh_add(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u32 addr_len, + u32 secid, + struct netlbl_audit *audit_info) { int ret_val; int ifindex; @@ -720,12 +720,12 @@ unlhsh_condremove_failure: * Returns zero on success, negative values on failure. * */ -static int netlbl_unlhsh_remove(struct net *net, - const char *dev_name, - const void *addr, - const void *mask, - u32 addr_len, - struct netlbl_audit *audit_info) +int netlbl_unlhsh_remove(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u32 addr_len, + struct netlbl_audit *audit_info) { int ret_val; struct net_device *dev; diff --git a/net/netlabel/netlabel_unlabeled.h b/net/netlabel/netlabel_unlabeled.h index 06b1301ac072..7aba63595137 100644 --- a/net/netlabel/netlabel_unlabeled.h +++ b/net/netlabel/netlabel_unlabeled.h @@ -221,6 +221,21 @@ int netlbl_unlabel_genl_init(void); /* General Unlabeled init function */ int netlbl_unlabel_init(u32 size); +/* Static/Fallback label management functions */ +int netlbl_unlhsh_add(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u32 addr_len, + u32 secid, + struct netlbl_audit *audit_info); +int netlbl_unlhsh_remove(struct net *net, + const char *dev_name, + const void *addr, + const void *mask, + u32 addr_len, + struct netlbl_audit *audit_info); + /* Process Unlabeled incoming network packets */ int netlbl_unlabel_getattr(const struct sk_buff *skb, u16 family, -- cgit v1.2.3 From f3298dc4f2277874d40cb4fc3a6e277317d6603b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 10 Dec 2008 03:16:51 -0500 Subject: sanitize audit_socketcall * don't bother with allocations * now that it can't fail, make it return void Signed-off-by: Al Viro --- include/linux/audit.h | 4 ++-- kernel/auditsc.c | 66 +++++++++++++++++++++++++++++---------------------- net/socket.c | 4 +--- 3 files changed, 41 insertions(+), 33 deletions(-) (limited to 'net') diff --git a/include/linux/audit.h b/include/linux/audit.h index 26c4f6f65a46..466a953d4bf6 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -446,7 +446,7 @@ extern void audit_log_task_context(struct audit_buffer *ab); extern int __audit_ipc_obj(struct kern_ipc_perm *ipcp); extern int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); extern int audit_bprm(struct linux_binprm *bprm); -extern int audit_socketcall(int nargs, unsigned long *args); +extern void audit_socketcall(int nargs, unsigned long *args); extern int audit_sockaddr(int len, void *addr); extern int __audit_fd_pair(int fd1, int fd2); extern int audit_set_macxattr(const char *name); @@ -549,7 +549,7 @@ extern int audit_signals; #define audit_ipc_obj(i) ({ 0; }) #define audit_ipc_set_perm(q,u,g,m) ({ 0; }) #define audit_bprm(p) ({ 0; }) -#define audit_socketcall(n,a) ({ 0; }) +#define audit_socketcall(n,a) ((void)0) #define audit_fd_pair(n,a) ({ 0; }) #define audit_sockaddr(len, addr) ({ 0; }) #define audit_set_macxattr(n) do { ; } while (0) diff --git a/kernel/auditsc.c b/kernel/auditsc.c index c2e43ebb1b68..5cda66466e14 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -168,12 +168,6 @@ struct audit_aux_data_execve { struct mm_struct *mm; }; -struct audit_aux_data_socketcall { - struct audit_aux_data d; - int nargs; - unsigned long args[0]; -}; - struct audit_aux_data_fd_pair { struct audit_aux_data d; int fd[2]; @@ -247,6 +241,14 @@ struct audit_context { struct audit_tree_refs *trees, *first_trees; int tree_count; + int type; + union { + struct { + int nargs; + long args[6]; + } socketcall; + }; + #if AUDIT_DEBUG int put_count; int ino_count; @@ -1226,6 +1228,27 @@ static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name) audit_log_format(ab, " cap_fe=%d cap_fver=%x", name->fcap.fE, name->fcap_ver); } +static void show_special(struct audit_context *context) +{ + struct audit_buffer *ab; + int i; + + ab = audit_log_start(context, GFP_KERNEL, context->type); + if (!ab) + return; + + switch (context->type) { + case AUDIT_SOCKETCALL: { + int nargs = context->socketcall.nargs; + audit_log_format(ab, "nargs=%d", nargs); + for (i = 0; i < nargs; i++) + audit_log_format(ab, " a%d=%lx", i, + context->socketcall.args[i]); + break; } + } + audit_log_end(ab); +} + static void audit_log_exit(struct audit_context *context, struct task_struct *tsk) { const struct cred *cred; @@ -1372,13 +1395,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_execve_info(context, &ab, axi); break; } - case AUDIT_SOCKETCALL: { - struct audit_aux_data_socketcall *axs = (void *)aux; - audit_log_format(ab, "nargs=%d", axs->nargs); - for (i=0; inargs; i++) - audit_log_format(ab, " a%d=%lx", i, axs->args[i]); - break; } - case AUDIT_FD_PAIR: { struct audit_aux_data_fd_pair *axs = (void *)aux; audit_log_format(ab, "fd0=%d fd1=%d", axs->fd[0], axs->fd[1]); @@ -1410,6 +1426,9 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_end(ab); } + if (context->type) + show_special(context); + if (context->sockaddr_len) { ab = audit_log_start(context, GFP_KERNEL, AUDIT_SOCKADDR); if (ab) { @@ -1689,6 +1708,7 @@ void audit_syscall_exit(int valid, long return_code) context->target_pid = 0; context->target_sid = 0; context->sockaddr_len = 0; + context->type = 0; kfree(context->filterkey); context->filterkey = NULL; tsk->audit_context = context; @@ -2406,27 +2426,17 @@ int audit_bprm(struct linux_binprm *bprm) * @nargs: number of args * @args: args array * - * Returns 0 for success or NULL context or < 0 on error. */ -int audit_socketcall(int nargs, unsigned long *args) +void audit_socketcall(int nargs, unsigned long *args) { - struct audit_aux_data_socketcall *ax; struct audit_context *context = current->audit_context; if (likely(!context || context->dummy)) - return 0; - - ax = kmalloc(sizeof(*ax) + nargs * sizeof(unsigned long), GFP_KERNEL); - if (!ax) - return -ENOMEM; - - ax->nargs = nargs; - memcpy(ax->args, args, nargs * sizeof(unsigned long)); + return; - ax->d.type = AUDIT_SOCKETCALL; - ax->d.next = context->aux; - context->aux = (void *)ax; - return 0; + context->type = AUDIT_SOCKETCALL; + context->socketcall.nargs = nargs; + memcpy(context->socketcall.args, args, nargs * sizeof(unsigned long)); } /** diff --git a/net/socket.c b/net/socket.c index 2c730fc718ab..b41a92093e40 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2065,9 +2065,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) if (copy_from_user(a, args, nargs[call])) return -EFAULT; - err = audit_socketcall(nargs[call] / sizeof(unsigned long), a); - if (err) - return err; + audit_socketcall(nargs[call] / sizeof(unsigned long), a); a0 = a[0]; a1 = a[1]; -- cgit v1.2.3 From 157cf649a735a2f7e8dba0ed08e6e38b6c30d886 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 14 Dec 2008 04:57:47 -0500 Subject: sanitize audit_fd_pair() * no allocations * return void Signed-off-by: Al Viro --- fs/pipe.c | 7 +------ include/linux/audit.h | 9 ++++----- kernel/auditsc.c | 44 ++++++++++++++------------------------------ net/socket.c | 9 +-------- 4 files changed, 20 insertions(+), 49 deletions(-) (limited to 'net') diff --git a/fs/pipe.c b/fs/pipe.c index aaf797bd57b9..891697112f66 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -1016,10 +1016,7 @@ int do_pipe_flags(int *fd, int flags) goto err_fdr; fdw = error; - error = audit_fd_pair(fdr, fdw); - if (error < 0) - goto err_fdw; - + audit_fd_pair(fdr, fdw); fd_install(fdr, fr); fd_install(fdw, fw); fd[0] = fdr; @@ -1027,8 +1024,6 @@ int do_pipe_flags(int *fd, int flags) return 0; - err_fdw: - put_unused_fd(fdw); err_fdr: put_unused_fd(fdr); err_read_pipe: diff --git a/include/linux/audit.h b/include/linux/audit.h index 54978bdd2bd4..bd59cd1e3219 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -448,7 +448,7 @@ extern void __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mod extern int audit_bprm(struct linux_binprm *bprm); extern void audit_socketcall(int nargs, unsigned long *args); extern int audit_sockaddr(int len, void *addr); -extern int __audit_fd_pair(int fd1, int fd2); +extern void __audit_fd_pair(int fd1, int fd2); extern int audit_set_macxattr(const char *name); extern void __audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr); extern void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout); @@ -464,11 +464,10 @@ static inline void audit_ipc_obj(struct kern_ipc_perm *ipcp) if (unlikely(!audit_dummy_context())) __audit_ipc_obj(ipcp); } -static inline int audit_fd_pair(int fd1, int fd2) +static inline void audit_fd_pair(int fd1, int fd2) { if (unlikely(!audit_dummy_context())) - return __audit_fd_pair(fd1, fd2); - return 0; + __audit_fd_pair(fd1, fd2); } static inline void audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode) { @@ -537,7 +536,7 @@ extern int audit_signals; #define audit_ipc_set_perm(q,u,g,m) ((void)0) #define audit_bprm(p) ({ 0; }) #define audit_socketcall(n,a) ((void)0) -#define audit_fd_pair(n,a) ({ 0; }) +#define audit_fd_pair(n,a) ((void)0) #define audit_sockaddr(len, addr) ({ 0; }) #define audit_set_macxattr(n) do { ; } while (0) #define audit_mq_open(o,m,a) ((void)0) diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 83e946f1cdde..327e65d50674 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -131,11 +131,6 @@ struct audit_aux_data_execve { struct mm_struct *mm; }; -struct audit_aux_data_fd_pair { - struct audit_aux_data d; - int fd[2]; -}; - struct audit_aux_data_pids { struct audit_aux_data d; pid_t target_pid[AUDIT_AUX_PIDS]; @@ -241,6 +236,7 @@ struct audit_context { struct mq_attr attr; } mq_open; }; + int fds[2]; #if AUDIT_DEBUG int put_count; @@ -1382,11 +1378,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_execve_info(context, &ab, axi); break; } - case AUDIT_FD_PAIR: { - struct audit_aux_data_fd_pair *axs = (void *)aux; - audit_log_format(ab, "fd0=%d fd1=%d", axs->fd[0], axs->fd[1]); - break; } - case AUDIT_BPRM_FCAPS: { struct audit_aux_data_bprm_fcaps *axs = (void *)aux; audit_log_format(ab, "fver=%x", axs->fcap_ver); @@ -1416,6 +1407,15 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts if (context->type) show_special(context, &call_panic); + if (context->fds[0] >= 0) { + ab = audit_log_start(context, GFP_KERNEL, AUDIT_FD_PAIR); + if (ab) { + audit_log_format(ab, "fd0=%d fd1=%d", + context->fds[0], context->fds[1]); + audit_log_end(ab); + } + } + if (context->sockaddr_len) { ab = audit_log_start(context, GFP_KERNEL, AUDIT_SOCKADDR); if (ab) { @@ -1696,6 +1696,7 @@ void audit_syscall_exit(int valid, long return_code) context->target_sid = 0; context->sockaddr_len = 0; context->type = 0; + context->fds[0] = -1; kfree(context->filterkey); context->filterkey = NULL; tsk->audit_context = context; @@ -2291,29 +2292,12 @@ void audit_socketcall(int nargs, unsigned long *args) * @fd1: the first file descriptor * @fd2: the second file descriptor * - * Returns 0 for success or NULL context or < 0 on error. */ -int __audit_fd_pair(int fd1, int fd2) +void __audit_fd_pair(int fd1, int fd2) { struct audit_context *context = current->audit_context; - struct audit_aux_data_fd_pair *ax; - - if (likely(!context)) { - return 0; - } - - ax = kmalloc(sizeof(*ax), GFP_KERNEL); - if (!ax) { - return -ENOMEM; - } - - ax->fd[0] = fd1; - ax->fd[1] = fd2; - - ax->d.type = AUDIT_FD_PAIR; - ax->d.next = context->aux; - context->aux = (void *)ax; - return 0; + context->fds[0] = fd1; + context->fds[1] = fd2; } /** diff --git a/net/socket.c b/net/socket.c index b41a92093e40..06603d73c411 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1313,13 +1313,7 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, goto out_fd1; } - err = audit_fd_pair(fd1, fd2); - if (err < 0) { - fput(newfile1); - fput(newfile2); - goto out_fd; - } - + audit_fd_pair(fd1, fd2); fd_install(fd1, newfile1); fd_install(fd2, newfile2); /* fd1 and fd2 may be already another descriptors. @@ -1349,7 +1343,6 @@ out_fd2: out_fd1: put_filp(newfile2); sock_release(sock2); -out_fd: put_unused_fd(fd1); put_unused_fd(fd2); goto out; -- cgit v1.2.3 From 14deae41566b5cdd992c01d0069518ced5227c83 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 4 Jan 2009 16:04:39 -0800 Subject: ipv6: Fix sporadic sendmsg -EINVAL when sending to multicast groups. Thanks to excellent diagnosis by Eduard Guzovsky. The core problem is that on a network with lots of active multicast traffic, the neighbour cache can fill up. If we try to allocate a new route and thus neighbour cache entry, the bog-standard GC attempt the neighbour layer does in ineffective because route entries hold a reference to the existing neighbour entries and GC can only liberate entries with no references. IPV4 already has a way to handle this, by doing a route cache GC in such situations (when neigh attach returns -ENOBUFS). So simply mimick this on the ipv6 side. Tested-by: Eduard Guzovsky Signed-off-by: David S. Miller --- include/net/ndisc.h | 4 ++-- net/ipv6/route.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 49 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/ndisc.h b/include/net/ndisc.h index ce532f2222ce..1459ed3e2697 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -155,9 +155,9 @@ static inline struct neighbour * ndisc_get_neigh(struct net_device *dev, const s { if (dev) - return __neigh_lookup(&nd_tbl, addr, dev, 1); + return __neigh_lookup_errno(&nd_tbl, addr, dev); - return NULL; + return ERR_PTR(-ENODEV); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 18c486cf4987..76f06b94ab9f 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -627,6 +627,9 @@ static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *dad rt = ip6_rt_copy(ort); if (rt) { + struct neighbour *neigh; + int attempts = !in_softirq(); + if (!(rt->rt6i_flags&RTF_GATEWAY)) { if (rt->rt6i_dst.plen != 128 && ipv6_addr_equal(&rt->rt6i_dst.addr, daddr)) @@ -646,7 +649,35 @@ static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *dad } #endif - rt->rt6i_nexthop = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway); + retry: + neigh = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway); + if (IS_ERR(neigh)) { + struct net *net = dev_net(rt->rt6i_dev); + int saved_rt_min_interval = + net->ipv6.sysctl.ip6_rt_gc_min_interval; + int saved_rt_elasticity = + net->ipv6.sysctl.ip6_rt_gc_elasticity; + + if (attempts-- > 0) { + net->ipv6.sysctl.ip6_rt_gc_elasticity = 1; + net->ipv6.sysctl.ip6_rt_gc_min_interval = 0; + + ip6_dst_gc(net->ipv6.ip6_dst_ops); + + net->ipv6.sysctl.ip6_rt_gc_elasticity = + saved_rt_elasticity; + net->ipv6.sysctl.ip6_rt_gc_min_interval = + saved_rt_min_interval; + goto retry; + } + + if (net_ratelimit()) + printk(KERN_WARNING + "Neighbour table overflow.\n"); + dst_free(&rt->u.dst); + return NULL; + } + rt->rt6i_nexthop = neigh; } @@ -945,8 +976,11 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, dev_hold(dev); if (neigh) neigh_hold(neigh); - else + else { neigh = ndisc_get_neigh(dev, addr); + if (IS_ERR(neigh)) + neigh = NULL; + } rt->rt6i_dev = dev; rt->rt6i_idev = idev; @@ -1887,6 +1921,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, { struct net *net = dev_net(idev->dev); struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); + struct neighbour *neigh; if (rt == NULL) return ERR_PTR(-ENOMEM); @@ -1909,11 +1944,18 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, rt->rt6i_flags |= RTF_ANYCAST; else rt->rt6i_flags |= RTF_LOCAL; - rt->rt6i_nexthop = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway); - if (rt->rt6i_nexthop == NULL) { + neigh = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway); + if (IS_ERR(neigh)) { dst_free(&rt->u.dst); - return ERR_PTR(-ENOMEM); + + /* We are casting this because that is the return + * value type. But an errno encoded pointer is the + * same regardless of the underlying pointer type, + * and that's what we are returning. So this is OK. + */ + return (struct rt6_info *) neigh; } + rt->rt6i_nexthop = neigh; ipv6_addr_copy(&rt->rt6i_dst.addr, addr); rt->rt6i_dst.plen = 128; -- cgit v1.2.3 From b530256d2e0f1a75fab31f9821129fff1bb49faa Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sun, 4 Jan 2009 16:13:19 -0800 Subject: gro: Use gso_size to store MSS In order to allow GRO packets without frag_list at all, we need to store the MSS in the packet itself. The obvious place is gso_size. The only thing to watch out for is if the packet ends up not being GRO then we need to clear gso_size before pushing the packet into the stack. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/dev.c | 2 ++ net/core/skbuff.c | 1 + net/ipv4/tcp.c | 5 +---- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 09c66a449da6..1e1a68066457 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2365,6 +2365,7 @@ static int napi_gro_complete(struct sk_buff *skb) } out: + skb_shinfo(skb)->gso_size = 0; __skb_push(skb, -skb_network_offset(skb)); return netif_receive_skb(skb); } @@ -2446,6 +2447,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) } NAPI_GRO_CB(skb)->count = 1; + skb_shinfo(skb)->gso_size = skb->len; skb->next = napi->gro_list; napi->gro_list = skb; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b8d0abb26433..3aafb10325b8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2613,6 +2613,7 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) *NAPI_GRO_CB(nskb) = *NAPI_GRO_CB(p); skb_shinfo(nskb)->frag_list = p; + skb_shinfo(nskb)->gso_size = skb_shinfo(p)->gso_size; skb_header_release(p); nskb->prev = p; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f28acf11fc67..4d655e945413 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2519,9 +2519,7 @@ found: flush |= memcmp(th + 1, th2 + 1, thlen - sizeof(*th)); total = p->len; - mss = total; - if (skb_shinfo(p)->frag_list) - mss = skb_shinfo(p)->frag_list->len; + mss = skb_shinfo(p)->gso_size; flush |= skb->len > mss || skb->len <= 0; flush |= ntohl(th2->seq) + total != ntohl(th->seq); @@ -2557,7 +2555,6 @@ int tcp_gro_complete(struct sk_buff *skb) skb->csum_offset = offsetof(struct tcphdr, check); skb->ip_summed = CHECKSUM_PARTIAL; - skb_shinfo(skb)->gso_size = skb_shinfo(skb)->frag_list->len; skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count; if (th->cwr) -- cgit v1.2.3 From 5d38a079ce3971f932bbdc0dc5b887806fabd5dc Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sun, 4 Jan 2009 16:13:40 -0800 Subject: gro: Add page frag support This patch allows GRO to merge page frags (skb_shinfo(skb)->frags) in one skb, rather than using the less efficient frag_list. It also adds a new interface, napi_gro_frags to allow drivers to inject page frags directly into the stack without allocating an skb. This is intended to be the GRO equivalent for LRO's lro_receive_frags interface. The existing GSO interface can already handle page frags with or without an appended frag_list so nothing needs to be changed there. The merging itself is rather simple. We store any new frag entries after the last existing entry, without checking whether the first new entry can be merged with the last existing entry. Making this check would actually be easy but since no existing driver can produce contiguous frags anyway it would just be mental masturbation. If the total number of entries would exceed the capacity of a single skb, we simply resort to using frag_list as we do now. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 16 ++++++++- net/core/dev.c | 91 ++++++++++++++++++++++++++++++++++++++++++++--- net/core/skbuff.c | 14 +++++++- 3 files changed, 114 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 41e1224651cf..c28bbba3c23d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -313,10 +313,11 @@ struct napi_struct { #ifdef CONFIG_NETPOLL spinlock_t poll_lock; int poll_owner; - struct net_device *dev; #endif + struct net_device *dev; struct list_head dev_list; struct sk_buff *gro_list; + struct sk_buff *skb; }; enum @@ -990,6 +991,9 @@ struct napi_gro_cb { /* Number of segments aggregated. */ int count; + + /* Free the skb? */ + int free; }; #define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb) @@ -1011,6 +1015,14 @@ struct packet_type { struct list_head list; }; +struct napi_gro_fraginfo { + skb_frag_t frags[MAX_SKB_FRAGS]; + unsigned int nr_frags; + unsigned int ip_summed; + unsigned int len; + __wsum csum; +}; + #include #include @@ -1363,6 +1375,8 @@ extern int netif_receive_skb(struct sk_buff *skb); extern void napi_gro_flush(struct napi_struct *napi); extern int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); +extern int napi_gro_frags(struct napi_struct *napi, + struct napi_gro_fraginfo *info); extern void netif_nit_deliver(struct sk_buff *skb); extern int dev_valid_name(const char *name); extern int dev_ioctl(struct net *net, unsigned int cmd, void __user *); diff --git a/net/core/dev.c b/net/core/dev.c index 1e1a68066457..382df6c09eec 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -132,6 +132,9 @@ /* Instead of increasing this, you should create a hash table. */ #define MAX_GRO_SKBS 8 +/* This should be increased if a protocol with a bigger head is added. */ +#define GRO_MAX_HEAD (MAX_HEADER + 128) + /* * The list of packet types we will receive (as opposed to discard) * and the routines to invoke. @@ -2345,7 +2348,7 @@ static int napi_gro_complete(struct sk_buff *skb) struct list_head *head = &ptype_base[ntohs(type) & PTYPE_HASH_MASK]; int err = -ENOENT; - if (!skb_shinfo(skb)->frag_list) + if (NAPI_GRO_CB(skb)->count == 1) goto out; rcu_read_lock(); @@ -2384,7 +2387,7 @@ void napi_gro_flush(struct napi_struct *napi) } EXPORT_SYMBOL(napi_gro_flush); -int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) +static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { struct sk_buff **pp = NULL; struct packet_type *ptype; @@ -2393,6 +2396,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) int count = 0; int same_flow; int mac_len; + int free; if (!(skb->dev->features & NETIF_F_GRO)) goto normal; @@ -2409,6 +2413,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) skb->mac_len = mac_len; NAPI_GRO_CB(skb)->same_flow = 0; NAPI_GRO_CB(skb)->flush = 0; + NAPI_GRO_CB(skb)->free = 0; for (p = napi->gro_list; p; p = p->next) { count++; @@ -2428,6 +2433,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) goto normal; same_flow = NAPI_GRO_CB(skb)->same_flow; + free = NAPI_GRO_CB(skb)->free; if (pp) { struct sk_buff *nskb = *pp; @@ -2452,13 +2458,86 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) napi->gro_list = skb; ok: - return NET_RX_SUCCESS; + return free; normal: - return netif_receive_skb(skb); + return -1; +} + +int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) +{ + switch (__napi_gro_receive(napi, skb)) { + case -1: + return netif_receive_skb(skb); + + case 1: + kfree_skb(skb); + break; + } + + return NET_RX_SUCCESS; } EXPORT_SYMBOL(napi_gro_receive); +int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) +{ + struct net_device *dev = napi->dev; + struct sk_buff *skb = napi->skb; + int err = NET_RX_DROP; + + napi->skb = NULL; + + if (!skb) { + skb = netdev_alloc_skb(dev, GRO_MAX_HEAD + NET_IP_ALIGN); + if (!skb) + goto out; + + skb_reserve(skb, NET_IP_ALIGN); + } + + BUG_ON(info->nr_frags > MAX_SKB_FRAGS); + skb_shinfo(skb)->nr_frags = info->nr_frags; + memcpy(skb_shinfo(skb)->frags, info->frags, sizeof(info->frags)); + + skb->data_len = info->len; + skb->len += info->len; + skb->truesize += info->len; + + if (!pskb_may_pull(skb, ETH_HLEN)) + goto reuse; + + err = NET_RX_SUCCESS; + + skb->protocol = eth_type_trans(skb, dev); + + skb->ip_summed = info->ip_summed; + skb->csum = info->csum; + + switch (__napi_gro_receive(napi, skb)) { + case -1: + return netif_receive_skb(skb); + + case 0: + goto out; + } + +reuse: + skb_shinfo(skb)->nr_frags = 0; + + skb->len -= skb->data_len; + skb->truesize -= skb->data_len; + skb->data_len = 0; + + __skb_pull(skb, skb_headlen(skb)); + skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); + + napi->skb = skb; + +out: + return err; +} +EXPORT_SYMBOL(napi_gro_frags); + static int process_backlog(struct napi_struct *napi, int quota) { int work = 0; @@ -2537,11 +2616,12 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi, { INIT_LIST_HEAD(&napi->poll_list); napi->gro_list = NULL; + napi->skb = NULL; napi->poll = poll; napi->weight = weight; list_add(&napi->dev_list, &dev->napi_list); -#ifdef CONFIG_NETPOLL napi->dev = dev; +#ifdef CONFIG_NETPOLL spin_lock_init(&napi->poll_lock); napi->poll_owner = -1; #endif @@ -2554,6 +2634,7 @@ void netif_napi_del(struct napi_struct *napi) struct sk_buff *skb, *next; list_del_init(&napi->dev_list); + kfree(napi->skb); for (skb = napi->gro_list; skb; skb = next) { next = skb->next; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3aafb10325b8..5110b359c758 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2594,6 +2594,17 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) if (skb_shinfo(p)->frag_list) goto merge; + else if (!skb_headlen(p) && !skb_headlen(skb) && + skb_shinfo(p)->nr_frags + skb_shinfo(skb)->nr_frags < + MAX_SKB_FRAGS) { + memcpy(skb_shinfo(p)->frags + skb_shinfo(p)->nr_frags, + skb_shinfo(skb)->frags, + skb_shinfo(skb)->nr_frags * sizeof(skb_frag_t)); + + skb_shinfo(p)->nr_frags += skb_shinfo(skb)->nr_frags; + NAPI_GRO_CB(skb)->free = 1; + goto done; + } headroom = skb_headroom(p); nskb = netdev_alloc_skb(p->dev, headroom); @@ -2628,11 +2639,12 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) p = nskb; merge: - NAPI_GRO_CB(p)->count++; p->prev->next = skb; p->prev = skb; skb_header_release(skb); +done: + NAPI_GRO_CB(p)->count++; p->data_len += skb->len; p->truesize += skb->len; p->len += skb->len; -- cgit v1.2.3 From f32f8b72e02e851972a0172603104046aa5fec96 Mon Sep 17 00:00:00 2001 From: Simon Holm Thøgersen Date: Sun, 4 Jan 2009 17:11:24 -0800 Subject: net/rfkill/rfkill.c: fix unused rfkill_led_trigger() warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 4dec9b807be757780ca3611a959ac22c28d292a7 ("rfkill: strip pointless notifier chain") removed the only user of rfkill_led_trigger() that was not guarded by #ifdef CONFIG_RFKILL_LEDS. Therefore, move rfkill_led_trigger() completely inside #ifdef CONFIG_RFKILL_LEDS and avoid the compile time warning: net/rfkill/rfkill.c:59: warning: 'rfkill_led_trigger' defined but not used Signed-off-by: Simon Holm Thøgersen Signed-off-by: David S. Miller --- net/rfkill/rfkill.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c index 3c94f76d5525..3eaa39403c13 100644 --- a/net/rfkill/rfkill.c +++ b/net/rfkill/rfkill.c @@ -54,10 +54,10 @@ static unsigned long rfkill_states_lockdflt[BITS_TO_LONGS(RFKILL_TYPE_MAX)]; static bool rfkill_epo_lock_active; +#ifdef CONFIG_RFKILL_LEDS static void rfkill_led_trigger(struct rfkill *rfkill, enum rfkill_state state) { -#ifdef CONFIG_RFKILL_LEDS struct led_trigger *led = &rfkill->led_trigger; if (!led->name) @@ -66,10 +66,8 @@ static void rfkill_led_trigger(struct rfkill *rfkill, led_trigger_event(led, LED_OFF); else led_trigger_event(led, LED_FULL); -#endif /* CONFIG_RFKILL_LEDS */ } -#ifdef CONFIG_RFKILL_LEDS static void rfkill_led_trigger_activate(struct led_classdev *led) { struct rfkill *rfkill = container_of(led->trigger, -- cgit v1.2.3 From 22604c866889c4b2e12b73cbf1683bda1b72a313 Mon Sep 17 00:00:00 2001 From: Michael Marineau Date: Sun, 4 Jan 2009 17:18:51 -0800 Subject: net: Fix for initial link state in 2.6.28 From: Michael Marineau Commit b47300168e770b60ab96c8924854c3b0eb4260eb "Do not fire linkwatch events until the device is registered." was made as a workaround for drivers that call netif_carrier_off before registering the device. Unfortunately this causes these drivers to incorrectly report their link status as IF_OPER_UNKNOWN which can falsely set the IFF_RUNNING flag when the interface is first brought up. This issues was previously pointed out[1] but was dismissed saying that IFF_RUNNING is not related to the link status. From my digging IFF_RUNNING, as reported to userspace, is based on the link state. It is set based on __LINK_STATE_START and IF_OPER_UP or IF_OPER_UNKNOWN. See [2], [3], and [4]. (Whether or not the kernel has IFF_RUNNING set in flags is not reported to user space so it may well be independent of the link, I don't know if and when it may get set.) The end result depends slightly depending on the driver. The the two I tested were e1000e and b44. With e1000e if the system is booted without a network cable attached the interface will falsely report RUNNING when it is brought up causing NetworkManager to attempt to start it and eventually time out. With b44 when the system is booted with a network cable attached and brought up with dhcpcd it will time out the first time. The attached patch that will still set the operstate variable correctly to IF_OPER_UP/DOWN/etc when linkwatch_fire_event is called but then return rather than skipping the linkwatch_fire_event call entirely as the previous fix did. (sorry it isn't inline, I don't have a patch friendly email client at the moment) Signed-off-by: David S. Miller --- net/core/link_watch.c | 7 ++++++- net/sched/sch_generic.c | 4 ---- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/link_watch.c b/net/core/link_watch.c index bf8f7af699d7..1e401e12dc72 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -178,7 +178,6 @@ static void __linkwatch_run_queue(int urgent_only) */ clear_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state); - rfc2863_policy(dev); if (dev->flags & IFF_UP) { if (netif_carrier_ok(dev)) dev_activate(dev); @@ -215,6 +214,12 @@ void linkwatch_fire_event(struct net_device *dev) { bool urgent = linkwatch_urgent_event(dev); + rfc2863_policy(dev); + + /* Some drivers call netif_carrier_off early */ + if (dev->reg_state == NETREG_UNINITIALIZED) + return; + if (!test_and_set_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state)) { dev_hold(dev); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 5f5efe4e6072..23a8e6141a00 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -270,8 +270,6 @@ static void dev_watchdog_down(struct net_device *dev) void netif_carrier_on(struct net_device *dev) { if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) { - if (dev->reg_state == NETREG_UNINITIALIZED) - return; linkwatch_fire_event(dev); if (netif_running(dev)) __netdev_watchdog_up(dev); @@ -288,8 +286,6 @@ EXPORT_SYMBOL(netif_carrier_on); void netif_carrier_off(struct net_device *dev) { if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) { - if (dev->reg_state == NETREG_UNINITIALIZED) - return; linkwatch_fire_event(dev); } } -- cgit v1.2.3 From 914d11647b6d6fe81bdf0c059612ee36282b8cee Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Sun, 4 Jan 2009 17:27:31 -0800 Subject: ipv6: IPV6_PKTINFO relied userspace providing correct length MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Ilpo Järvinen Reported-by: Eric Sesterhenn Signed-off-by: David S. Miller --- net/ipv6/ipv6_sockglue.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index eeeaad2e8b5c..40f324655e24 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -404,7 +404,7 @@ sticky_done: else if (optlen < sizeof(struct in6_pktinfo) || optval == NULL) goto e_inval; - if (copy_from_user(&pkt, optval, optlen)) { + if (copy_from_user(&pkt, optval, sizeof(struct in6_pktinfo))) { retv = -EFAULT; break; } -- cgit v1.2.3 From 858eb711ba64f8a001d7003295b8078bcab33b6d Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Sun, 4 Jan 2009 17:29:21 -0800 Subject: DCB: fix kfree(skb) Use kfree_skb instead of kfree for struct sk_buff pointers. Signed-off-by: Roel Kluin Signed-off-by: David S. Miller --- net/dcb/dcbnl.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c index 5dbfe5fdc0d6..8379496de82b 100644 --- a/net/dcb/dcbnl.c +++ b/net/dcb/dcbnl.c @@ -191,7 +191,7 @@ static int dcbnl_reply(u8 value, u8 event, u8 cmd, u8 attr, u32 pid, return 0; nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); return ret; } @@ -272,7 +272,7 @@ static int dcbnl_getpfccfg(struct net_device *netdev, struct nlattr **tb, return 0; nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: return -EINVAL; } @@ -314,7 +314,7 @@ static int dcbnl_getperm_hwaddr(struct net_device *netdev, struct nlattr **tb, nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: return -EINVAL; } @@ -380,7 +380,7 @@ static int dcbnl_getcap(struct net_device *netdev, struct nlattr **tb, return 0; nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: return -EINVAL; } @@ -458,7 +458,7 @@ static int dcbnl_getnumtcs(struct net_device *netdev, struct nlattr **tb, return 0; nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: return ret; } @@ -687,7 +687,7 @@ err_pg: nla_nest_cancel(dcbnl_skb, pg_nest); nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: ret = -EINVAL; return ret; @@ -949,7 +949,7 @@ err_bcn: nla_nest_cancel(dcbnl_skb, bcn_nest); nlmsg_failure: err: - kfree(dcbnl_skb); + kfree_skb(dcbnl_skb); err_out: ret = -EINVAL; return ret; -- cgit v1.2.3 From 6e5c172cf7ca1ab878cc6a6a4c1d52fef60f3ee0 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Sun, 4 Jan 2009 17:31:18 -0800 Subject: can: update can-bcm for hrtimer hardirq callbacks Since commit ca109491f612aab5c8152207631c0444f63da97f ("hrtimer: removing all ur callback modes") the hrtimer callbacks are processed only in hardirq context. This patch moves some functionality into tasklets to run in softirq context. Additionally some duplicated code was removed in bcm_rx_thr_flush() and an avoidable memcpy was removed from bcm_rx_handler(). Signed-off-by: Oliver Hartkopp Signed-off-by: David S. Miller --- net/can/bcm.c | 208 +++++++++++++++++++++++++++++++++++----------------------- 1 file changed, 125 insertions(+), 83 deletions(-) (limited to 'net') diff --git a/net/can/bcm.c b/net/can/bcm.c index da0d426c0ce4..6248ae2502c7 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -70,7 +70,7 @@ #define CAN_BCM_VERSION CAN_VERSION static __initdata const char banner[] = KERN_INFO - "can: broadcast manager protocol (rev " CAN_BCM_VERSION ")\n"; + "can: broadcast manager protocol (rev " CAN_BCM_VERSION " t)\n"; MODULE_DESCRIPTION("PF_CAN broadcast manager protocol"); MODULE_LICENSE("Dual BSD/GPL"); @@ -90,6 +90,7 @@ struct bcm_op { unsigned long frames_abs, frames_filtered; struct timeval ival1, ival2; struct hrtimer timer, thrtimer; + struct tasklet_struct tsklet, thrtsklet; ktime_t rx_stamp, kt_ival1, kt_ival2, kt_lastmsg; int rx_ifindex; int count; @@ -341,6 +342,23 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, } } +static void bcm_tx_timeout_tsklet(unsigned long data) +{ + struct bcm_op *op = (struct bcm_op *)data; + struct bcm_msg_head msg_head; + + /* create notification to user */ + msg_head.opcode = TX_EXPIRED; + msg_head.flags = op->flags; + msg_head.count = op->count; + msg_head.ival1 = op->ival1; + msg_head.ival2 = op->ival2; + msg_head.can_id = op->can_id; + msg_head.nframes = 0; + + bcm_send_to_user(op, &msg_head, NULL, 0); +} + /* * bcm_tx_timeout_handler - performes cyclic CAN frame transmissions */ @@ -352,20 +370,8 @@ static enum hrtimer_restart bcm_tx_timeout_handler(struct hrtimer *hrtimer) if (op->kt_ival1.tv64 && (op->count > 0)) { op->count--; - if (!op->count && (op->flags & TX_COUNTEVT)) { - struct bcm_msg_head msg_head; - - /* create notification to user */ - msg_head.opcode = TX_EXPIRED; - msg_head.flags = op->flags; - msg_head.count = op->count; - msg_head.ival1 = op->ival1; - msg_head.ival2 = op->ival2; - msg_head.can_id = op->can_id; - msg_head.nframes = 0; - - bcm_send_to_user(op, &msg_head, NULL, 0); - } + if (!op->count && (op->flags & TX_COUNTEVT)) + tasklet_schedule(&op->tsklet); } if (op->kt_ival1.tv64 && (op->count > 0)) { @@ -402,6 +408,9 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) if (op->frames_filtered > ULONG_MAX/100) op->frames_filtered = op->frames_abs = 0; + /* this element is not throttled anymore */ + data->can_dlc &= (BCM_CAN_DLC_MASK|RX_RECV); + head.opcode = RX_CHANGED; head.flags = op->flags; head.count = op->count; @@ -420,37 +429,32 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) */ static void bcm_rx_update_and_send(struct bcm_op *op, struct can_frame *lastdata, - struct can_frame *rxdata) + const struct can_frame *rxdata) { memcpy(lastdata, rxdata, CFSIZ); - /* mark as used */ - lastdata->can_dlc |= RX_RECV; + /* mark as used and throttled by default */ + lastdata->can_dlc |= (RX_RECV|RX_THR); - /* throtteling mode inactive OR data update already on the run ? */ - if (!op->kt_ival2.tv64 || hrtimer_callback_running(&op->thrtimer)) { + /* throtteling mode inactive ? */ + if (!op->kt_ival2.tv64) { /* send RX_CHANGED to the user immediately */ - bcm_rx_changed(op, rxdata); + bcm_rx_changed(op, lastdata); return; } - if (hrtimer_active(&op->thrtimer)) { - /* mark as 'throttled' */ - lastdata->can_dlc |= RX_THR; + /* with active throttling timer we are just done here */ + if (hrtimer_active(&op->thrtimer)) return; - } - if (!op->kt_lastmsg.tv64) { - /* send first RX_CHANGED to the user immediately */ - bcm_rx_changed(op, rxdata); - op->kt_lastmsg = ktime_get(); - return; - } + /* first receiption with enabled throttling mode */ + if (!op->kt_lastmsg.tv64) + goto rx_changed_settime; + /* got a second frame inside a potential throttle period? */ if (ktime_us_delta(ktime_get(), op->kt_lastmsg) < ktime_to_us(op->kt_ival2)) { - /* mark as 'throttled' and start timer */ - lastdata->can_dlc |= RX_THR; + /* do not send the saved data - only start throttle timer */ hrtimer_start(&op->thrtimer, ktime_add(op->kt_lastmsg, op->kt_ival2), HRTIMER_MODE_ABS); @@ -458,7 +462,8 @@ static void bcm_rx_update_and_send(struct bcm_op *op, } /* the gap was that big, that throttling was not needed here */ - bcm_rx_changed(op, rxdata); +rx_changed_settime: + bcm_rx_changed(op, lastdata); op->kt_lastmsg = ktime_get(); } @@ -467,7 +472,7 @@ static void bcm_rx_update_and_send(struct bcm_op *op, * received data stored in op->last_frames[] */ static void bcm_rx_cmp_to_index(struct bcm_op *op, int index, - struct can_frame *rxdata) + const struct can_frame *rxdata) { /* * no one uses the MSBs of can_dlc for comparation, @@ -511,14 +516,12 @@ static void bcm_rx_starttimer(struct bcm_op *op) hrtimer_start(&op->timer, op->kt_ival1, HRTIMER_MODE_REL); } -/* - * bcm_rx_timeout_handler - when the (cyclic) CAN frame receiption timed out - */ -static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) +static void bcm_rx_timeout_tsklet(unsigned long data) { - struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); + struct bcm_op *op = (struct bcm_op *)data; struct bcm_msg_head msg_head; + /* create notification to user */ msg_head.opcode = RX_TIMEOUT; msg_head.flags = op->flags; msg_head.count = op->count; @@ -528,6 +531,17 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) msg_head.nframes = 0; bcm_send_to_user(op, &msg_head, NULL, 0); +} + +/* + * bcm_rx_timeout_handler - when the (cyclic) CAN frame receiption timed out + */ +static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) +{ + struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); + + /* schedule before NET_RX_SOFTIRQ */ + tasklet_hi_schedule(&op->tsklet); /* no restart of the timer is done here! */ @@ -540,10 +554,26 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) return HRTIMER_NORESTART; } +/* + * bcm_rx_do_flush - helper for bcm_rx_thr_flush + */ +static inline int bcm_rx_do_flush(struct bcm_op *op, int update, int index) +{ + if ((op->last_frames) && (op->last_frames[index].can_dlc & RX_THR)) { + if (update) + bcm_rx_changed(op, &op->last_frames[index]); + return 1; + } + return 0; +} + /* * bcm_rx_thr_flush - Check for throttled data and send it to the userspace + * + * update == 0 : just check if throttled data is available (any irq context) + * update == 1 : check and send throttled data to userspace (soft_irq context) */ -static int bcm_rx_thr_flush(struct bcm_op *op) +static int bcm_rx_thr_flush(struct bcm_op *op, int update) { int updated = 0; @@ -551,27 +581,25 @@ static int bcm_rx_thr_flush(struct bcm_op *op) int i; /* for MUX filter we start at index 1 */ - for (i = 1; i < op->nframes; i++) { - if ((op->last_frames) && - (op->last_frames[i].can_dlc & RX_THR)) { - op->last_frames[i].can_dlc &= ~RX_THR; - bcm_rx_changed(op, &op->last_frames[i]); - updated++; - } - } + for (i = 1; i < op->nframes; i++) + updated += bcm_rx_do_flush(op, update, i); } else { /* for RX_FILTER_ID and simple filter */ - if (op->last_frames && (op->last_frames[0].can_dlc & RX_THR)) { - op->last_frames[0].can_dlc &= ~RX_THR; - bcm_rx_changed(op, &op->last_frames[0]); - updated++; - } + updated += bcm_rx_do_flush(op, update, 0); } return updated; } +static void bcm_rx_thr_tsklet(unsigned long data) +{ + struct bcm_op *op = (struct bcm_op *)data; + + /* push the changed data to the userspace */ + bcm_rx_thr_flush(op, 1); +} + /* * bcm_rx_thr_handler - the time for blocked content updates is over now: * Check for throttled data and send it to the userspace @@ -580,7 +608,9 @@ static enum hrtimer_restart bcm_rx_thr_handler(struct hrtimer *hrtimer) { struct bcm_op *op = container_of(hrtimer, struct bcm_op, thrtimer); - if (bcm_rx_thr_flush(op)) { + tasklet_schedule(&op->thrtsklet); + + if (bcm_rx_thr_flush(op, 0)) { hrtimer_forward(hrtimer, ktime_get(), op->kt_ival2); return HRTIMER_RESTART; } else { @@ -596,48 +626,38 @@ static enum hrtimer_restart bcm_rx_thr_handler(struct hrtimer *hrtimer) static void bcm_rx_handler(struct sk_buff *skb, void *data) { struct bcm_op *op = (struct bcm_op *)data; - struct can_frame rxframe; + const struct can_frame *rxframe = (struct can_frame *)skb->data; int i; /* disable timeout */ hrtimer_cancel(&op->timer); - if (skb->len == sizeof(rxframe)) { - memcpy(&rxframe, skb->data, sizeof(rxframe)); - /* save rx timestamp */ - op->rx_stamp = skb->tstamp; - /* save originator for recvfrom() */ - op->rx_ifindex = skb->dev->ifindex; - /* update statistics */ - op->frames_abs++; - kfree_skb(skb); + if (op->can_id != rxframe->can_id) + goto rx_freeskb; - } else { - kfree_skb(skb); - return; - } - - if (op->can_id != rxframe.can_id) - return; + /* save rx timestamp */ + op->rx_stamp = skb->tstamp; + /* save originator for recvfrom() */ + op->rx_ifindex = skb->dev->ifindex; + /* update statistics */ + op->frames_abs++; if (op->flags & RX_RTR_FRAME) { /* send reply for RTR-request (placed in op->frames[0]) */ bcm_can_tx(op); - return; + goto rx_freeskb; } if (op->flags & RX_FILTER_ID) { /* the easiest case */ - bcm_rx_update_and_send(op, &op->last_frames[0], &rxframe); - bcm_rx_starttimer(op); - return; + bcm_rx_update_and_send(op, &op->last_frames[0], rxframe); + goto rx_freeskb_starttimer; } if (op->nframes == 1) { /* simple compare with index 0 */ - bcm_rx_cmp_to_index(op, 0, &rxframe); - bcm_rx_starttimer(op); - return; + bcm_rx_cmp_to_index(op, 0, rxframe); + goto rx_freeskb_starttimer; } if (op->nframes > 1) { @@ -649,15 +669,19 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) */ for (i = 1; i < op->nframes; i++) { - if ((GET_U64(&op->frames[0]) & GET_U64(&rxframe)) == + if ((GET_U64(&op->frames[0]) & GET_U64(rxframe)) == (GET_U64(&op->frames[0]) & GET_U64(&op->frames[i]))) { - bcm_rx_cmp_to_index(op, i, &rxframe); + bcm_rx_cmp_to_index(op, i, rxframe); break; } } - bcm_rx_starttimer(op); } + +rx_freeskb_starttimer: + bcm_rx_starttimer(op); +rx_freeskb: + kfree_skb(skb); } /* @@ -681,6 +705,12 @@ static void bcm_remove_op(struct bcm_op *op) hrtimer_cancel(&op->timer); hrtimer_cancel(&op->thrtimer); + if (op->tsklet.func) + tasklet_kill(&op->tsklet); + + if (op->thrtsklet.func) + tasklet_kill(&op->thrtsklet); + if ((op->frames) && (op->frames != &op->sframe)) kfree(op->frames); @@ -891,6 +921,10 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, hrtimer_init(&op->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); op->timer.function = bcm_tx_timeout_handler; + /* initialize tasklet for tx countevent notification */ + tasklet_init(&op->tsklet, bcm_tx_timeout_tsklet, + (unsigned long) op); + /* currently unused in tx_ops */ hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); @@ -1054,9 +1088,17 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, hrtimer_init(&op->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); op->timer.function = bcm_rx_timeout_handler; + /* initialize tasklet for rx timeout notification */ + tasklet_init(&op->tsklet, bcm_rx_timeout_tsklet, + (unsigned long) op); + hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); op->thrtimer.function = bcm_rx_thr_handler; + /* initialize tasklet for rx throttle handling */ + tasklet_init(&op->thrtsklet, bcm_rx_thr_tsklet, + (unsigned long) op); + /* add this bcm_op to the list of the rx_ops */ list_add(&op->list, &bo->rx_ops); @@ -1102,7 +1144,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, */ op->kt_lastmsg = ktime_set(0, 0); hrtimer_cancel(&op->thrtimer); - bcm_rx_thr_flush(op); + bcm_rx_thr_flush(op, 1); } if ((op->flags & STARTTIMER) && op->kt_ival1.tv64) -- cgit v1.2.3 From ddebc973c56b51b4e5d84d606f0430d81b895d67 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 4 Jan 2009 21:42:53 -0800 Subject: dccp: Lockless integration of CCID congestion-control plugins Based on Arnaldo's earlier patch, this patch integrates the standardised CCID congestion control plugins (CCID-2 and CCID-3) of DCCP with dccp.ko: * enables a faster connection path by eliminating the need to always go through the CCID registration lock; * updates the implementation to use only a single array whose size equals the number of configured CCIDs instead of the maximum (256); * since the CCIDs are now fixed array elements, synchronization is no longer needed, simplifying use and implementation. CCID-2 is suggested as minimum for a basic DCCP implementation (RFC 4340, 10); CCID-3 is a standards-track CCID supported by RFC 4342 and RFC 5348. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/Kconfig | 4 -- net/dccp/Makefile | 9 ++- net/dccp/ackvec.h | 4 +- net/dccp/ccid.c | 156 +++++++++++++++++++++++++++++++----------------- net/dccp/ccid.h | 11 ++-- net/dccp/ccids/Kconfig | 70 +++++++--------------- net/dccp/ccids/Makefile | 8 --- net/dccp/ccids/ccid2.c | 22 +------ net/dccp/ccids/ccid3.c | 23 +------ net/dccp/dccp.h | 2 - net/dccp/proto.c | 7 +++ 11 files changed, 148 insertions(+), 168 deletions(-) (limited to 'net') diff --git a/net/dccp/Kconfig b/net/dccp/Kconfig index 7aa2a7acc7ec..ad6dffd9070e 100644 --- a/net/dccp/Kconfig +++ b/net/dccp/Kconfig @@ -1,7 +1,6 @@ menuconfig IP_DCCP tristate "The DCCP Protocol (EXPERIMENTAL)" depends on INET && EXPERIMENTAL - select IP_DCCP_CCID2 ---help--- Datagram Congestion Control Protocol (RFC 4340) @@ -25,9 +24,6 @@ config INET_DCCP_DIAG def_tristate y if (IP_DCCP = y && INET_DIAG = y) def_tristate m -config IP_DCCP_ACKVEC - bool - source "net/dccp/ccids/Kconfig" menu "DCCP Kernel Hacking" diff --git a/net/dccp/Makefile b/net/dccp/Makefile index f4f8793aafff..5ff2e7b35690 100644 --- a/net/dccp/Makefile +++ b/net/dccp/Makefile @@ -2,14 +2,19 @@ obj-$(CONFIG_IP_DCCP) += dccp.o dccp_ipv4.o dccp-y := ccid.o feat.o input.o minisocks.o options.o output.o proto.o timer.o +# +# CCID algorithms to be used by dccp.ko +# +# CCID-2 is default (RFC 4340, p. 77) and has Ack Vectors as dependency +dccp-y += ccids/ccid2.o ackvec.o +dccp-$(CONFIG_IP_DCCP_CCID3) += ccids/ccid3.o + dccp_ipv4-y := ipv4.o # build dccp_ipv6 as module whenever either IPv6 or DCCP is a module obj-$(subst y,$(CONFIG_IP_DCCP),$(CONFIG_IPV6)) += dccp_ipv6.o dccp_ipv6-y := ipv6.o -dccp-$(CONFIG_IP_DCCP_ACKVEC) += ackvec.o - obj-$(CONFIG_INET_DCCP_DIAG) += dccp_diag.o obj-$(CONFIG_NET_DCCPPROBE) += dccp_probe.o diff --git a/net/dccp/ackvec.h b/net/dccp/ackvec.h index 4ccee030524e..569a33a79baa 100644 --- a/net/dccp/ackvec.h +++ b/net/dccp/ackvec.h @@ -84,7 +84,7 @@ struct dccp_ackvec_record { struct sock; struct sk_buff; -#ifdef CONFIG_IP_DCCP_ACKVEC +#ifndef ___OLD_INTERFACE_TO_BE_REMOVED___ extern int dccp_ackvec_init(void); extern void dccp_ackvec_exit(void); @@ -106,7 +106,7 @@ static inline int dccp_ackvec_pending(const struct dccp_ackvec *av) { return av->av_vec_len; } -#else /* CONFIG_IP_DCCP_ACKVEC */ +#else /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ static inline int dccp_ackvec_init(void) { return 0; diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index bcc643f992ae..538d3b1da623 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -13,6 +13,70 @@ #include "ccid.h" +static struct ccid_operations *ccids[] = { + &ccid2_ops, +#ifdef CONFIG_IP_DCCP_CCID3 + &ccid3_ops, +#endif +}; + +static struct ccid_operations *ccid_by_number(const u8 id) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(ccids); i++) + if (ccids[i]->ccid_id == id) + return ccids[i]; + return NULL; +} + +/* check that up to @array_len members in @ccid_array are supported */ +bool ccid_support_check(u8 const *ccid_array, u8 array_len) +{ + while (array_len > 0) + if (ccid_by_number(ccid_array[--array_len]) == NULL) + return false; + return true; +} + +/** + * ccid_get_builtin_ccids - Populate a list of built-in CCIDs + * @ccid_array: pointer to copy into + * @array_len: value to return length into + * This function allocates memory - caller must see that it is freed after use. + */ +int ccid_get_builtin_ccids(u8 **ccid_array, u8 *array_len) +{ + *ccid_array = kmalloc(ARRAY_SIZE(ccids), gfp_any()); + if (*ccid_array == NULL) + return -ENOBUFS; + + for (*array_len = 0; *array_len < ARRAY_SIZE(ccids); *array_len += 1) + (*ccid_array)[*array_len] = ccids[*array_len]->ccid_id; + return 0; +} + +int ccid_getsockopt_builtin_ccids(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + u8 *ccid_array, array_len; + int err = 0; + + if (len < ARRAY_SIZE(ccids)) + return -EINVAL; + + if (ccid_get_builtin_ccids(&ccid_array, &array_len)) + return -ENOBUFS; + + if (put_user(array_len, optlen) || + copy_to_user(optval, ccid_array, array_len)) + err = -EFAULT; + + kfree(ccid_array); + return err; +} + +#ifdef ___OLD_INTERFACE_TO_BE_REMOVED___ static u8 builtin_ccids[] = { DCCPC_CCID2, /* CCID2 is supported by default */ #if defined(CONFIG_IP_DCCP_CCID3) || defined(CONFIG_IP_DCCP_CCID3_MODULE) @@ -62,6 +126,7 @@ static inline void ccids_read_unlock(void) #define ccids_read_lock() do { } while(0) #define ccids_read_unlock() do { } while(0) #endif +#endif /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ static struct kmem_cache *ccid_kmem_cache_create(int obj_size, const char *fmt,...) { @@ -93,6 +158,7 @@ static void ccid_kmem_cache_destroy(struct kmem_cache *slab) } } +#ifdef ___OLD_INTERFACE_TO_BE_REMOVED___ /* check that up to @array_len members in @ccid_array are supported */ bool ccid_support_check(u8 const *ccid_array, u8 array_len) { @@ -133,8 +199,9 @@ int ccid_getsockopt_builtin_ccids(struct sock *sk, int len, return -EFAULT; return 0; } +#endif /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ -int ccid_register(struct ccid_operations *ccid_ops) +static int ccid_activate(struct ccid_operations *ccid_ops) { int err = -ENOBUFS; @@ -152,79 +219,40 @@ int ccid_register(struct ccid_operations *ccid_ops) if (ccid_ops->ccid_hc_tx_slab == NULL) goto out_free_rx_slab; - ccids_write_lock(); - err = -EEXIST; - if (ccids[ccid_ops->ccid_id] == NULL) { - ccids[ccid_ops->ccid_id] = ccid_ops; - err = 0; - } - ccids_write_unlock(); - if (err != 0) - goto out_free_tx_slab; - - pr_info("CCID: Registered CCID %d (%s)\n", + pr_info("CCID: Activated CCID %d (%s)\n", ccid_ops->ccid_id, ccid_ops->ccid_name); + err = 0; out: return err; -out_free_tx_slab: - ccid_kmem_cache_destroy(ccid_ops->ccid_hc_tx_slab); - ccid_ops->ccid_hc_tx_slab = NULL; - goto out; out_free_rx_slab: ccid_kmem_cache_destroy(ccid_ops->ccid_hc_rx_slab); ccid_ops->ccid_hc_rx_slab = NULL; goto out; } -EXPORT_SYMBOL_GPL(ccid_register); - -int ccid_unregister(struct ccid_operations *ccid_ops) +static void ccid_deactivate(struct ccid_operations *ccid_ops) { - ccids_write_lock(); - ccids[ccid_ops->ccid_id] = NULL; - ccids_write_unlock(); - ccid_kmem_cache_destroy(ccid_ops->ccid_hc_tx_slab); ccid_ops->ccid_hc_tx_slab = NULL; ccid_kmem_cache_destroy(ccid_ops->ccid_hc_rx_slab); ccid_ops->ccid_hc_rx_slab = NULL; - pr_info("CCID: Unregistered CCID %d (%s)\n", + pr_info("CCID: Deactivated CCID %d (%s)\n", ccid_ops->ccid_id, ccid_ops->ccid_name); - return 0; } -EXPORT_SYMBOL_GPL(ccid_unregister); - struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp) { - struct ccid_operations *ccid_ops; + struct ccid_operations *ccid_ops = ccid_by_number(id); struct ccid *ccid = NULL; - ccids_read_lock(); -#ifdef CONFIG_MODULES - if (ccids[id] == NULL) { - /* We only try to load if in process context */ - ccids_read_unlock(); - if (gfp & GFP_ATOMIC) - goto out; - request_module("net-dccp-ccid-%d", id); - ccids_read_lock(); - } -#endif - ccid_ops = ccids[id]; if (ccid_ops == NULL) - goto out_unlock; - - if (!try_module_get(ccid_ops->ccid_owner)) - goto out_unlock; - - ccids_read_unlock(); + goto out; ccid = kmem_cache_alloc(rx ? ccid_ops->ccid_hc_rx_slab : ccid_ops->ccid_hc_tx_slab, gfp); if (ccid == NULL) - goto out_module_put; + goto out; ccid->ccid_ops = ccid_ops; if (rx) { memset(ccid + 1, 0, ccid_ops->ccid_hc_rx_obj_size); @@ -239,15 +267,10 @@ struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp) } out: return ccid; -out_unlock: - ccids_read_unlock(); - goto out; out_free_ccid: kmem_cache_free(rx ? ccid_ops->ccid_hc_rx_slab : ccid_ops->ccid_hc_tx_slab, ccid); ccid = NULL; -out_module_put: - module_put(ccid_ops->ccid_owner); goto out; } @@ -270,10 +293,6 @@ static void ccid_delete(struct ccid *ccid, struct sock *sk, int rx) ccid_ops->ccid_hc_tx_exit(sk); kmem_cache_free(ccid_ops->ccid_hc_tx_slab, ccid); } - ccids_read_lock(); - if (ccids[ccid_ops->ccid_id] != NULL) - module_put(ccid_ops->ccid_owner); - ccids_read_unlock(); } void ccid_hc_rx_delete(struct ccid *ccid, struct sock *sk) @@ -289,3 +308,28 @@ void ccid_hc_tx_delete(struct ccid *ccid, struct sock *sk) } EXPORT_SYMBOL_GPL(ccid_hc_tx_delete); + +int __init ccid_initialize_builtins(void) +{ + int i, err; + + for (i = 0; i < ARRAY_SIZE(ccids); i++) { + err = ccid_activate(ccids[i]); + if (err) + goto unwind_registrations; + } + return 0; + +unwind_registrations: + while(--i >= 0) + ccid_deactivate(ccids[i]); + return err; +} + +void ccid_cleanup_builtins(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(ccids); i++) + ccid_deactivate(ccids[i]); +} diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index 18f69423a708..75c21c52ed7a 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -29,7 +29,6 @@ struct tcp_info; * @ccid_id: numerical CCID ID (up to %CCID_MAX, cf. table 5 in RFC 4340, 10.) * @ccid_ccmps: the CCMPS including network/transport headers (0 when disabled) * @ccid_name: alphabetical identifier string for @ccid_id - * @ccid_owner: module which implements/owns this CCID * @ccid_hc_{r,t}x_slab: memory pool for the receiver/sender half-connection * @ccid_hc_{r,t}x_obj_size: size of the receiver/sender half-connection socket * @@ -48,7 +47,6 @@ struct ccid_operations { unsigned char ccid_id; __u32 ccid_ccmps; const char *ccid_name; - struct module *ccid_owner; struct kmem_cache *ccid_hc_rx_slab, *ccid_hc_tx_slab; __u32 ccid_hc_rx_obj_size, @@ -90,8 +88,13 @@ struct ccid_operations { int __user *optlen); }; -extern int ccid_register(struct ccid_operations *ccid_ops); -extern int ccid_unregister(struct ccid_operations *ccid_ops); +extern struct ccid_operations ccid2_ops; +#ifdef CONFIG_IP_DCCP_CCID3 +extern struct ccid_operations ccid3_ops; +#endif + +extern int ccid_initialize_builtins(void); +extern void ccid_cleanup_builtins(void); struct ccid { struct ccid_operations *ccid_ops; diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index 12275943eab8..b30f049cf1d3 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -1,80 +1,52 @@ menu "DCCP CCIDs Configuration (EXPERIMENTAL)" depends on EXPERIMENTAL -config IP_DCCP_CCID2 - tristate "CCID2 (TCP-Like) (EXPERIMENTAL)" - def_tristate IP_DCCP - select IP_DCCP_ACKVEC - ---help--- - CCID 2, TCP-like Congestion Control, denotes Additive Increase, - Multiplicative Decrease (AIMD) congestion control with behavior - modelled directly on TCP, including congestion window, slow start, - timeouts, and so forth [RFC 2581]. CCID 2 achieves maximum - bandwidth over the long term, consistent with the use of end-to-end - congestion control, but halves its congestion window in response to - each congestion event. This leads to the abrupt rate changes - typical of TCP. Applications should use CCID 2 if they prefer - maximum bandwidth utilization to steadiness of rate. This is often - the case for applications that are not playing their data directly - to the user. For example, a hypothetical application that - transferred files over DCCP, using application-level retransmissions - for lost packets, would prefer CCID 2 to CCID 3. On-line games may - also prefer CCID 2. See RFC 4341 for further details. - - CCID2 is the default CCID used by DCCP. - config IP_DCCP_CCID2_DEBUG - bool "CCID2 debugging messages" - depends on IP_DCCP_CCID2 - ---help--- - Enable CCID2-specific debugging messages. + bool "CCID-2 debugging messages" + ---help--- + Enable CCID-2 specific debugging messages. - When compiling CCID2 as a module, this debugging output can - additionally be toggled by setting the ccid2_debug module - parameter to 0 or 1. + The debugging output can additionally be toggled by setting the + ccid2_debug parameter to 0 or 1. - If in doubt, say N. + If in doubt, say N. config IP_DCCP_CCID3 - tristate "CCID3 (TCP-Friendly) (EXPERIMENTAL)" - def_tristate IP_DCCP + bool "CCID-3 (TCP-Friendly) (EXPERIMENTAL)" + def_bool y if (IP_DCCP = y || IP_DCCP = m) select IP_DCCP_TFRC_LIB ---help--- - CCID 3 denotes TCP-Friendly Rate Control (TFRC), an equation-based + CCID-3 denotes TCP-Friendly Rate Control (TFRC), an equation-based rate-controlled congestion control mechanism. TFRC is designed to be reasonably fair when competing for bandwidth with TCP-like flows, where a flow is "reasonably fair" if its sending rate is generally within a factor of two of the sending rate of a TCP flow under the same conditions. However, TFRC has a much lower variation of - throughput over time compared with TCP, which makes CCID 3 more - suitable than CCID 2 for applications such streaming media where a + throughput over time compared with TCP, which makes CCID-3 more + suitable than CCID-2 for applications such streaming media where a relatively smooth sending rate is of importance. - CCID 3 is further described in RFC 4342, + CCID-3 is further described in RFC 4342, http://www.ietf.org/rfc/rfc4342.txt The TFRC congestion control algorithms were initially described in - RFC 3448. + RFC 5448. This text was extracted from RFC 4340 (sec. 10.2), http://www.ietf.org/rfc/rfc4340.txt - - To compile this CCID as a module, choose M here: the module will be - called dccp_ccid3. - If in doubt, say M. + If in doubt, say N. config IP_DCCP_CCID3_DEBUG - bool "CCID3 debugging messages" - depends on IP_DCCP_CCID3 - ---help--- - Enable CCID3-specific debugging messages. + bool "CCID-3 debugging messages" + depends on IP_DCCP_CCID3 + ---help--- + Enable CCID-3 specific debugging messages. - When compiling CCID3 as a module, this debugging output can - additionally be toggled by setting the ccid3_debug module - parameter to 0 or 1. + The debugging output can additionally be toggled by setting the + ccid3_debug parameter to 0 or 1. - If in doubt, say N. + If in doubt, say N. config IP_DCCP_CCID3_RTO int "Use higher bound for nofeedback timer" diff --git a/net/dccp/ccids/Makefile b/net/dccp/ccids/Makefile index 438f20bccff7..cdaefffbb777 100644 --- a/net/dccp/ccids/Makefile +++ b/net/dccp/ccids/Makefile @@ -1,9 +1 @@ -obj-$(CONFIG_IP_DCCP_CCID3) += dccp_ccid3.o - -dccp_ccid3-y := ccid3.o - -obj-$(CONFIG_IP_DCCP_CCID2) += dccp_ccid2.o - -dccp_ccid2-y := ccid2.o - obj-y += lib/ diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index c9ea19a4d85e..d235294ace23 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -768,10 +768,9 @@ static void ccid2_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) } } -static struct ccid_operations ccid2 = { +struct ccid_operations ccid2_ops = { .ccid_id = DCCPC_CCID2, .ccid_name = "TCP-like", - .ccid_owner = THIS_MODULE, .ccid_hc_tx_obj_size = sizeof(struct ccid2_hc_tx_sock), .ccid_hc_tx_init = ccid2_hc_tx_init, .ccid_hc_tx_exit = ccid2_hc_tx_exit, @@ -784,22 +783,5 @@ static struct ccid_operations ccid2 = { #ifdef CONFIG_IP_DCCP_CCID2_DEBUG module_param(ccid2_debug, bool, 0644); -MODULE_PARM_DESC(ccid2_debug, "Enable debug messages"); +MODULE_PARM_DESC(ccid2_debug, "Enable CCID-2 debug messages"); #endif - -static __init int ccid2_module_init(void) -{ - return ccid_register(&ccid2); -} -module_init(ccid2_module_init); - -static __exit void ccid2_module_exit(void) -{ - ccid_unregister(&ccid2); -} -module_exit(ccid2_module_exit); - -MODULE_AUTHOR("Andrea Bittau "); -MODULE_DESCRIPTION("DCCP TCP-Like (CCID2) CCID"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS("net-dccp-ccid-2"); diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 3b8bd7ca6761..a27b7f4c19c5 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -940,10 +940,9 @@ static int ccid3_hc_rx_getsockopt(struct sock *sk, const int optname, int len, return 0; } -static struct ccid_operations ccid3 = { +struct ccid_operations ccid3_ops = { .ccid_id = DCCPC_CCID3, .ccid_name = "TCP-Friendly Rate Control", - .ccid_owner = THIS_MODULE, .ccid_hc_tx_obj_size = sizeof(struct ccid3_hc_tx_sock), .ccid_hc_tx_init = ccid3_hc_tx_init, .ccid_hc_tx_exit = ccid3_hc_tx_exit, @@ -964,23 +963,5 @@ static struct ccid_operations ccid3 = { #ifdef CONFIG_IP_DCCP_CCID3_DEBUG module_param(ccid3_debug, bool, 0644); -MODULE_PARM_DESC(ccid3_debug, "Enable debug messages"); +MODULE_PARM_DESC(ccid3_debug, "Enable CCID-3 debug messages"); #endif - -static __init int ccid3_module_init(void) -{ - return ccid_register(&ccid3); -} -module_init(ccid3_module_init); - -static __exit void ccid3_module_exit(void) -{ - ccid_unregister(&ccid3); -} -module_exit(ccid3_module_exit); - -MODULE_AUTHOR("Ian McDonald , " - "Arnaldo Carvalho de Melo "); -MODULE_DESCRIPTION("DCCP TFRC CCID3 CCID"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS("net-dccp-ccid-3"); diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 0bc4c9a02e19..f2230fc168e1 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -432,10 +432,8 @@ static inline int dccp_ack_pending(const struct sock *sk) { const struct dccp_sock *dp = dccp_sk(sk); return dp->dccps_timestamp_echo != 0 || -#ifdef CONFIG_IP_DCCP_ACKVEC (dp->dccps_hc_rx_ackvec != NULL && dccp_ackvec_pending(dp->dccps_hc_rx_ackvec)) || -#endif inet_csk_ack_scheduled(sk); } diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 1747ccae8e8d..945b4d5d23b3 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -1118,9 +1118,15 @@ static int __init dccp_init(void) if (rc) goto out_ackvec_exit; + rc = ccid_initialize_builtins(); + if (rc) + goto out_sysctl_exit; + dccp_timestamping_init(); out: return rc; +out_sysctl_exit: + dccp_sysctl_exit(); out_ackvec_exit: dccp_ackvec_exit(); out_free_dccp_mib: @@ -1143,6 +1149,7 @@ out_free_percpu: static void __exit dccp_fini(void) { + ccid_cleanup_builtins(); dccp_mib_exit(); free_pages((unsigned long)dccp_hashinfo.bhash, get_order(dccp_hashinfo.bhash_size * -- cgit v1.2.3 From e5fd56ca4eb3a130882bbef69d6952ef6aca5c8d Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 4 Jan 2009 21:43:23 -0800 Subject: dccp: Clean up ccid.c after integration of CCID plugins This patch cleans up after integrating the CCID modules and, in addition, * moves the if/else cases from ccid_delete() into ccid_hc_{tx,rx}_delete(); * removes the 'gfp' argument to ccid_new() - since it is always gfp_any(). Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/ackvec.h | 49 -------------------- net/dccp/ccid.c | 136 +++++------------------------------------------------- net/dccp/ccid.h | 3 +- net/dccp/feat.c | 2 +- 4 files changed, 14 insertions(+), 176 deletions(-) (limited to 'net') diff --git a/net/dccp/ackvec.h b/net/dccp/ackvec.h index 569a33a79baa..45f95e55f873 100644 --- a/net/dccp/ackvec.h +++ b/net/dccp/ackvec.h @@ -84,7 +84,6 @@ struct dccp_ackvec_record { struct sock; struct sk_buff; -#ifndef ___OLD_INTERFACE_TO_BE_REMOVED___ extern int dccp_ackvec_init(void); extern void dccp_ackvec_exit(void); @@ -106,52 +105,4 @@ static inline int dccp_ackvec_pending(const struct dccp_ackvec *av) { return av->av_vec_len; } -#else /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ -static inline int dccp_ackvec_init(void) -{ - return 0; -} - -static inline void dccp_ackvec_exit(void) -{ -} - -static inline struct dccp_ackvec *dccp_ackvec_alloc(const gfp_t priority) -{ - return NULL; -} - -static inline void dccp_ackvec_free(struct dccp_ackvec *av) -{ -} - -static inline int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, - const u64 ackno, const u8 state) -{ - return -1; -} - -static inline void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, - struct sock *sk, const u64 ackno) -{ -} - -static inline int dccp_ackvec_parse(struct sock *sk, const struct sk_buff *skb, - const u64 *ackno, const u8 opt, - const u8 *value, const u8 len) -{ - return -1; -} - -static inline int dccp_insert_option_ackvec(const struct sock *sk, - const struct sk_buff *skb) -{ - return -1; -} - -static inline int dccp_ackvec_pending(const struct dccp_ackvec *av) -{ - return 0; -} -#endif /* CONFIG_IP_DCCP_ACKVEC */ #endif /* _ACKVEC_H */ diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index 538d3b1da623..19b214ad5e06 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -76,58 +76,6 @@ int ccid_getsockopt_builtin_ccids(struct sock *sk, int len, return err; } -#ifdef ___OLD_INTERFACE_TO_BE_REMOVED___ -static u8 builtin_ccids[] = { - DCCPC_CCID2, /* CCID2 is supported by default */ -#if defined(CONFIG_IP_DCCP_CCID3) || defined(CONFIG_IP_DCCP_CCID3_MODULE) - DCCPC_CCID3, -#endif -}; - -static struct ccid_operations *ccids[CCID_MAX]; -#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) -static atomic_t ccids_lockct = ATOMIC_INIT(0); -static DEFINE_SPINLOCK(ccids_lock); - -/* - * The strategy is: modifications ccids vector are short, do not sleep and - * veeery rare, but read access should be free of any exclusive locks. - */ -static void ccids_write_lock(void) -{ - spin_lock(&ccids_lock); - while (atomic_read(&ccids_lockct) != 0) { - spin_unlock(&ccids_lock); - yield(); - spin_lock(&ccids_lock); - } -} - -static inline void ccids_write_unlock(void) -{ - spin_unlock(&ccids_lock); -} - -static inline void ccids_read_lock(void) -{ - atomic_inc(&ccids_lockct); - smp_mb__after_atomic_inc(); - spin_unlock_wait(&ccids_lock); -} - -static inline void ccids_read_unlock(void) -{ - atomic_dec(&ccids_lockct); -} - -#else -#define ccids_write_lock() do { } while(0) -#define ccids_write_unlock() do { } while(0) -#define ccids_read_lock() do { } while(0) -#define ccids_read_unlock() do { } while(0) -#endif -#endif /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ - static struct kmem_cache *ccid_kmem_cache_create(int obj_size, const char *fmt,...) { struct kmem_cache *slab; @@ -158,49 +106,6 @@ static void ccid_kmem_cache_destroy(struct kmem_cache *slab) } } -#ifdef ___OLD_INTERFACE_TO_BE_REMOVED___ -/* check that up to @array_len members in @ccid_array are supported */ -bool ccid_support_check(u8 const *ccid_array, u8 array_len) -{ - u8 i, j, found; - - for (i = 0, found = 0; i < array_len; i++, found = 0) { - for (j = 0; !found && j < ARRAY_SIZE(builtin_ccids); j++) - found = (ccid_array[i] == builtin_ccids[j]); - if (!found) - return false; - } - return true; -} - -/** - * ccid_get_builtin_ccids - Provide copy of `builtin' CCID array - * @ccid_array: pointer to copy into - * @array_len: value to return length into - * This function allocates memory - caller must see that it is freed after use. - */ -int ccid_get_builtin_ccids(u8 **ccid_array, u8 *array_len) -{ - *ccid_array = kmemdup(builtin_ccids, sizeof(builtin_ccids), gfp_any()); - if (*ccid_array == NULL) - return -ENOBUFS; - *array_len = ARRAY_SIZE(builtin_ccids); - return 0; -} - -int ccid_getsockopt_builtin_ccids(struct sock *sk, int len, - char __user *optval, int __user *optlen) -{ - if (len < sizeof(builtin_ccids)) - return -EINVAL; - - if (put_user(sizeof(builtin_ccids), optlen) || - copy_to_user(optval, builtin_ccids, sizeof(builtin_ccids))) - return -EFAULT; - return 0; -} -#endif /* ___OLD_INTERFACE_TO_BE_REMOVED___ */ - static int ccid_activate(struct ccid_operations *ccid_ops) { int err = -ENOBUFS; @@ -241,7 +146,7 @@ static void ccid_deactivate(struct ccid_operations *ccid_ops) ccid_ops->ccid_id, ccid_ops->ccid_name); } -struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp) +struct ccid *ccid_new(const u8 id, struct sock *sk, bool rx) { struct ccid_operations *ccid_ops = ccid_by_number(id); struct ccid *ccid = NULL; @@ -250,7 +155,7 @@ struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp) goto out; ccid = kmem_cache_alloc(rx ? ccid_ops->ccid_hc_rx_slab : - ccid_ops->ccid_hc_tx_slab, gfp); + ccid_ops->ccid_hc_tx_slab, gfp_any()); if (ccid == NULL) goto out; ccid->ccid_ops = ccid_ops; @@ -274,41 +179,24 @@ out_free_ccid: goto out; } -EXPORT_SYMBOL_GPL(ccid_new); - -static void ccid_delete(struct ccid *ccid, struct sock *sk, int rx) -{ - struct ccid_operations *ccid_ops; - - if (ccid == NULL) - return; - - ccid_ops = ccid->ccid_ops; - if (rx) { - if (ccid_ops->ccid_hc_rx_exit != NULL) - ccid_ops->ccid_hc_rx_exit(sk); - kmem_cache_free(ccid_ops->ccid_hc_rx_slab, ccid); - } else { - if (ccid_ops->ccid_hc_tx_exit != NULL) - ccid_ops->ccid_hc_tx_exit(sk); - kmem_cache_free(ccid_ops->ccid_hc_tx_slab, ccid); - } -} - void ccid_hc_rx_delete(struct ccid *ccid, struct sock *sk) { - ccid_delete(ccid, sk, 1); + if (ccid != NULL) { + if (ccid->ccid_ops->ccid_hc_rx_exit != NULL) + ccid->ccid_ops->ccid_hc_rx_exit(sk); + kmem_cache_free(ccid->ccid_ops->ccid_hc_rx_slab, ccid); + } } -EXPORT_SYMBOL_GPL(ccid_hc_rx_delete); - void ccid_hc_tx_delete(struct ccid *ccid, struct sock *sk) { - ccid_delete(ccid, sk, 0); + if (ccid != NULL) { + if (ccid->ccid_ops->ccid_hc_tx_exit != NULL) + ccid->ccid_ops->ccid_hc_tx_exit(sk); + kmem_cache_free(ccid->ccid_ops->ccid_hc_tx_slab, ccid); + } } -EXPORT_SYMBOL_GPL(ccid_hc_tx_delete); - int __init ccid_initialize_builtins(void) { int i, err; diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index 75c21c52ed7a..facedd20b531 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -111,8 +111,7 @@ extern int ccid_get_builtin_ccids(u8 **ccid_array, u8 *array_len); extern int ccid_getsockopt_builtin_ccids(struct sock *sk, int len, char __user *, int __user *); -extern struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, - gfp_t gfp); +extern struct ccid *ccid_new(const u8 id, struct sock *sk, bool rx); static inline int ccid_get_current_rx_ccid(struct dccp_sock *dp) { diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 30f9fb76b921..741b2db24735 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -34,7 +34,7 @@ static int dccp_hdlr_ccid(struct sock *sk, u64 ccid, bool rx) { struct dccp_sock *dp = dccp_sk(sk); - struct ccid *new_ccid = ccid_new(ccid, sk, rx, gfp_any()); + struct ccid *new_ccid = ccid_new(ccid, sk, rx); if (new_ccid == NULL) return -ENOMEM; -- cgit v1.2.3 From 129fa44785a399248ae2466b6cb5c655e96668f7 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 4 Jan 2009 21:45:33 -0800 Subject: dccp: Integrate the TFRC library with DCCP This patch integrates the TFRC library, which is a dependency of CCID-3 (and CCID-4), with the new use of CCIDs in the DCCP module. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/Makefile | 6 ++++-- net/dccp/ccid.c | 8 +++++++- net/dccp/ccids/Kconfig | 9 ++------- net/dccp/ccids/Makefile | 1 - net/dccp/ccids/lib/Makefile | 3 --- net/dccp/ccids/lib/loss_interval.c | 3 --- net/dccp/ccids/lib/packet_history.c | 9 --------- net/dccp/ccids/lib/tfrc.c | 19 ++++--------------- net/dccp/ccids/lib/tfrc.h | 11 ++++++++++- net/dccp/ccids/lib/tfrc_equation.c | 4 ---- net/dccp/feat.c | 4 ---- net/dccp/input.c | 2 -- 12 files changed, 27 insertions(+), 52 deletions(-) delete mode 100644 net/dccp/ccids/Makefile delete mode 100644 net/dccp/ccids/lib/Makefile (limited to 'net') diff --git a/net/dccp/Makefile b/net/dccp/Makefile index 5ff2e7b35690..2991efcc8dea 100644 --- a/net/dccp/Makefile +++ b/net/dccp/Makefile @@ -8,6 +8,10 @@ dccp-y := ccid.o feat.o input.o minisocks.o options.o output.o proto.o timer.o # CCID-2 is default (RFC 4340, p. 77) and has Ack Vectors as dependency dccp-y += ccids/ccid2.o ackvec.o dccp-$(CONFIG_IP_DCCP_CCID3) += ccids/ccid3.o +dccp-$(CONFIG_IP_DCCP_TFRC_LIB) += ccids/lib/tfrc.o \ + ccids/lib/tfrc_equation.o \ + ccids/lib/packet_history.o \ + ccids/lib/loss_interval.o dccp_ipv4-y := ipv4.o @@ -22,5 +26,3 @@ dccp-$(CONFIG_SYSCTL) += sysctl.o dccp_diag-y := diag.o dccp_probe-y := probe.o - -obj-y += ccids/ diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index 19b214ad5e06..f3e9ba1cfd01 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -12,6 +12,7 @@ */ #include "ccid.h" +#include "ccids/lib/tfrc.h" static struct ccid_operations *ccids[] = { &ccid2_ops, @@ -199,7 +200,10 @@ void ccid_hc_tx_delete(struct ccid *ccid, struct sock *sk) int __init ccid_initialize_builtins(void) { - int i, err; + int i, err = tfrc_lib_init(); + + if (err) + return err; for (i = 0; i < ARRAY_SIZE(ccids); i++) { err = ccid_activate(ccids[i]); @@ -211,6 +215,7 @@ int __init ccid_initialize_builtins(void) unwind_registrations: while(--i >= 0) ccid_deactivate(ccids[i]); + tfrc_lib_exit(); return err; } @@ -220,4 +225,5 @@ void ccid_cleanup_builtins(void) for (i = 0; i < ARRAY_SIZE(ccids); i++) ccid_deactivate(ccids[i]); + tfrc_lib_exit(); } diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index b30f049cf1d3..b28bf962edc3 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -14,7 +14,6 @@ config IP_DCCP_CCID2_DEBUG config IP_DCCP_CCID3 bool "CCID-3 (TCP-Friendly) (EXPERIMENTAL)" def_bool y if (IP_DCCP = y || IP_DCCP = m) - select IP_DCCP_TFRC_LIB ---help--- CCID-3 denotes TCP-Friendly Rate Control (TFRC), an equation-based rate-controlled congestion control mechanism. TFRC is designed to @@ -80,12 +79,8 @@ config IP_DCCP_CCID3_RTO therefore not be performed on WANs. config IP_DCCP_TFRC_LIB - tristate - default n + def_bool y if IP_DCCP_CCID3 config IP_DCCP_TFRC_DEBUG - bool - depends on IP_DCCP_TFRC_LIB - default y if IP_DCCP_CCID3_DEBUG - + def_bool y if IP_DCCP_CCID3_DEBUG endmenu diff --git a/net/dccp/ccids/Makefile b/net/dccp/ccids/Makefile deleted file mode 100644 index cdaefffbb777..000000000000 --- a/net/dccp/ccids/Makefile +++ /dev/null @@ -1 +0,0 @@ -obj-y += lib/ diff --git a/net/dccp/ccids/lib/Makefile b/net/dccp/ccids/lib/Makefile deleted file mode 100644 index 68c93e3d89dc..000000000000 --- a/net/dccp/ccids/lib/Makefile +++ /dev/null @@ -1,3 +0,0 @@ -obj-$(CONFIG_IP_DCCP_TFRC_LIB) += dccp_tfrc_lib.o - -dccp_tfrc_lib-y := tfrc.o tfrc_equation.o packet_history.o loss_interval.o diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c index 5b3ce0688c5c..4d1e40127264 100644 --- a/net/dccp/ccids/lib/loss_interval.c +++ b/net/dccp/ccids/lib/loss_interval.c @@ -60,7 +60,6 @@ void tfrc_lh_cleanup(struct tfrc_loss_hist *lh) lh->ring[LIH_INDEX(lh->counter)] = NULL; } } -EXPORT_SYMBOL_GPL(tfrc_lh_cleanup); static void tfrc_lh_calc_i_mean(struct tfrc_loss_hist *lh) { @@ -121,7 +120,6 @@ u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb) return (lh->i_mean < old_i_mean); } -EXPORT_SYMBOL_GPL(tfrc_lh_update_i_mean); /* Determine if `new_loss' does begin a new loss interval [RFC 4342, 10.2] */ static inline u8 tfrc_lh_is_new_loss(struct tfrc_loss_interval *cur, @@ -169,7 +167,6 @@ int tfrc_lh_interval_add(struct tfrc_loss_hist *lh, struct tfrc_rx_hist *rh, } return 1; } -EXPORT_SYMBOL_GPL(tfrc_lh_interval_add); int __init tfrc_li_init(void) { diff --git a/net/dccp/ccids/lib/packet_history.c b/net/dccp/ccids/lib/packet_history.c index 6cc108afdc3b..b7785b3581ec 100644 --- a/net/dccp/ccids/lib/packet_history.c +++ b/net/dccp/ccids/lib/packet_history.c @@ -94,7 +94,6 @@ int tfrc_tx_hist_add(struct tfrc_tx_hist_entry **headp, u64 seqno) *headp = entry; return 0; } -EXPORT_SYMBOL_GPL(tfrc_tx_hist_add); void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp) { @@ -109,7 +108,6 @@ void tfrc_tx_hist_purge(struct tfrc_tx_hist_entry **headp) *headp = NULL; } -EXPORT_SYMBOL_GPL(tfrc_tx_hist_purge); u32 tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head, const u64 seqno, const ktime_t now) @@ -127,7 +125,6 @@ u32 tfrc_tx_hist_rtt(struct tfrc_tx_hist_entry *head, const u64 seqno, return rtt; } -EXPORT_SYMBOL_GPL(tfrc_tx_hist_rtt); /* @@ -172,7 +169,6 @@ void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h, tfrc_rx_hist_entry_from_skb(entry, skb, ndp); } -EXPORT_SYMBOL_GPL(tfrc_rx_hist_add_packet); /* has the packet contained in skb been seen before? */ int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb) @@ -189,7 +185,6 @@ int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb) return 0; } -EXPORT_SYMBOL_GPL(tfrc_rx_hist_duplicate); static void tfrc_rx_hist_swap(struct tfrc_rx_hist *h, const u8 a, const u8 b) { @@ -390,7 +385,6 @@ int tfrc_rx_handle_loss(struct tfrc_rx_hist *h, } return is_new_loss; } -EXPORT_SYMBOL_GPL(tfrc_rx_handle_loss); int tfrc_rx_hist_alloc(struct tfrc_rx_hist *h) { @@ -412,7 +406,6 @@ out_free: } return -ENOBUFS; } -EXPORT_SYMBOL_GPL(tfrc_rx_hist_alloc); void tfrc_rx_hist_purge(struct tfrc_rx_hist *h) { @@ -424,7 +417,6 @@ void tfrc_rx_hist_purge(struct tfrc_rx_hist *h) h->ring[i] = NULL; } } -EXPORT_SYMBOL_GPL(tfrc_rx_hist_purge); /** * tfrc_rx_hist_rtt_last_s - reference entry to compute RTT samples against @@ -495,4 +487,3 @@ keep_ref_for_next_time: return sample; } -EXPORT_SYMBOL_GPL(tfrc_rx_hist_sample_rtt); diff --git a/net/dccp/ccids/lib/tfrc.c b/net/dccp/ccids/lib/tfrc.c index 185916218e07..60c412ccfeef 100644 --- a/net/dccp/ccids/lib/tfrc.c +++ b/net/dccp/ccids/lib/tfrc.c @@ -1,20 +1,18 @@ /* - * TFRC: main module holding the pieces of the TFRC library together + * TFRC library initialisation * * Copyright (c) 2007 The University of Aberdeen, Scotland, UK * Copyright (c) 2007 Arnaldo Carvalho de Melo */ -#include -#include #include "tfrc.h" #ifdef CONFIG_IP_DCCP_TFRC_DEBUG int tfrc_debug; module_param(tfrc_debug, bool, 0644); -MODULE_PARM_DESC(tfrc_debug, "Enable debug messages"); +MODULE_PARM_DESC(tfrc_debug, "Enable TFRC debug messages"); #endif -static int __init tfrc_module_init(void) +int __init tfrc_lib_init(void) { int rc = tfrc_li_init(); @@ -38,18 +36,9 @@ out: return rc; } -static void __exit tfrc_module_exit(void) +void __exit tfrc_lib_exit(void) { tfrc_rx_packet_history_exit(); tfrc_tx_packet_history_exit(); tfrc_li_exit(); } - -module_init(tfrc_module_init); -module_exit(tfrc_module_exit); - -MODULE_AUTHOR("Gerrit Renker , " - "Ian McDonald , " - "Arnaldo Carvalho de Melo "); -MODULE_DESCRIPTION("DCCP TFRC library"); -MODULE_LICENSE("GPL"); diff --git a/net/dccp/ccids/lib/tfrc.h b/net/dccp/ccids/lib/tfrc.h index ed9857527acf..e9720b143275 100644 --- a/net/dccp/ccids/lib/tfrc.h +++ b/net/dccp/ccids/lib/tfrc.h @@ -17,7 +17,8 @@ #include #include #include "../../dccp.h" -/* internal includes that this module exports: */ + +/* internal includes that this library exports: */ #include "loss_interval.h" #include "packet_history.h" @@ -66,4 +67,12 @@ extern void tfrc_rx_packet_history_exit(void); extern int tfrc_li_init(void); extern void tfrc_li_exit(void); + +#ifdef CONFIG_IP_DCCP_TFRC_LIB +extern int tfrc_lib_init(void); +extern void tfrc_lib_exit(void); +#else +#define tfrc_lib_init() (0) +#define tfrc_lib_exit() +#endif #endif /* _TFRC_H_ */ diff --git a/net/dccp/ccids/lib/tfrc_equation.c b/net/dccp/ccids/lib/tfrc_equation.c index 2f20a29cffe4..c5d3a9e5a5a4 100644 --- a/net/dccp/ccids/lib/tfrc_equation.c +++ b/net/dccp/ccids/lib/tfrc_equation.c @@ -659,8 +659,6 @@ u32 tfrc_calc_x(u16 s, u32 R, u32 p) return scaled_div32(result, f); } -EXPORT_SYMBOL_GPL(tfrc_calc_x); - /** * tfrc_calc_x_reverse_lookup - try to find p given f(p) * @@ -693,5 +691,3 @@ u32 tfrc_calc_x_reverse_lookup(u32 fvalue) index = tfrc_binsearch(fvalue, 0); return (index + 1) * 1000000 / TFRC_CALC_X_ARRSIZE; } - -EXPORT_SYMBOL_GPL(tfrc_calc_x_reverse_lookup); diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 741b2db24735..4152308958ab 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -1214,8 +1214,6 @@ const char *dccp_feat_typename(const u8 type) return NULL; } -EXPORT_SYMBOL_GPL(dccp_feat_typename); - const char *dccp_feat_name(const u8 feat) { static const char *feature_names[] = { @@ -1240,6 +1238,4 @@ const char *dccp_feat_name(const u8 feat) return feature_names[feat]; } - -EXPORT_SYMBOL_GPL(dccp_feat_name); #endif /* CONFIG_IP_DCCP_DEBUG */ diff --git a/net/dccp/input.c b/net/dccp/input.c index 5eb443f656c1..7648f316310f 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -741,5 +741,3 @@ u32 dccp_sample_rtt(struct sock *sk, long delta) return delta; } - -EXPORT_SYMBOL_GPL(dccp_sample_rtt); -- cgit v1.2.3 From 4f7d54f59bc470f0aaa932f747a95232d7ebf8b1 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 5 Jan 2009 00:00:12 -0800 Subject: tcp: don't mask EOF and socket errors on nonblocking splice receive Currently, setting SPLICE_F_NONBLOCK on splice from a TCP socket results in masking of EOF (RDHUP) and error conditions on the socket by an -EAGAIN return. Move the NONBLOCK check in tcp_splice_read() to be after the EOF and error checks to fix this. Signed-off-by: Lennert Buytenhek Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4d655e945413..bce1b068f2a7 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -580,10 +580,6 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, else if (!ret) { if (spliced) break; - if (flags & SPLICE_F_NONBLOCK) { - ret = -EAGAIN; - break; - } if (sock_flag(sk, SOCK_DONE)) break; if (sk->sk_err) { @@ -601,6 +597,10 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, ret = -ENOTCONN; break; } + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } if (!timeo) { ret = -EAGAIN; break; -- cgit v1.2.3 From 7945cc6464a4db0caf6dfacdfe05806051c4cb7b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 5 Jan 2009 00:59:00 -0800 Subject: tcp: Kill extraneous SPLICE_F_NONBLOCK checks. In splice TCP receive, the SPLICE_F_NONBLOCK flag is used to compute the "timeo" value. So checking it again inside of the main receive loop to trigger -EAGAIN processing is entirely unnecessary. Noticed by Jarek P. and Lennert Buytenhek. Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index bce1b068f2a7..35bcddf8a932 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -597,10 +597,6 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, ret = -ENOTCONN; break; } - if (flags & SPLICE_F_NONBLOCK) { - ret = -EAGAIN; - break; - } if (!timeo) { ret = -EAGAIN; break; -- cgit v1.2.3 From 56ff5efad96182f4d3cb3dc6b07396762c658f16 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 9 Dec 2008 09:34:39 -0500 Subject: zero i_uid/i_gid on inode allocation ... and don't bother in callers. Don't bother with zeroing i_blocks, while we are at it - it's already been zeroed. i_mode is not worth the effort; it has no common default value. Signed-off-by: Al Viro --- arch/powerpc/platforms/cell/spufs/inode.c | 1 - arch/s390/hypfs/inode.c | 1 - drivers/infiniband/hw/ipath/ipath_fs.c | 3 --- drivers/isdn/capi/capifs.c | 2 -- drivers/misc/ibmasm/ibmasmfs.c | 2 -- drivers/oprofile/oprofilefs.c | 3 --- drivers/usb/core/inode.c | 1 - drivers/usb/gadget/inode.c | 1 - fs/autofs/inode.c | 2 -- fs/autofs4/inode.c | 4 ---- fs/binfmt_misc.c | 3 --- fs/configfs/inode.c | 3 --- fs/cramfs/inode.c | 2 -- fs/debugfs/inode.c | 3 --- fs/devpts/inode.c | 4 ---- fs/hugetlbfs/inode.c | 1 - fs/inode.c | 2 ++ fs/libfs.c | 5 ----- fs/ocfs2/dlm/dlmfs.c | 2 -- fs/omfs/inode.c | 1 - fs/openpromfs/inode.c | 3 --- fs/proc/base.c | 4 ---- fs/proc/proc_sysctl.c | 1 - fs/ramfs/inode.c | 1 - fs/romfs/inode.c | 1 - fs/sysfs/inode.c | 3 --- ipc/mqueue.c | 1 - kernel/cgroup.c | 1 - net/sunrpc/rpc_pipe.c | 2 -- security/inode.c | 3 --- security/selinux/selinuxfs.c | 2 -- 31 files changed, 2 insertions(+), 66 deletions(-) (limited to 'net') diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 6296bfd9cb0b..e309ef70a531 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -97,7 +97,6 @@ spufs_new_inode(struct super_block *sb, int mode) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; out: return inode; diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 9d4f8e6c0800..5a805df216bb 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -106,7 +106,6 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode) ret->i_mode = mode; ret->i_uid = hypfs_info->uid; ret->i_gid = hypfs_info->gid; - ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; if (mode & S_IFDIR) ret->i_nlink = 2; diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c index 53912c327bfe..8dc2bb781605 100644 --- a/drivers/infiniband/hw/ipath/ipath_fs.c +++ b/drivers/infiniband/hw/ipath/ipath_fs.c @@ -57,9 +57,6 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry, } inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_private = data; if ((mode & S_IFMT) == S_IFDIR) { diff --git a/drivers/isdn/capi/capifs.c b/drivers/isdn/capi/capifs.c index 0aa66ec4cbdd..b129409925af 100644 --- a/drivers/isdn/capi/capifs.c +++ b/drivers/isdn/capi/capifs.c @@ -111,8 +111,6 @@ capifs_fill_super(struct super_block *s, void *data, int silent) goto fail; inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_blocks = 0; - inode->i_uid = inode->i_gid = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c index 22a7e8ba211d..de966a6fb7e6 100644 --- a/drivers/misc/ibmasm/ibmasmfs.c +++ b/drivers/misc/ibmasm/ibmasmfs.c @@ -146,8 +146,6 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode) if (ret) { ret->i_mode = mode; - ret->i_uid = ret->i_gid = 0; - ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; } return ret; diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c index ddc4c59f02dc..b7e4cee24269 100644 --- a/drivers/oprofile/oprofilefs.c +++ b/drivers/oprofile/oprofilefs.c @@ -29,9 +29,6 @@ static struct inode *oprofilefs_get_inode(struct super_block *sb, int mode) if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } return inode; diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c index 185be760833e..2a129cb7bb56 100644 --- a/drivers/usb/core/inode.c +++ b/drivers/usb/core/inode.c @@ -279,7 +279,6 @@ static struct inode *usbfs_get_inode (struct super_block *sb, int mode, dev_t de inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { default: diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index eeb26c0f88e5..317b48fdbf01 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -2001,7 +2001,6 @@ gadgetfs_make_inode (struct super_block *sb, inode->i_mode = mode; inode->i_uid = default_uid; inode->i_gid = default_gid; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_private = data; diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index c773680d5c60..e1734f2d6e26 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -251,13 +251,11 @@ struct inode *autofs_iget(struct super_block *sb, unsigned long ino) inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; inode->i_nlink = 2; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_blocks = 0; if (ino == AUTOFS_ROOT_INO) { inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &autofs_root_inode_operations; inode->i_fop = &autofs_root_operations; - inode->i_uid = inode->i_gid = 0; /* Changed in read_super */ goto done; } diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 7b19802cfef4..cfc23e53b6f4 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -455,11 +455,7 @@ struct inode *autofs4_get_inode(struct super_block *sb, if (sb->s_root) { inode->i_uid = sb->s_root->d_inode->i_uid; inode->i_gid = sb->s_root->d_inode->i_gid; - } else { - inode->i_uid = 0; - inode->i_gid = 0; } - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; if (S_ISDIR(inf->mode)) { diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index f2744ab4e5b3..e1158cb4fbd6 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -496,9 +496,6 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode) if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb); } diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 4803ccc94480..5d349d38e056 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -117,8 +117,6 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr) static inline void set_default_inode_attr(struct inode * inode, mode_t mode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } @@ -136,7 +134,6 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd) { struct inode * inode = new_inode(configfs_sb); if (inode) { - inode->i_blocks = 0; inode->i_mapping->a_ops = &configfs_aops; inode->i_mapping->backing_dev_info = &configfs_backing_dev_info; inode->i_op = &configfs_inode_operations; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index f40423eb1a14..a07338d2d140 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -83,8 +83,6 @@ static struct inode *get_cramfs_inode(struct super_block *sb, inode->i_op = &page_symlink_inode_operations; inode->i_data.a_ops = &cramfs_aops; } else { - inode->i_size = 0; - inode->i_blocks = 0; init_special_inode(inode, inode->i_mode, old_decode_dev(cramfs_inode->size)); } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 3dbe2169cf36..81ae9ea3c6e1 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -37,9 +37,6 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { default: diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index fff96e152c0c..5f3231b9633f 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -189,8 +189,6 @@ static int mknod_ptmx(struct super_block *sb) } inode->i_ino = 2; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; mode = S_IFCHR|opts->ptmxmode; @@ -300,8 +298,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent) goto free_fsi; inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_blocks = 0; - inode->i_uid = inode->i_gid = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 7d479ce3aceb..0ab0c6f5f438 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -506,7 +506,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, inode->i_mode = mode; inode->i_uid = uid; inode->i_gid = gid; - inode->i_blocks = 0; inode->i_mapping->a_ops = &hugetlbfs_aops; inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/inode.c b/fs/inode.c index 7de1cda92489..bd48e5e6d3e8 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -131,6 +131,8 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode) inode->i_op = &empty_iops; inode->i_fop = &empty_fops; inode->i_nlink = 1; + inode->i_uid = 0; + inode->i_gid = 0; atomic_set(&inode->i_writecount, 0); inode->i_size = 0; inode->i_blocks = 0; diff --git a/fs/libfs.c b/fs/libfs.c index e960a8321902..7de05f7ce746 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -231,7 +231,6 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, */ root->i_ino = 1; root->i_mode = S_IFDIR | S_IRUSR | S_IWUSR; - root->i_uid = root->i_gid = 0; root->i_atime = root->i_mtime = root->i_ctime = CURRENT_TIME; dentry = d_alloc(NULL, &d_name); if (!dentry) { @@ -436,8 +435,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files */ inode->i_ino = 1; inode->i_mode = S_IFDIR | 0755; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; @@ -464,8 +461,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files if (!inode) goto out; inode->i_mode = S_IFREG | files->mode; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_fop = files->ops; inode->i_ino = i; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 6f7a77d54020..1c9efb406a96 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -341,7 +341,6 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inc_nlink(inode); @@ -367,7 +366,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent, inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c index 6afe57c84f84..633e9dc972bb 100644 --- a/fs/omfs/inode.c +++ b/fs/omfs/inode.c @@ -39,7 +39,6 @@ struct inode *omfs_new_inode(struct inode *dir, int mode) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->a_ops = &omfs_aops; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index d41bdc784de4..ffcd04f0012c 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -256,9 +256,6 @@ found: break; } - inode->i_gid = 0; - inode->i_uid = 0; - d_add(dentry, inode); return NULL; } diff --git a/fs/proc/base.c b/fs/proc/base.c index cad92c1ac2b3..10fd5223d600 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1426,8 +1426,6 @@ static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_st if (!ei->pid) goto out_unlock; - inode->i_uid = 0; - inode->i_gid = 0; if (task_dumpable(task)) { rcu_read_lock(); cred = __task_cred(task); @@ -2349,8 +2347,6 @@ static struct dentry *proc_base_instantiate(struct inode *dir, if (!ei->pid) goto out_iput; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_mode = p->mode; if (S_ISDIR(inode->i_mode)) inode->i_nlink = 2; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 06ed10b7da9e..94fcfff6863a 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -31,7 +31,6 @@ static struct inode *proc_sys_make_inode(struct super_block *sb, inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */ inode->i_mode = table->mode; - inode->i_uid = inode->i_gid = 0; if (!table->child) { inode->i_mode |= S_IFREG; inode->i_op = &proc_sys_inode_operations; diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index a83a3518ae33..b7e6ac706b87 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -57,7 +57,6 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mapping->a_ops = &ramfs_aops; inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info; mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER); diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index 60d2f822e87b..c97d4c931715 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -524,7 +524,6 @@ romfs_iget(struct super_block *sb, unsigned long ino) i->i_size = be32_to_cpu(ri.size); i->i_mtime.tv_sec = i->i_atime.tv_sec = i->i_ctime.tv_sec = 0; i->i_mtime.tv_nsec = i->i_atime.tv_nsec = i->i_ctime.tv_nsec = 0; - i->i_uid = i->i_gid = 0; /* Precalculate the data offset */ ino = romfs_strnlen(i, ino+ROMFH_SIZE, ROMFS_MAXFN); diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index eb53c632f856..dfa3d94cfc74 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -107,8 +107,6 @@ int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) static inline void set_default_inode_attr(struct inode * inode, mode_t mode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } @@ -149,7 +147,6 @@ static void sysfs_init_inode(struct sysfs_dirent *sd, struct inode *inode) { struct bin_attribute *bin_attr; - inode->i_blocks = 0; inode->i_mapping->a_ops = &sysfs_aops; inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info; inode->i_op = &sysfs_inode_operations; diff --git a/ipc/mqueue.c b/ipc/mqueue.c index d9393f8e4c3e..41b72f02fa70 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -120,7 +120,6 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode, inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_mtime = inode->i_ctime = inode->i_atime = CURRENT_TIME; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 48348dde6d81..f7c5099a0572 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -573,7 +573,6 @@ static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_mapping->backing_dev_info = &cgroup_backing_dev_info; } diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 192453248870..577385a4a5dc 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -522,8 +522,6 @@ rpc_get_inode(struct super_block *sb, int mode) if (!inode) return NULL; inode->i_mode = mode; - inode->i_uid = inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch(mode & S_IFMT) { case S_IFDIR: diff --git a/security/inode.c b/security/inode.c index efea5a605466..007ef252dde7 100644 --- a/security/inode.c +++ b/security/inode.c @@ -61,9 +61,6 @@ static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev) if (inode) { inode->i_mode = mode; - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { default: diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c index e5520996a75b..8f612c8becb5 100644 --- a/security/selinux/selinuxfs.c +++ b/security/selinux/selinuxfs.c @@ -847,8 +847,6 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode) if (ret) { ret->i_mode = mode; - ret->i_uid = ret->i_gid = 0; - ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; } return ret; -- cgit v1.2.3 From c276e098d3ee33059b4a1c747354226cec58487c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 5 Jan 2009 16:01:51 -0800 Subject: Revert "net: Fix for initial link state in 2.6.28" This reverts commit 22604c866889c4b2e12b73cbf1683bda1b72a313. We can't fix this issue in this way, because we now can try to take the dev_base_lock rwlock as a writer in software interrupt context and that is not allowed without major surgery elsewhere. This initial link state problem needs to be solved in some other way. Signed-off-by: David S. Miller --- net/core/link_watch.c | 7 +------ net/sched/sch_generic.c | 4 ++++ 2 files changed, 5 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 1e401e12dc72..bf8f7af699d7 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -178,6 +178,7 @@ static void __linkwatch_run_queue(int urgent_only) */ clear_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state); + rfc2863_policy(dev); if (dev->flags & IFF_UP) { if (netif_carrier_ok(dev)) dev_activate(dev); @@ -214,12 +215,6 @@ void linkwatch_fire_event(struct net_device *dev) { bool urgent = linkwatch_urgent_event(dev); - rfc2863_policy(dev); - - /* Some drivers call netif_carrier_off early */ - if (dev->reg_state == NETREG_UNINITIALIZED) - return; - if (!test_and_set_bit(__LINK_STATE_LINKWATCH_PENDING, &dev->state)) { dev_hold(dev); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 23a8e6141a00..5f5efe4e6072 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -270,6 +270,8 @@ static void dev_watchdog_down(struct net_device *dev) void netif_carrier_on(struct net_device *dev) { if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; linkwatch_fire_event(dev); if (netif_running(dev)) __netdev_watchdog_up(dev); @@ -286,6 +288,8 @@ EXPORT_SYMBOL(netif_carrier_on); void netif_carrier_off(struct net_device *dev) { if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; linkwatch_fire_event(dev); } } -- cgit v1.2.3 From 55cdea9ed9cf2d76993e40ed7a1fc649a14db07c Mon Sep 17 00:00:00 2001 From: Hendrik Brueckner Date: Mon, 5 Jan 2009 18:07:07 -0800 Subject: af_iucv: New error return codes for connect() If the iucv_path_connect() call fails then return an error code that corresponds to the iucv_path_connect() failure condition; instead of returning -ECONNREFUSED for any failure. This helps to improve error handling for user space applications (e.g. inform the user that the z/VM guest is not authorized to connect to other guest virtual machines). The error return codes are based on those described in connect(2). Signed-off-by: Hendrik Brueckner Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/iucv/af_iucv.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index af3192d2a5a3..1077bc4e6e2a 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -494,7 +494,21 @@ static int iucv_sock_connect(struct socket *sock, struct sockaddr *addr, if (err) { iucv_path_free(iucv->path); iucv->path = NULL; - err = -ECONNREFUSED; + switch (err) { + case 0x0b: /* Target communicator is not logged on */ + err = -ENETUNREACH; + break; + case 0x0d: /* Max connections for this guest exceeded */ + case 0x0e: /* Max connections for target guest exceeded */ + err = -EAGAIN; + break; + case 0x0f: /* Missing IUCV authorization */ + err = -EACCES; + break; + default: + err = -ECONNREFUSED; + break; + } goto done; } -- cgit v1.2.3 From 18becbc5479f88d5adc218374ca62b8b93ec2545 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Mon, 5 Jan 2009 18:07:46 -0800 Subject: af_iucv: avoid left over IUCV connections from failing connects For certain types of AFIUCV socket connect failures IUCV connections are left over. Add some cleanup-statements to avoid cluttered IUCV connections. Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/iucv/af_iucv.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'net') diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 1077bc4e6e2a..6b5f193e5f48 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -521,6 +521,13 @@ static int iucv_sock_connect(struct socket *sock, struct sockaddr *addr, release_sock(sk); return -ECONNREFUSED; } + + if (err) { + iucv_path_sever(iucv->path, NULL); + iucv_path_free(iucv->path); + iucv->path = NULL; + } + done: release_sock(sk); return err; -- cgit v1.2.3 From 65dbd7c2778f1921ef1ee2a73e47a2a126fed30f Mon Sep 17 00:00:00 2001 From: Hendrik Brueckner Date: Mon, 5 Jan 2009 18:08:23 -0800 Subject: af_iucv: Free iucv path/socket in path_pending callback Free iucv path after iucv_path_sever() calls in iucv_callback_connreq() (path_pending() iucv callback). If iucv_path_accept() fails, free path and free/kill newly created socket. Signed-off-by: Hendrik Brueckner Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/iucv/af_iucv.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 6b5f193e5f48..eb8a2a0b6eb7 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1042,12 +1042,14 @@ static int iucv_callback_connreq(struct iucv_path *path, ASCEBC(user_data, sizeof(user_data)); if (sk->sk_state != IUCV_LISTEN) { err = iucv_path_sever(path, user_data); + iucv_path_free(path); goto fail; } /* Check for backlog size */ if (sk_acceptq_is_full(sk)) { err = iucv_path_sever(path, user_data); + iucv_path_free(path); goto fail; } @@ -1055,6 +1057,7 @@ static int iucv_callback_connreq(struct iucv_path *path, nsk = iucv_sock_alloc(NULL, SOCK_STREAM, GFP_ATOMIC); if (!nsk) { err = iucv_path_sever(path, user_data); + iucv_path_free(path); goto fail; } @@ -1078,6 +1081,8 @@ static int iucv_callback_connreq(struct iucv_path *path, err = iucv_path_accept(path, &af_iucv_handler, nuser_data, nsk); if (err) { err = iucv_path_sever(path, user_data); + iucv_path_free(path); + iucv_sock_kill(nsk); goto fail; } -- cgit v1.2.3 From f1d3e4dca3f8d4f55656477e83d0afe0ea7cbaed Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 5 Jan 2009 18:09:02 -0800 Subject: iucv: fix cpu hotplug If the iucv module is compiled in/loaded but no user is registered cpu hot remove doesn't work. Reason for that is that the iucv cpu hotplug notifier on CPU_DOWN_PREPARE checks if the iucv_buffer_cpumask would be empty after the corresponding bit would be cleared. However the bit was never set since iucv wasn't enable. That causes all cpu hot unplug operations to fail in this scenario. To fix this use iucv_path_table as an indicator wether iucv is enabled or not. Signed-off-by: Heiko Carstens Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/iucv/iucv.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index 8f57d4f4328a..032f61e98595 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -517,6 +517,7 @@ static int iucv_enable(void) size_t alloc_size; int cpu, rc; + get_online_cpus(); rc = -ENOMEM; alloc_size = iucv_max_pathid * sizeof(struct iucv_path); iucv_path_table = kzalloc(alloc_size, GFP_KERNEL); @@ -524,19 +525,17 @@ static int iucv_enable(void) goto out; /* Declare per cpu buffers. */ rc = -EIO; - get_online_cpus(); for_each_online_cpu(cpu) smp_call_function_single(cpu, iucv_declare_cpu, NULL, 1); if (cpus_empty(iucv_buffer_cpumask)) /* No cpu could declare an iucv buffer. */ - goto out_path; + goto out; put_online_cpus(); return 0; - -out_path: - put_online_cpus(); - kfree(iucv_path_table); out: + kfree(iucv_path_table); + iucv_path_table = NULL; + put_online_cpus(); return rc; } @@ -551,8 +550,9 @@ static void iucv_disable(void) { get_online_cpus(); on_each_cpu(iucv_retrieve_cpu, NULL, 1); - put_online_cpus(); kfree(iucv_path_table); + iucv_path_table = NULL; + put_online_cpus(); } static int __cpuinit iucv_cpu_notify(struct notifier_block *self, @@ -589,10 +589,14 @@ static int __cpuinit iucv_cpu_notify(struct notifier_block *self, case CPU_ONLINE_FROZEN: case CPU_DOWN_FAILED: case CPU_DOWN_FAILED_FROZEN: + if (!iucv_path_table) + break; smp_call_function_single(cpu, iucv_declare_cpu, NULL, 1); break; case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: + if (!iucv_path_table) + break; cpumask = iucv_buffer_cpumask; cpu_clear(cpu, cpumask); if (cpus_empty(cpumask)) -- cgit v1.2.3 From 6f57321422e0d359e83c978c2b03db77b967b7d5 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Mon, 5 Jan 2009 18:14:19 -0800 Subject: pkt_sched: cls_u32: Fix locking in u32_change() New nodes are inserted in u32_change() under rtnl_lock() with wmb(), so without tcf_tree_lock() like in other classifiers (e.g. cls_fw). This isn't enough without rmb() on the read side, but on the other hand adding such barriers doesn't give any savings, so the lock is added instead. Reported-by: m0sia Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/cls_u32.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 05d178008cbc..07372f60bee3 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -638,8 +638,9 @@ static int u32_change(struct tcf_proto *tp, unsigned long base, u32 handle, break; n->next = *ins; - wmb(); + tcf_tree_lock(tp); *ins = n; + tcf_tree_unlock(tp); *arg = (unsigned long)n; return 0; -- cgit v1.2.3 From 025dfdafe77f20b3890981a394774baab7b9c827 Mon Sep 17 00:00:00 2001 From: Frederik Schwarzer Date: Thu, 16 Oct 2008 19:02:37 +0200 Subject: trivial: fix then -> than typos in comments and documentation - (better, more, bigger ...) then -> (...) than Signed-off-by: Frederik Schwarzer Signed-off-by: Jiri Kosina --- Documentation/hwmon/abituguru-datasheet | 6 +++--- Documentation/networking/rxrpc.txt | 2 +- Documentation/scsi/ChangeLog.lpfc | 2 +- arch/blackfin/kernel/kgdb.c | 2 +- arch/ia64/kernel/kprobes.c | 2 +- arch/m68k/Kconfig | 2 +- arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c | 2 +- arch/powerpc/kernel/kprobes.c | 2 +- arch/powerpc/oprofile/cell/spu_profiler.c | 2 +- arch/s390/Kconfig | 2 +- arch/s390/kernel/kprobes.c | 2 +- arch/sparc/kernel/kprobes.c | 2 +- arch/x86/kernel/kprobes.c | 2 +- arch/x86/kernel/mfgpt_32.c | 2 +- drivers/hwmon/fschmd.c | 2 +- drivers/infiniband/hw/mlx4/cq.c | 2 +- drivers/message/i2o/i2o_scsi.c | 2 +- drivers/mtd/devices/pmc551.c | 2 +- drivers/mtd/ubi/eba.c | 2 +- drivers/mtd/ubi/io.c | 2 +- drivers/mtd/ubi/scan.c | 2 +- drivers/mtd/ubi/ubi-media.h | 4 ++-- drivers/mtd/ubi/vtbl.c | 2 +- drivers/mtd/ubi/wl.c | 4 ++-- drivers/net/bnx2x_link.c | 2 +- drivers/net/e1000/e1000_hw.c | 4 ++-- drivers/net/slip.h | 2 +- drivers/net/tehuti.c | 4 ++-- drivers/net/tokenring/smctr.c | 2 +- drivers/net/wireless/ipw2x00/ipw2100.c | 2 +- drivers/net/wireless/rt2x00/rt2x00crypto.c | 4 ++-- drivers/net/wireless/strip.c | 2 +- drivers/s390/block/dasd_eer.c | 4 ++-- drivers/s390/char/vmlogrdr.c | 4 ++-- drivers/scsi/lpfc/lpfc_hbadisc.c | 4 ++-- drivers/scsi/lpfc/lpfc_sli.c | 10 +++++----- drivers/serial/crisv10.c | 4 ++-- drivers/video/console/vgacon.c | 2 +- fs/ocfs2/cluster/heartbeat.c | 2 +- fs/proc/task_nommu.c | 2 +- fs/ubifs/Kconfig | 2 +- fs/ubifs/budget.c | 4 ++-- fs/ubifs/gc.c | 2 +- fs/ubifs/journal.c | 2 +- fs/ubifs/shrinker.c | 2 +- fs/xfs/linux-2.6/xfs_super.c | 2 +- include/linux/mtd/mtd.h | 2 +- include/linux/spi/spi.h | 4 ++-- include/mtd/ubi-user.h | 2 +- kernel/pid.c | 2 +- kernel/time/jiffies.c | 2 +- net/sctp/auth.c | 4 ++-- net/sctp/sm_statefuns.c | 6 +++--- net/sctp/socket.c | 2 +- net/sctp/tsnmap.c | 2 +- sound/usb/usx2y/usbusx2y.c | 2 +- 56 files changed, 76 insertions(+), 76 deletions(-) (limited to 'net') diff --git a/Documentation/hwmon/abituguru-datasheet b/Documentation/hwmon/abituguru-datasheet index aef5a9b36846..4d184f2db0ea 100644 --- a/Documentation/hwmon/abituguru-datasheet +++ b/Documentation/hwmon/abituguru-datasheet @@ -74,7 +74,7 @@ a sensor. Notice that some banks have both a read and a write address this is how the uGuru determines if a read from or a write to the bank is taking place, thus when reading you should always use the read address and when writing the -write address. The write address is always one (1) more then the read address. +write address. The write address is always one (1) more than the read address. uGuru ready @@ -224,7 +224,7 @@ Bit 3: Beep if alarm (RW) Bit 4: 1 if alarm cause measured temp is over the warning threshold (R) Bit 5: 1 if alarm cause measured volt is over the max threshold (R) Bit 6: 1 if alarm cause measured volt is under the min threshold (R) -Bit 7: Volt sensor: Shutdown if alarm persist for more then 4 seconds (RW) +Bit 7: Volt sensor: Shutdown if alarm persist for more than 4 seconds (RW) Temp sensor: Shutdown if temp is over the shutdown threshold (RW) * This bit is only honored/used by the uGuru if a temp sensor is connected @@ -293,7 +293,7 @@ Byte 0: Alarm behaviour for the selected sensor. A 1 enables the described behaviour. Bit 0: Give an alarm if measured rpm is under the min threshold (RW) Bit 3: Beep if alarm (RW) -Bit 7: Shutdown if alarm persist for more then 4 seconds (RW) +Bit 7: Shutdown if alarm persist for more than 4 seconds (RW) Byte 1: min threshold (scale as bank 0x26) diff --git a/Documentation/networking/rxrpc.txt b/Documentation/networking/rxrpc.txt index c3669a3fb4af..60d05eb77c64 100644 --- a/Documentation/networking/rxrpc.txt +++ b/Documentation/networking/rxrpc.txt @@ -540,7 +540,7 @@ A client would issue an operation by: MSG_MORE should be set in msghdr::msg_flags on all but the last part of the request. Multiple requests may be made simultaneously. - If a call is intended to go to a destination other then the default + If a call is intended to go to a destination other than the default specified through connect(), then msghdr::msg_name should be set on the first request message of that call. diff --git a/Documentation/scsi/ChangeLog.lpfc b/Documentation/scsi/ChangeLog.lpfc index ae3f962a7cfc..ff19a52fe004 100644 --- a/Documentation/scsi/ChangeLog.lpfc +++ b/Documentation/scsi/ChangeLog.lpfc @@ -733,7 +733,7 @@ Changes from 20040920 to 20041018 I/O completion path a little more, especially taking care of fast-pathing the non-error case. Also removes tons of dead members and defines from lpfc_scsi.h - e.g. lpfc_target is down - to nothing more then the lpfc_nodelist pointer. + to nothing more than the lpfc_nodelist pointer. * Added binary sysfs file to issue mbox commands * Replaced #if __BIG_ENDIAN with #if __BIG_ENDIAN_BITFIELD for compatibility with the user space applications. diff --git a/arch/blackfin/kernel/kgdb.c b/arch/blackfin/kernel/kgdb.c index b795a207742c..1c5afaeb9504 100644 --- a/arch/blackfin/kernel/kgdb.c +++ b/arch/blackfin/kernel/kgdb.c @@ -105,7 +105,7 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) * Extracts ebp, esp and eip values understandable by gdb from the values * saved by switch_to. * thread.esp points to ebp. flags and ebp are pushed in switch_to hence esp - * prior to entering switch_to is 8 greater then the value that is saved. + * prior to entering switch_to is 8 greater than the value that is saved. * If switch_to changes, change following code appropriately. */ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index f07688da947c..0017b9de2ddf 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -434,7 +434,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) /* * It is possible to have multiple instances associated with a given * task either because an multiple functions in the call path - * have a return probe installed on them, and/or more then one return + * have a return probe installed on them, and/or more than one return * return probe was registered for a target function. * * We can handle this because: diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index c825bde17cb3..fb87c08c6b57 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -303,7 +303,7 @@ config M68KFPU_EMU_EXTRAPREC correct rounding, the emulator can (often) do the same but this extra calculation can cost quite some time, so you can disable it here. The emulator will then "only" calculate with a 64 bit - mantissa and round slightly incorrect, what is more then enough + mantissa and round slightly incorrect, what is more than enough for normal usage. config M68KFPU_EMU_ONLY diff --git a/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c b/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c index 97862f45496d..caf5e9a0acc7 100644 --- a/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c +++ b/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c @@ -148,7 +148,7 @@ int read_eeprom(char *buffer, int eeprom_size, int size) send_byte(W_HEADER); recv_ack(); - /* EEPROM with size of more then 2K need two byte addressing */ + /* EEPROM with size of more than 2K need two byte addressing */ if (eeprom_size > 2048) { send_byte(0x00); recv_ack(); diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index de79915452c8..b29005a5a8f5 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -316,7 +316,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p, /* * It is possible to have multiple instances associated with a given * task either because an multiple functions in the call path - * have a return probe installed on them, and/or more then one return + * have a return probe installed on them, and/or more than one return * return probe was registered for a target function. * * We can handle this because: diff --git a/arch/powerpc/oprofile/cell/spu_profiler.c b/arch/powerpc/oprofile/cell/spu_profiler.c index dd499c3e9da7..83faa958b9d4 100644 --- a/arch/powerpc/oprofile/cell/spu_profiler.c +++ b/arch/powerpc/oprofile/cell/spu_profiler.c @@ -49,7 +49,7 @@ void set_spu_profiling_frequency(unsigned int freq_khz, unsigned int cycles_rese * of precision. This is close enough for the purpose at hand. * * The value of the timeout should be small enough that the hw - * trace buffer will not get more then about 1/3 full for the + * trace buffer will not get more than about 1/3 full for the * maximum user specified (the LFSR value) hw sampling frequency. * This is to ensure the trace buffer will never fill even if the * kernel thread scheduling varies under a heavy system load. diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 19577aeffd7b..a94a3c3ae932 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -299,7 +299,7 @@ config WARN_STACK This option enables the compiler options -mwarn-framesize and -mwarn-dynamicstack. If the compiler supports these options it will generate warnings for function which either use alloca or - create a stack frame bigger then CONFIG_WARN_STACK_SIZE. + create a stack frame bigger than CONFIG_WARN_STACK_SIZE. Say N if you are unsure. diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 569079ec4ff0..267f6698680a 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -381,7 +381,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p, /* * It is possible to have multiple instances associated with a given * task either because an multiple functions in the call path - * have a return probe installed on them, and/or more then one return + * have a return probe installed on them, and/or more than one return * return probe was registered for a target function. * * We can handle this because: diff --git a/arch/sparc/kernel/kprobes.c b/arch/sparc/kernel/kprobes.c index 201a6e547e4a..3bc6527c95af 100644 --- a/arch/sparc/kernel/kprobes.c +++ b/arch/sparc/kernel/kprobes.c @@ -517,7 +517,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) /* * It is possible to have multiple instances associated with a given * task either because an multiple functions in the call path - * have a return probe installed on them, and/or more then one return + * have a return probe installed on them, and/or more than one return * return probe was registered for a target function. * * We can handle this because: diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 6c27679ec6aa..a116e6d5726c 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -694,7 +694,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs) /* * It is possible to have multiple instances associated with a given * task either because multiple functions in the call path have - * return probes installed on them, and/or more then one + * return probes installed on them, and/or more than one * return probe was registered for a target function. * * We can handle this because: diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c index c12314c9e86f..8815f3c7fec7 100644 --- a/arch/x86/kernel/mfgpt_32.c +++ b/arch/x86/kernel/mfgpt_32.c @@ -252,7 +252,7 @@ EXPORT_SYMBOL_GPL(geode_mfgpt_alloc_timer); /* * The MFPGT timers on the CS5536 provide us with suitable timers to use * as clock event sources - not as good as a HPET or APIC, but certainly - * better then the PIT. This isn't a general purpose MFGPT driver, but + * better than the PIT. This isn't a general purpose MFGPT driver, but * a simplified one designed specifically to act as a clock event source. * For full details about the MFGPT, please consult the CS5536 data sheet. */ diff --git a/drivers/hwmon/fschmd.c b/drivers/hwmon/fschmd.c index 967170368933..8b2d756595d9 100644 --- a/drivers/hwmon/fschmd.c +++ b/drivers/hwmon/fschmd.c @@ -75,7 +75,7 @@ static const u8 FSCHMD_REG_VOLT[3] = { 0x45, 0x42, 0x48 }; /* minimum pwm at which the fan is driven (pwm can by increased depending on the temp. Notice that for the scy some fans share there minimum speed. - Also notice that with the scy the sensor order is different then with the + Also notice that with the scy the sensor order is different than with the other chips, this order was in the 2.4 driver and kept for consistency. */ static const u8 FSCHMD_REG_FAN_MIN[5][6] = { { 0x55, 0x65 }, /* pos */ diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index a3c5af1d7ec0..de5263beab4a 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -367,7 +367,7 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata) if (err) goto out; } else { - /* Can't be smaller then the number of outstanding CQEs */ + /* Can't be smaller than the number of outstanding CQEs */ outst_cqe = mlx4_ib_get_outstanding_cqes(cq); if (entries < outst_cqe + 1) { err = 0; diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index 1bcdbbb9e7d3..3d45817e6dcd 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -390,7 +390,7 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m, * @i2o_dev: the I2O device which was added * * If a I2O device is added we catch the notification, because I2O classes - * other then SCSI peripheral will not be received through + * other than SCSI peripheral will not be received through * i2o_scsi_probe(). */ static void i2o_scsi_notify_device_add(struct i2o_device *i2o_dev) diff --git a/drivers/mtd/devices/pmc551.c b/drivers/mtd/devices/pmc551.c index d38bca64bb15..d2fd550f7e09 100644 --- a/drivers/mtd/devices/pmc551.c +++ b/drivers/mtd/devices/pmc551.c @@ -34,7 +34,7 @@ * aperture size, not the dram size, and the V370PDC supplies no * other method for memory size discovery. This problem is * mostly only relevant when compiled as a module, as the - * unloading of the module with an aperture size smaller then + * unloading of the module with an aperture size smaller than * the ram will cause the driver to detect the onboard memory * size to be equal to the aperture size when the module is * reloaded. Soooo, to help, the module supports an msize diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 048a606cebde..25def348e5ba 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -717,7 +717,7 @@ write_error: * to the real data size, although the @buf buffer has to contain the * alignment. In all other cases, @len has to be aligned. * - * It is prohibited to write more then once to logical eraseblocks of static + * It is prohibited to write more than once to logical eraseblocks of static * volumes. This function returns zero in case of success and a negative error * code in case of failure. */ diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index a74118c05745..fe81039f2a7c 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -465,7 +465,7 @@ out: * This function synchronously erases physical eraseblock @pnum. If @torture * flag is not zero, the physical eraseblock is checked by means of writing * different patterns to it and reading them back. If the torturing is enabled, - * the physical eraseblock is erased more then once. + * the physical eraseblock is erased more than once. * * This function returns the number of erasures made in case of success, %-EIO * if the erasure failed or the torturing test failed, and other negative error diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 41d47e1cf15c..ecde202a5a12 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -478,7 +478,7 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, return 0; } else { /* - * This logical eraseblock is older then the one found + * This logical eraseblock is older than the one found * previously. */ if (cmp_res & 4) diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h index 2ad940409053..8419fdccc79c 100644 --- a/drivers/mtd/ubi/ubi-media.h +++ b/drivers/mtd/ubi/ubi-media.h @@ -135,7 +135,7 @@ enum { * The erase counter header takes 64 bytes and has a plenty of unused space for * future usage. The unused fields are zeroed. The @version field is used to * indicate the version of UBI implementation which is supposed to be able to - * work with this UBI image. If @version is greater then the current UBI + * work with this UBI image. If @version is greater than the current UBI * version, the image is rejected. This may be useful in future if something * is changed radically. This field is duplicated in the volume identifier * header. @@ -187,7 +187,7 @@ struct ubi_ec_hdr { * (sequence number) is used to distinguish between older and newer versions of * logical eraseblocks. * - * There are 2 situations when there may be more then one physical eraseblock + * There are 2 situations when there may be more than one physical eraseblock * corresponding to the same logical eraseblock, i.e., having the same @vol_id * and @lnum values in the volume identifier header. Suppose we have a logical * eraseblock L and it is mapped to the physical eraseblock P. diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 333c8941552f..1afc61e7455d 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -577,7 +577,7 @@ static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si, if (vtbl[i].flags & UBI_VTBL_AUTORESIZE_FLG) { /* Auto re-size flag may be set only for one volume */ if (ubi->autoresize_vol_id != -1) { - ubi_err("more then one auto-resize volume (%d " + ubi_err("more than one auto-resize volume (%d " "and %d)", ubi->autoresize_vol_id, i); kfree(vol); return -EINVAL; diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 14901cb82c18..891534f8210d 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -128,7 +128,7 @@ * situation when the picked physical eraseblock is constantly erased after the * data is written to it. So, we have a constant which limits the highest erase * counter of the free physical eraseblock to pick. Namely, the WL sub-system - * does not pick eraseblocks with erase counter greater then the lowest erase + * does not pick eraseblocks with erase counter greater than the lowest erase * counter plus %WL_FREE_MAX_DIFF. */ #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD) @@ -917,7 +917,7 @@ static int ensure_wear_leveling(struct ubi_device *ubi) /* * We schedule wear-leveling only if the difference between the * lowest erase counter of used physical eraseblocks and a high - * erase counter of free physical eraseblocks is greater then + * erase counter of free physical eraseblocks is greater than * %UBI_WL_THRESHOLD. */ e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb); diff --git a/drivers/net/bnx2x_link.c b/drivers/net/bnx2x_link.c index 67de94f1f30e..fefa6ab13064 100644 --- a/drivers/net/bnx2x_link.c +++ b/drivers/net/bnx2x_link.c @@ -3359,7 +3359,7 @@ static u8 bnx2x_format_ver(u32 num, u8 *str, u16 len) u8 shift = 8*4; u8 digit; if (len < 10) { - /* Need more then 10chars for this format */ + /* Need more than 10chars for this format */ *str_ptr = '\0'; return -EINVAL; } diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index d04eef53571e..e1a3fc1303ee 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -6758,7 +6758,7 @@ static s32 e1000_get_cable_length(struct e1000_hw *hw, u16 *min_length, * returns: - E1000_ERR_XXX * E1000_SUCCESS * - * For phy's older then IGP, this function simply reads the polarity bit in the + * For phy's older than IGP, this function simply reads the polarity bit in the * Phy Status register. For IGP phy's, this bit is valid only if link speed is * 10 Mbps. If the link speed is 100 Mbps there is no polarity so this bit will * return 0. If the link speed is 1000 Mbps the polarity status is in the @@ -6834,7 +6834,7 @@ static s32 e1000_check_polarity(struct e1000_hw *hw, * returns: - E1000_ERR_XXX * E1000_SUCCESS * - * For phy's older then IGP, this function reads the Downshift bit in the Phy + * For phy's older than IGP, this function reads the Downshift bit in the Phy * Specific Status register. For IGP phy's, it reads the Downgrade bit in the * Link Health register. In IGP this bit is latched high, so the driver must * read it immediately after link is established. diff --git a/drivers/net/slip.h b/drivers/net/slip.h index 853e0f6ec710..9ea5c11287d2 100644 --- a/drivers/net/slip.h +++ b/drivers/net/slip.h @@ -75,7 +75,7 @@ struct slip { unsigned long tx_errors; /* Planned stuff */ unsigned long rx_dropped; /* No memory for skb */ unsigned long tx_dropped; /* When MTU change */ - unsigned long rx_over_errors; /* Frame bigger then SLIP buf. */ + unsigned long rx_over_errors; /* Frame bigger than SLIP buf. */ #ifdef SL_INCLUDE_CSLIP unsigned long tx_compressed; unsigned long rx_compressed; diff --git a/drivers/net/tehuti.c b/drivers/net/tehuti.c index a10a83a11d9f..a7a4dc4d6313 100644 --- a/drivers/net/tehuti.c +++ b/drivers/net/tehuti.c @@ -1004,7 +1004,7 @@ static inline void bdx_rxdb_free_elem(struct rxdb *db, int n) * skb for rx. It assumes that Rx is desabled in HW * funcs are grouped for better cache usage * - * RxD fifo is smaller then RxF fifo by design. Upon high load, RxD will be + * RxD fifo is smaller than RxF fifo by design. Upon high load, RxD will be * filled and packets will be dropped by nic without getting into host or * cousing interrupt. Anyway, in that condition, host has no chance to proccess * all packets, but dropping in nic is cheaper, since it takes 0 cpu cycles @@ -1826,7 +1826,7 @@ static void bdx_tx_free(struct bdx_priv *priv) * * Pushes desc to TxD fifo and overlaps it if needed. * NOTE: this func does not check for available space. this is responsibility - * of the caller. Neither does it check that data size is smaller then + * of the caller. Neither does it check that data size is smaller than * fifo size. */ static void bdx_tx_push_desc(struct bdx_priv *priv, void *data, int size) diff --git a/drivers/net/tokenring/smctr.c b/drivers/net/tokenring/smctr.c index a011666342ff..50eb29ce3c87 100644 --- a/drivers/net/tokenring/smctr.c +++ b/drivers/net/tokenring/smctr.c @@ -3064,7 +3064,7 @@ static int smctr_load_node_addr(struct net_device *dev) * will consequently cause a timeout. * * NOTE 1: If the monitor_state is MS_BEACON_TEST_STATE, all transmit - * queues other then the one used for the lobe_media_test should be + * queues other than the one used for the lobe_media_test should be * disabled.!? * * NOTE 2: If the monitor_state is MS_BEACON_TEST_STATE and the receive_mask diff --git a/drivers/net/wireless/ipw2x00/ipw2100.c b/drivers/net/wireless/ipw2x00/ipw2100.c index 1667065b86a7..753de1a9c4b3 100644 --- a/drivers/net/wireless/ipw2x00/ipw2100.c +++ b/drivers/net/wireless/ipw2x00/ipw2100.c @@ -1332,7 +1332,7 @@ static int ipw2100_power_cycle_adapter(struct ipw2100_priv *priv) IPW_AUX_HOST_RESET_REG_STOP_MASTER); /* Step 2. Wait for stop Master Assert - * (not more then 50us, otherwise ret error */ + * (not more than 50us, otherwise ret error */ i = 5; do { udelay(IPW_WAIT_RESET_MASTER_ASSERT_COMPLETE_DELAY); diff --git a/drivers/net/wireless/rt2x00/rt2x00crypto.c b/drivers/net/wireless/rt2x00/rt2x00crypto.c index 37ad0d2fb64c..aee9cba13eb3 100644 --- a/drivers/net/wireless/rt2x00/rt2x00crypto.c +++ b/drivers/net/wireless/rt2x00/rt2x00crypto.c @@ -184,8 +184,8 @@ void rt2x00crypto_rx_insert_iv(struct sk_buff *skb, unsigned int align, * Make room for new data, note that we increase both * headsize and tailsize when required. The tailsize is * only needed when ICV data needs to be inserted and - * the padding is smaller then the ICV data. - * When alignment requirements is greater then the + * the padding is smaller than the ICV data. + * When alignment requirements is greater than the * ICV data we must trim the skb to the correct size * because we need to remove the extra bytes. */ diff --git a/drivers/net/wireless/strip.c b/drivers/net/wireless/strip.c index dd0de3a9ed4e..7015f2480550 100644 --- a/drivers/net/wireless/strip.c +++ b/drivers/net/wireless/strip.c @@ -236,7 +236,7 @@ struct strip { unsigned long tx_errors; /* Planned stuff */ unsigned long rx_dropped; /* No memory for skb */ unsigned long tx_dropped; /* When MTU change */ - unsigned long rx_over_errors; /* Frame bigger then STRIP buf. */ + unsigned long rx_over_errors; /* Frame bigger than STRIP buf. */ unsigned long pps_timer; /* Timer to determine pps */ unsigned long rx_pps_count; /* Counter to determine pps */ diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c index 892e2878d61b..f8e05ce98621 100644 --- a/drivers/s390/block/dasd_eer.c +++ b/drivers/s390/block/dasd_eer.c @@ -535,8 +535,8 @@ static int dasd_eer_open(struct inode *inp, struct file *filp) eerb->buffer_page_count > INT_MAX / PAGE_SIZE) { kfree(eerb); MESSAGE(KERN_WARNING, "can't open device since module " - "parameter eer_pages is smaller then 1 or" - " bigger then %d", (int)(INT_MAX / PAGE_SIZE)); + "parameter eer_pages is smaller than 1 or" + " bigger than %d", (int)(INT_MAX / PAGE_SIZE)); unlock_kernel(); return -EINVAL; } diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c index aabbeb909cc6..d8a2289fcb69 100644 --- a/drivers/s390/char/vmlogrdr.c +++ b/drivers/s390/char/vmlogrdr.c @@ -427,7 +427,7 @@ static int vmlogrdr_receive_data(struct vmlogrdr_priv_t *priv) buffer = priv->buffer + sizeof(int); } /* - * If the record is bigger then our buffer, we receive only + * If the record is bigger than our buffer, we receive only * a part of it. We can get the rest later. */ if (iucv_data_count > NET_BUFFER_SIZE) @@ -437,7 +437,7 @@ static int vmlogrdr_receive_data(struct vmlogrdr_priv_t *priv) 0, buffer, iucv_data_count, &priv->residual_length); spin_unlock_bh(&priv->priv_lock); - /* An rc of 5 indicates that the record was bigger then + /* An rc of 5 indicates that the record was bigger than * the buffer, which is OK for us. A 9 indicates that the * record was purged befor we could receive it. */ diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index 8c64494444bf..311ed6dea726 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -1964,10 +1964,10 @@ lpfc_set_disctmo(struct lpfc_vport *vport) uint32_t tmo; if (vport->port_state == LPFC_LOCAL_CFG_LINK) { - /* For FAN, timeout should be greater then edtov */ + /* For FAN, timeout should be greater than edtov */ tmo = (((phba->fc_edtov + 999) / 1000) + 1); } else { - /* Normal discovery timeout should be > then ELS/CT timeout + /* Normal discovery timeout should be > than ELS/CT timeout * FC spec states we need 3 * ratov for CT requests */ tmo = ((phba->fc_ratov * 3) + 3); diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 01dfdc8696f8..a36a120561e2 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -420,7 +420,7 @@ lpfc_sli_next_iocb_slot (struct lpfc_hba *phba, struct lpfc_sli_ring *pring) if (unlikely(pring->local_getidx >= max_cmd_idx)) { lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0315 Ring %d issue: portCmdGet %d " - "is bigger then cmd ring %d\n", + "is bigger than cmd ring %d\n", pring->ringno, pring->local_getidx, max_cmd_idx); @@ -1628,12 +1628,12 @@ lpfc_sli_rsp_pointers_error(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) { struct lpfc_pgp *pgp = &phba->port_gp[pring->ringno]; /* - * Ring handler: portRspPut is bigger then + * Ring handler: portRspPut is bigger than * rsp ring */ lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0312 Ring %d handler: portRspPut %d " - "is bigger then rsp ring %d\n", + "is bigger than rsp ring %d\n", pring->ringno, le32_to_cpu(pgp->rspPutInx), pring->numRiocb); @@ -2083,12 +2083,12 @@ lpfc_sli_handle_slow_ring_event(struct lpfc_hba *phba, portRspPut = le32_to_cpu(pgp->rspPutInx); if (portRspPut >= portRspMax) { /* - * Ring handler: portRspPut is bigger then + * Ring handler: portRspPut is bigger than * rsp ring */ lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0303 Ring %d handler: portRspPut %d " - "is bigger then rsp ring %d\n", + "is bigger than rsp ring %d\n", pring->ringno, portRspPut, portRspMax); phba->link_state = LPFC_HBA_ERROR; diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c index 8b2c619a09f2..e642c22c80e2 100644 --- a/drivers/serial/crisv10.c +++ b/drivers/serial/crisv10.c @@ -1203,7 +1203,7 @@ static void e100_disable_txdma_channel(struct e100_serial *info) unsigned long flags; /* Disable output DMA channel for the serial port in question - * ( set to something other then serialX) + * ( set to something other than serialX) */ local_irq_save(flags); DFLOW(DEBUG_LOG(info->line, "disable_txdma_channel %i\n", info->line)); @@ -1266,7 +1266,7 @@ static void e100_disable_rxdma_channel(struct e100_serial *info) unsigned long flags; /* Disable input DMA channel for the serial port in question - * ( set to something other then serialX) + * ( set to something other than serialX) */ local_irq_save(flags); if (info->line == 0) { diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index e6210725b9ab..d012edda6d11 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -1332,7 +1332,7 @@ static void vgacon_save_screen(struct vc_data *c) c->vc_y = screen_info.orig_y; } - /* We can't copy in more then the size of the video buffer, + /* We can't copy in more than the size of the video buffer, * or we'll be copying in VGA BIOS */ if (!vga_is_gfx) diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 6ebaa58e2c03..04697ba7f73e 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -854,7 +854,7 @@ static int o2hb_thread(void *data) while (!kthread_should_stop() && !reg->hr_unclean_stop) { /* We track the time spent inside - * o2hb_do_disk_heartbeat so that we avoid more then + * o2hb_do_disk_heartbeat so that we avoid more than * hr_timeout_ms between disk writes. On busy systems * this should result in a heartbeat which is less * likely to time itself out. */ diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 219bd79ea894..d4a8be32b902 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -9,7 +9,7 @@ /* * Logic: we've got two memory sums for each process, "shared", and - * "non-shared". Shared memory may get counted more then once, for + * "non-shared". Shared memory may get counted more than once, for * each process that owns it. Non-shared memory is counted * accurately. */ diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig index 91ceeda7e5bf..e35b54d5059d 100644 --- a/fs/ubifs/Kconfig +++ b/fs/ubifs/Kconfig @@ -40,7 +40,7 @@ config UBIFS_FS_ZLIB depends on UBIFS_FS default y help - Zlib copresses better then LZO but it is slower. Say 'Y' if unsure. + Zlib compresses better than LZO but it is slower. Say 'Y' if unsure. # Debugging-related stuff config UBIFS_FS_DEBUG diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 0e5e54d82924..175f9c590b77 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -142,7 +142,7 @@ static long long get_liability(struct ubifs_info *c) * * This function is called when an operation cannot be budgeted because there * is supposedly no free space. But in most cases there is some free space: - * o budgeting is pessimistic, so it always budgets more then it is actually + * o budgeting is pessimistic, so it always budgets more than it is actually * needed, so shrinking the liability is one way to make free space - the * cached data will take less space then it was budgeted for; * o GC may turn some dark space into free space (budgeting treats dark space @@ -606,7 +606,7 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) * @c: UBIFS file-system description object * * This function converts budget which was allocated for a new page of data to - * the budget of changing an existing page of data. The latter is smaller then + * the budget of changing an existing page of data. The latter is smaller than * the former, so this function only does simple re-calculation and does not * involve any write-back. */ diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index 0bef6501d58a..9832f9abe28e 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -45,7 +45,7 @@ #define SMALL_NODE_WM UBIFS_MAX_DENT_NODE_SZ /* - * GC may need to move more then one LEB to make progress. The below constants + * GC may need to move more than one LEB to make progress. The below constants * define "soft" and "hard" limits on the number of LEBs the garbage collector * may move. */ diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index 10ae25b7d1db..9b7c54e0cd2a 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -191,7 +191,7 @@ again: if (wbuf->lnum != -1 && avail >= len) { /* * Someone else has switched the journal head and we have - * enough space now. This happens when more then one process is + * enough space now. This happens when more than one process is * trying to write to the same journal head at the same time. */ dbg_jnl("return LEB %d back, already have LEB %d:%d", diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c index f248533841a2..e7bab52a1410 100644 --- a/fs/ubifs/shrinker.c +++ b/fs/ubifs/shrinker.c @@ -151,7 +151,7 @@ static int shrink_tnc(struct ubifs_info *c, int nr, int age, int *contention) * @contention: if any contention, this is set to %1 * * This function walks the list of mounted UBIFS file-systems and frees clean - * znodes which are older then @age, until at least @nr znodes are freed. + * znodes which are older than @age, until at least @nr znodes are freed. * Returns the number of freed znodes. */ static int shrink_tnc_trees(int nr, int age, int *contention) diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 36f6cc703ef2..be846d606ae8 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1348,7 +1348,7 @@ xfs_finish_flags( { int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); - /* Fail a mount where the logbuf is smaller then the log stripe */ + /* Fail a mount where the logbuf is smaller than the log stripe */ if (xfs_sb_version_haslogv2(&mp->m_sb)) { if (mp->m_logbsize <= 0 && mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) { diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index eae26bb6430a..64433eb411d7 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -83,7 +83,7 @@ typedef enum { * @datbuf: data buffer - if NULL only oob data are read/written * @oobbuf: oob data buffer * - * Note, it is allowed to read more then one OOB area at one go, but not write. + * Note, it is allowed to read more than one OOB area at one go, but not write. * The interface assumes that the OOB write requests program only one page's * OOB area. */ diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 82229317753d..68bb1c501d0d 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -327,9 +327,9 @@ extern struct spi_master *spi_busnum_to_master(u16 busnum); * @tx_dma: DMA address of tx_buf, if @spi_message.is_dma_mapped * @rx_dma: DMA address of rx_buf, if @spi_message.is_dma_mapped * @len: size of rx and tx buffers (in bytes) - * @speed_hz: Select a speed other then the device default for this + * @speed_hz: Select a speed other than the device default for this * transfer. If 0 the default (from @spi_device) is used. - * @bits_per_word: select a bits_per_word other then the device default + * @bits_per_word: select a bits_per_word other than the device default * for this transfer. If 0 the default (from @spi_device) is used. * @cs_change: affects chipselect after this transfer completes * @delay_usecs: microseconds to delay after this transfer before diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h index ccdc562e444e..2dc2eb2b8e22 100644 --- a/include/mtd/ubi-user.h +++ b/include/mtd/ubi-user.h @@ -253,7 +253,7 @@ struct ubi_mkvol_req { * * Re-sizing is possible for both dynamic and static volumes. But while dynamic * volumes may be re-sized arbitrarily, static volumes cannot be made to be - * smaller then the number of bytes they bear. To arbitrarily shrink a static + * smaller than the number of bytes they bear. To arbitrarily shrink a static * volume, it must be wiped out first (by means of volume update operation with * zero number of bytes). */ diff --git a/kernel/pid.c b/kernel/pid.c index 064e76afa507..af9224cdd6c0 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -475,7 +475,7 @@ pid_t task_session_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) EXPORT_SYMBOL(task_session_nr_ns); /* - * Used by proc to find the first pid that is greater then or equal to nr. + * Used by proc to find the first pid that is greater than or equal to nr. * * If there is a pid at nr this function is exactly the same as find_pid_ns. */ diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 1ca99557e929..06f197560f3b 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -45,7 +45,7 @@ * * The value 8 is somewhat carefully chosen, as anything * larger can result in overflows. NSEC_PER_JIFFY grows as - * HZ shrinks, so values greater then 8 overflow 32bits when + * HZ shrinks, so values greater than 8 overflow 32bits when * HZ=100. */ #define JIFFIES_SHIFT 8 diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 52db5f60daa0..20c576f530fa 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -141,8 +141,8 @@ void sctp_auth_destroy_keys(struct list_head *keys) /* Compare two byte vectors as numbers. Return values * are: * 0 - vectors are equal - * < 0 - vector 1 is smaller then vector2 - * > 0 - vector 1 is greater then vector2 + * < 0 - vector 1 is smaller than vector2 + * > 0 - vector 1 is greater than vector2 * * Algorithm is: * This is performed by selecting the numerically smaller key vector... diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 1c4e5d6c29c0..3a0cd075914f 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4268,9 +4268,9 @@ nomem: /* * Handle a protocol violation when the chunk length is invalid. - * "Invalid" length is identified as smaller then the minimal length a + * "Invalid" length is identified as smaller than the minimal length a * given chunk can be. For example, a SACK chunk has invalid length - * if it's length is set to be smaller then the size of sctp_sack_chunk_t. + * if its length is set to be smaller than the size of sctp_sack_chunk_t. * * We inform the other end by sending an ABORT with a Protocol Violation * error code. @@ -4300,7 +4300,7 @@ static sctp_disposition_t sctp_sf_violation_chunklen( /* * Handle a protocol violation when the parameter length is invalid. - * "Invalid" length is identified as smaller then the minimal length a + * "Invalid" length is identified as smaller than the minimal length a * given parameter can be. */ static sctp_disposition_t sctp_sf_violation_paramlen( diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b14a8f33e42d..ff0a8f88de04 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2717,7 +2717,7 @@ static int sctp_setsockopt_associnfo(struct sock *sk, char __user *optval, int o paths++; } - /* Only validate asocmaxrxt if we have more then + /* Only validate asocmaxrxt if we have more than * one path/transport. We do this because path * retransmissions are only counted when we have more * then one path. diff --git a/net/sctp/tsnmap.c b/net/sctp/tsnmap.c index 35c73e82553a..9bd64565021a 100644 --- a/net/sctp/tsnmap.c +++ b/net/sctp/tsnmap.c @@ -227,7 +227,7 @@ void sctp_tsnmap_skip(struct sctp_tsnmap *map, __u32 tsn) */ bitmap_zero(map->tsn_map, map->len); } else { - /* If the gap is smaller then the map size, + /* If the gap is smaller than the map size, * shift the map by 'gap' bits and update further. */ bitmap_shift_right(map->tsn_map, map->tsn_map, gap, map->len); diff --git a/sound/usb/usx2y/usbusx2y.c b/sound/usb/usx2y/usbusx2y.c index ca26c532e77e..11639bd72a51 100644 --- a/sound/usb/usx2y/usbusx2y.c +++ b/sound/usb/usx2y/usbusx2y.c @@ -238,7 +238,7 @@ static void i_usX2Y_In04Int(struct urb *urb) send = 0; for (j = 0; j < URBS_AsyncSeq && !err; ++j) if (0 == usX2Y->AS04.urb[j]->status) { - struct us428_p4out *p4out = us428ctls->p4out + send; // FIXME if more then 1 p4out is new, 1 gets lost. + struct us428_p4out *p4out = us428ctls->p4out + send; // FIXME if more than 1 p4out is new, 1 gets lost. usb_fill_bulk_urb(usX2Y->AS04.urb[j], usX2Y->chip.dev, usb_sndbulkpipe(usX2Y->chip.dev, 0x04), &p4out->val.vol, p4out->type == eLT_Light ? sizeof(struct us428_lights) : 5, -- cgit v1.2.3 From c9233eb7b0b11ef176d4bf68da2ce85464b6ec39 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 20 Oct 2008 11:51:57 -0400 Subject: sunrpc: add sv_maxconn field to svc_serv (try #3) svc_check_conn_limits() attempts to prevent denial of service attacks by having the service close old connections once it reaches a threshold. This threshold is based on the number of threads in the service: (serv->sv_nrthreads + 3) * 20 Once we reach this, we drop the oldest connections and a printk pops to warn the admin that they should increase the number of threads. Increasing the number of threads isn't an option however for services like lockd. We don't want to eliminate this check entirely for such services but we need some way to increase this limit. This patch adds a sv_maxconn field to the svc_serv struct. When it's set to 0, we use the current method to calculate the max number of connections. RPC services can then set this on an as-needed basis. Signed-off-by: Jeff Layton Acked-by: Neil Brown Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/svc.h | 5 ++++- net/sunrpc/svc_xprt.c | 22 ++++++++++++++++------ 2 files changed, 20 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 3afe7fb403b2..3435d24bfe55 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -58,10 +58,13 @@ struct svc_serv { struct svc_stat * sv_stats; /* RPC statistics */ spinlock_t sv_lock; unsigned int sv_nrthreads; /* # of server threads */ + unsigned int sv_maxconn; /* max connections allowed or + * '0' causing max to be based + * on number of threads. */ + unsigned int sv_max_payload; /* datagram payload size */ unsigned int sv_max_mesg; /* max_payload + 1 page for overheads */ unsigned int sv_xdrsize; /* XDR buffer size */ - struct list_head sv_permsocks; /* all permanent sockets */ struct list_head sv_tempsocks; /* all temporary sockets */ int sv_tmpcnt; /* count of temporary sockets */ diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index bf5b5cdafebf..3fe4f1004278 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -515,8 +515,10 @@ int svc_port_is_privileged(struct sockaddr *sin) } /* - * Make sure that we don't have too many active connections. If we - * have, something must be dropped. + * Make sure that we don't have too many active connections. If we have, + * something must be dropped. It's not clear what will happen if we allow + * "too many" connections, but when dealing with network-facing software, + * we have to code defensively. Here we do that by imposing hard limits. * * There's no point in trying to do random drop here for DoS * prevention. The NFS clients does 1 reconnect in 15 seconds. An @@ -525,19 +527,27 @@ int svc_port_is_privileged(struct sockaddr *sin) * The only somewhat efficient mechanism would be if drop old * connections from the same IP first. But right now we don't even * record the client IP in svc_sock. + * + * single-threaded services that expect a lot of clients will probably + * need to set sv_maxconn to override the default value which is based + * on the number of threads */ static void svc_check_conn_limits(struct svc_serv *serv) { - if (serv->sv_tmpcnt > (serv->sv_nrthreads+3)*20) { + unsigned int limit = serv->sv_maxconn ? serv->sv_maxconn : + (serv->sv_nrthreads+3) * 20; + + if (serv->sv_tmpcnt > limit) { struct svc_xprt *xprt = NULL; spin_lock_bh(&serv->sv_lock); if (!list_empty(&serv->sv_tempsocks)) { if (net_ratelimit()) { /* Try to help the admin */ printk(KERN_NOTICE "%s: too many open " - "connections, consider increasing the " - "number of nfsd threads\n", - serv->sv_name); + "connections, consider increasing %s\n", + serv->sv_name, serv->sv_maxconn ? + "the max number of connections." : + "the number of threads."); } /* * Always select the oldest connection. It's not fair, -- cgit v1.2.3 From 69b6ba3712b796a66595cfaf0a5ab4dfe1cf964a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 23 Dec 2008 16:30:11 -0500 Subject: SUNRPC: Ensure the server closes sockets in a timely fashion We want to ensure that connected sockets close down the connection when we set XPT_CLOSE, so that we don't keep it hanging while cleaning up all the stuff that is keeping a reference to the socket. Signed-off-by: Trond Myklebust Signed-off-by: J. Bruce Fields --- net/sunrpc/svcsock.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index ef3238d665ee..cccfa7deb9af 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -59,6 +59,7 @@ static void svc_udp_data_ready(struct sock *, int); static int svc_udp_recvfrom(struct svc_rqst *); static int svc_udp_sendto(struct svc_rqst *); static void svc_sock_detach(struct svc_xprt *); +static void svc_tcp_sock_detach(struct svc_xprt *); static void svc_sock_free(struct svc_xprt *); static struct svc_xprt *svc_create_socket(struct svc_serv *, int, @@ -1017,7 +1018,7 @@ static struct svc_xprt_ops svc_tcp_ops = { .xpo_recvfrom = svc_tcp_recvfrom, .xpo_sendto = svc_tcp_sendto, .xpo_release_rqst = svc_release_skb, - .xpo_detach = svc_sock_detach, + .xpo_detach = svc_tcp_sock_detach, .xpo_free = svc_sock_free, .xpo_prep_reply_hdr = svc_tcp_prep_reply_hdr, .xpo_has_wspace = svc_tcp_has_wspace, @@ -1287,6 +1288,24 @@ static void svc_sock_detach(struct svc_xprt *xprt) sk->sk_state_change = svsk->sk_ostate; sk->sk_data_ready = svsk->sk_odata; sk->sk_write_space = svsk->sk_owspace; + + if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) + wake_up_interruptible(sk->sk_sleep); +} + +/* + * Disconnect the socket, and reset the callbacks + */ +static void svc_tcp_sock_detach(struct svc_xprt *xprt) +{ + struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt); + + dprintk("svc: svc_tcp_sock_detach(%p)\n", svsk); + + svc_sock_detach(xprt); + + if (!test_bit(XPT_LISTENER, &xprt->xpt_flags)) + kernel_sock_shutdown(svsk->sk_sock, SHUT_RDWR); } /* -- cgit v1.2.3 From 6f49a57aa5a0c6d4e4e27c85f7af6c83325a12d1 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 6 Jan 2009 11:38:14 -0700 Subject: dmaengine: up-level reference counting to the module level Simply, if a client wants any dmaengine channel then prevent all dmaengine modules from being removed. Once the clients are done re-enable module removal. Why?, beyond reducing complication: 1/ Tracking reference counts per-transaction in an efficient manner, as is currently done, requires a complicated scheme to avoid cache-line bouncing effects. 2/ Per-transaction ref-counting gives the false impression that a dma-driver can be gracefully removed ahead of its user (net, md, or dma-slave) 3/ None of the in-tree dma-drivers talk to hot pluggable hardware, but if such an engine were built one day we still would not need to notify clients of remove events. The driver can simply return NULL to a ->prep() request, something that is much easier for a client to handle. Reviewed-by: Andrew Morton Acked-by: Maciej Sosnowski Signed-off-by: Dan Williams --- crypto/async_tx/async_tx.c | 4 - drivers/dma/dmaengine.c | 205 +++++++++++++++++++++++++++---------------- drivers/dma/dmatest.c | 2 - drivers/dma/dw_dmac.c | 2 +- drivers/mmc/host/atmel-mci.c | 4 +- include/linux/dmaengine.h | 21 ----- include/net/netdma.h | 4 +- net/ipv4/tcp.c | 1 - 8 files changed, 132 insertions(+), 111 deletions(-) (limited to 'net') diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c index 8cfac182165d..43fe4cbe71e6 100644 --- a/crypto/async_tx/async_tx.c +++ b/crypto/async_tx/async_tx.c @@ -198,8 +198,6 @@ dma_channel_add_remove(struct dma_client *client, /* add the channel to the generic management list */ master_ref = kmalloc(sizeof(*master_ref), GFP_KERNEL); if (master_ref) { - /* keep a reference until async_tx is unloaded */ - dma_chan_get(chan); init_dma_chan_ref(master_ref, chan); spin_lock_irqsave(&async_tx_lock, flags); list_add_tail_rcu(&master_ref->node, @@ -221,8 +219,6 @@ dma_channel_add_remove(struct dma_client *client, spin_lock_irqsave(&async_tx_lock, flags); list_for_each_entry(ref, &async_tx_master_list, node) if (ref->chan == chan) { - /* permit backing devices to go away */ - dma_chan_put(ref->chan); list_del_rcu(&ref->node); call_rcu(&ref->rcu, free_dma_chan_ref); found = 1; diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index b9008932a8f3..d4d925912c47 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -74,6 +74,7 @@ static DEFINE_MUTEX(dma_list_mutex); static LIST_HEAD(dma_device_list); static LIST_HEAD(dma_client_list); +static long dmaengine_ref_count; /* --- sysfs implementation --- */ @@ -105,19 +106,8 @@ static ssize_t show_bytes_transferred(struct device *dev, struct device_attribut static ssize_t show_in_use(struct device *dev, struct device_attribute *attr, char *buf) { struct dma_chan *chan = to_dma_chan(dev); - int in_use = 0; - - if (unlikely(chan->slow_ref) && - atomic_read(&chan->refcount.refcount) > 1) - in_use = 1; - else { - if (local_read(&(per_cpu_ptr(chan->local, - get_cpu())->refcount)) > 0) - in_use = 1; - put_cpu(); - } - return sprintf(buf, "%d\n", in_use); + return sprintf(buf, "%d\n", chan->client_count); } static struct device_attribute dma_attrs[] = { @@ -155,6 +145,78 @@ __dma_chan_satisfies_mask(struct dma_chan *chan, dma_cap_mask_t *want) return bitmap_equal(want->bits, has.bits, DMA_TX_TYPE_END); } +static struct module *dma_chan_to_owner(struct dma_chan *chan) +{ + return chan->device->dev->driver->owner; +} + +/** + * balance_ref_count - catch up the channel reference count + * @chan - channel to balance ->client_count versus dmaengine_ref_count + * + * balance_ref_count must be called under dma_list_mutex + */ +static void balance_ref_count(struct dma_chan *chan) +{ + struct module *owner = dma_chan_to_owner(chan); + + while (chan->client_count < dmaengine_ref_count) { + __module_get(owner); + chan->client_count++; + } +} + +/** + * dma_chan_get - try to grab a dma channel's parent driver module + * @chan - channel to grab + * + * Must be called under dma_list_mutex + */ +static int dma_chan_get(struct dma_chan *chan) +{ + int err = -ENODEV; + struct module *owner = dma_chan_to_owner(chan); + + if (chan->client_count) { + __module_get(owner); + err = 0; + } else if (try_module_get(owner)) + err = 0; + + if (err == 0) + chan->client_count++; + + /* allocate upon first client reference */ + if (chan->client_count == 1 && err == 0) { + int desc_cnt = chan->device->device_alloc_chan_resources(chan, NULL); + + if (desc_cnt < 0) { + err = desc_cnt; + chan->client_count = 0; + module_put(owner); + } else + balance_ref_count(chan); + } + + return err; +} + +/** + * dma_chan_put - drop a reference to a dma channel's parent driver module + * @chan - channel to release + * + * Must be called under dma_list_mutex + */ +static void dma_chan_put(struct dma_chan *chan) +{ + if (!chan->client_count) + return; /* this channel failed alloc_chan_resources */ + chan->client_count--; + module_put(dma_chan_to_owner(chan)); + if (chan->client_count == 0) + chan->device->device_free_chan_resources(chan); +} + /** * dma_client_chan_alloc - try to allocate channels to a client * @client: &dma_client @@ -165,7 +227,6 @@ static void dma_client_chan_alloc(struct dma_client *client) { struct dma_device *device; struct dma_chan *chan; - int desc; /* allocated descriptor count */ enum dma_state_client ack; /* Find a channel */ @@ -178,23 +239,16 @@ static void dma_client_chan_alloc(struct dma_client *client) list_for_each_entry(chan, &device->channels, device_node) { if (!dma_chan_satisfies_mask(chan, client->cap_mask)) continue; + if (!chan->client_count) + continue; + ack = client->event_callback(client, chan, + DMA_RESOURCE_AVAILABLE); - desc = chan->device->device_alloc_chan_resources( - chan, client); - if (desc >= 0) { - ack = client->event_callback(client, - chan, - DMA_RESOURCE_AVAILABLE); - - /* we are done once this client rejects - * an available resource - */ - if (ack == DMA_ACK) { - dma_chan_get(chan); - chan->client_count++; - } else if (ack == DMA_NAK) - return; - } + /* we are done once this client rejects + * an available resource + */ + if (ack == DMA_NAK) + return; } } } @@ -224,7 +278,6 @@ EXPORT_SYMBOL(dma_sync_wait); void dma_chan_cleanup(struct kref *kref) { struct dma_chan *chan = container_of(kref, struct dma_chan, refcount); - chan->device->device_free_chan_resources(chan); kref_put(&chan->device->refcount, dma_async_device_cleanup); } EXPORT_SYMBOL(dma_chan_cleanup); @@ -232,18 +285,12 @@ EXPORT_SYMBOL(dma_chan_cleanup); static void dma_chan_free_rcu(struct rcu_head *rcu) { struct dma_chan *chan = container_of(rcu, struct dma_chan, rcu); - int bias = 0x7FFFFFFF; - int i; - for_each_possible_cpu(i) - bias -= local_read(&per_cpu_ptr(chan->local, i)->refcount); - atomic_sub(bias, &chan->refcount.refcount); + kref_put(&chan->refcount, dma_chan_cleanup); } static void dma_chan_release(struct dma_chan *chan) { - atomic_add(0x7FFFFFFF, &chan->refcount.refcount); - chan->slow_ref = 1; call_rcu(&chan->rcu, dma_chan_free_rcu); } @@ -262,44 +309,37 @@ static void dma_clients_notify_available(void) mutex_unlock(&dma_list_mutex); } -/** - * dma_chans_notify_available - tell the clients that a channel is going away - * @chan: channel on its way out - */ -static void dma_clients_notify_removed(struct dma_chan *chan) -{ - struct dma_client *client; - enum dma_state_client ack; - - mutex_lock(&dma_list_mutex); - - list_for_each_entry(client, &dma_client_list, global_node) { - ack = client->event_callback(client, chan, - DMA_RESOURCE_REMOVED); - - /* client was holding resources for this channel so - * free it - */ - if (ack == DMA_ACK) { - dma_chan_put(chan); - chan->client_count--; - } - } - - mutex_unlock(&dma_list_mutex); -} - /** * dma_async_client_register - register a &dma_client * @client: ptr to a client structure with valid 'event_callback' and 'cap_mask' */ void dma_async_client_register(struct dma_client *client) { + struct dma_device *device, *_d; + struct dma_chan *chan; + int err; + /* validate client data */ BUG_ON(dma_has_cap(DMA_SLAVE, client->cap_mask) && !client->slave); mutex_lock(&dma_list_mutex); + dmaengine_ref_count++; + + /* try to grab channels */ + list_for_each_entry_safe(device, _d, &dma_device_list, global_node) + list_for_each_entry(chan, &device->channels, device_node) { + err = dma_chan_get(chan); + if (err == -ENODEV) { + /* module removed before we could use it */ + list_del_init(&device->global_node); + break; + } else if (err) + pr_err("dmaengine: failed to get %s: (%d)\n", + dev_name(&chan->dev), err); + } + + list_add_tail(&client->global_node, &dma_client_list); mutex_unlock(&dma_list_mutex); } @@ -315,23 +355,17 @@ void dma_async_client_unregister(struct dma_client *client) { struct dma_device *device; struct dma_chan *chan; - enum dma_state_client ack; if (!client) return; mutex_lock(&dma_list_mutex); - /* free all channels the client is holding */ + dmaengine_ref_count--; + BUG_ON(dmaengine_ref_count < 0); + /* drop channel references */ list_for_each_entry(device, &dma_device_list, global_node) - list_for_each_entry(chan, &device->channels, device_node) { - ack = client->event_callback(client, chan, - DMA_RESOURCE_REMOVED); - - if (ack == DMA_ACK) { - dma_chan_put(chan); - chan->client_count--; - } - } + list_for_each_entry(chan, &device->channels, device_node) + dma_chan_put(chan); list_del(&client->global_node); mutex_unlock(&dma_list_mutex); @@ -423,6 +457,21 @@ int dma_async_device_register(struct dma_device *device) } mutex_lock(&dma_list_mutex); + if (dmaengine_ref_count) + list_for_each_entry(chan, &device->channels, device_node) { + /* if clients are already waiting for channels we need + * to take references on their behalf + */ + if (dma_chan_get(chan) == -ENODEV) { + /* note we can only get here for the first + * channel as the remaining channels are + * guaranteed to get a reference + */ + rc = -ENODEV; + mutex_unlock(&dma_list_mutex); + goto err_out; + } + } list_add_tail(&device->global_node, &dma_device_list); mutex_unlock(&dma_list_mutex); @@ -456,7 +505,7 @@ static void dma_async_device_cleanup(struct kref *kref) } /** - * dma_async_device_unregister - unregisters DMA devices + * dma_async_device_unregister - unregister a DMA device * @device: &dma_device */ void dma_async_device_unregister(struct dma_device *device) @@ -468,7 +517,9 @@ void dma_async_device_unregister(struct dma_device *device) mutex_unlock(&dma_list_mutex); list_for_each_entry(chan, &device->channels, device_node) { - dma_clients_notify_removed(chan); + WARN_ONCE(chan->client_count, + "%s called while %d clients hold a reference\n", + __func__, chan->client_count); device_unregister(&chan->dev); dma_chan_release(chan); } diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c index ed9636bfb54a..db4050884713 100644 --- a/drivers/dma/dmatest.c +++ b/drivers/dma/dmatest.c @@ -215,7 +215,6 @@ static int dmatest_func(void *data) smp_rmb(); chan = thread->chan; - dma_chan_get(chan); while (!kthread_should_stop()) { total_tests++; @@ -293,7 +292,6 @@ static int dmatest_func(void *data) } ret = 0; - dma_chan_put(chan); kfree(thread->dstbuf); err_dstbuf: kfree(thread->srcbuf); diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c index 0778d99aea7c..377dafa37a20 100644 --- a/drivers/dma/dw_dmac.c +++ b/drivers/dma/dw_dmac.c @@ -773,7 +773,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan, dev_vdbg(&chan->dev, "alloc_chan_resources\n"); /* Channels doing slave DMA can only handle one client. */ - if (dwc->dws || client->slave) { + if (dwc->dws || (client && client->slave)) { if (chan->client_count) return -EBUSY; } diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c index 7a3f2436b011..6c11f4d4c4e9 100644 --- a/drivers/mmc/host/atmel-mci.c +++ b/drivers/mmc/host/atmel-mci.c @@ -593,10 +593,8 @@ atmci_submit_data_dma(struct atmel_mci *host, struct mmc_data *data) /* If we don't have a channel, we can't do DMA */ chan = host->dma.chan; - if (chan) { - dma_chan_get(chan); + if (chan) host->data_chan = chan; - } if (!chan) return -ENODEV; diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index e4ec7e7b8056..d18d37d1015d 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -165,7 +165,6 @@ struct dma_slave { */ struct dma_chan_percpu { - local_t refcount; /* stats */ unsigned long memcpy_count; unsigned long bytes_transferred; @@ -205,26 +204,6 @@ struct dma_chan { void dma_chan_cleanup(struct kref *kref); -static inline void dma_chan_get(struct dma_chan *chan) -{ - if (unlikely(chan->slow_ref)) - kref_get(&chan->refcount); - else { - local_inc(&(per_cpu_ptr(chan->local, get_cpu())->refcount)); - put_cpu(); - } -} - -static inline void dma_chan_put(struct dma_chan *chan) -{ - if (unlikely(chan->slow_ref)) - kref_put(&chan->refcount, dma_chan_cleanup); - else { - local_dec(&(per_cpu_ptr(chan->local, get_cpu())->refcount)); - put_cpu(); - } -} - /* * typedef dma_event_callback - function pointer to a DMA event callback * For each channel added to the system this routine is called for each client. diff --git a/include/net/netdma.h b/include/net/netdma.h index f28c6e064e8f..cbe2737f4a61 100644 --- a/include/net/netdma.h +++ b/include/net/netdma.h @@ -27,11 +27,11 @@ static inline struct dma_chan *get_softnet_dma(void) { struct dma_chan *chan; + rcu_read_lock(); chan = rcu_dereference(__get_cpu_var(softnet_data).net_dma); - if (chan) - dma_chan_get(chan); rcu_read_unlock(); + return chan; } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f28acf11fc67..75e0e0a2d8db 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1632,7 +1632,6 @@ skip_copy: /* Safe to free early-copied skbs now */ __skb_queue_purge(&sk->sk_async_wait_queue); - dma_chan_put(tp->ucopy.dma_chan); tp->ucopy.dma_chan = NULL; } if (tp->ucopy.pinned_list) { -- cgit v1.2.3 From 2ba05622b8b143b0c95968ba59bddfbd6d2f2559 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 6 Jan 2009 11:38:14 -0700 Subject: dmaengine: provide a common 'issue_pending_all' implementation async_tx and net_dma each have open-coded versions of issue_pending_all, so provide a common routine in dmaengine. The implementation needs to walk the global device list, so implement rcu to allow dma_issue_pending_all to run lockless. Clients protect themselves from channel removal events by holding a dmaengine reference. Reviewed-by: Andrew Morton Signed-off-by: Dan Williams --- crypto/async_tx/async_tx.c | 12 ------------ drivers/dma/dmaengine.c | 27 ++++++++++++++++++++++++--- include/linux/async_tx.h | 2 +- include/linux/dmaengine.h | 1 + net/core/dev.c | 9 +-------- 5 files changed, 27 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c index b88bb1f608fc..2cdf7a0867b7 100644 --- a/crypto/async_tx/async_tx.c +++ b/crypto/async_tx/async_tx.c @@ -45,18 +45,6 @@ static DEFINE_SPINLOCK(async_tx_lock); static LIST_HEAD(async_tx_master_list); -/* async_tx_issue_pending_all - start all transactions on all channels */ -void async_tx_issue_pending_all(void) -{ - struct dma_chan_ref *ref; - - rcu_read_lock(); - list_for_each_entry_rcu(ref, &async_tx_master_list, node) - ref->chan->device->device_issue_pending(ref->chan); - rcu_read_unlock(); -} -EXPORT_SYMBOL_GPL(async_tx_issue_pending_all); - static void free_dma_chan_ref(struct rcu_head *rcu) { diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 87a8cd4791ed..418eca28d472 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -70,6 +70,7 @@ #include #include #include +#include static DEFINE_MUTEX(dma_list_mutex); static LIST_HEAD(dma_device_list); @@ -365,6 +366,26 @@ struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type) } EXPORT_SYMBOL(dma_find_channel); +/** + * dma_issue_pending_all - flush all pending operations across all channels + */ +void dma_issue_pending_all(void) +{ + struct dma_device *device; + struct dma_chan *chan; + + WARN_ONCE(dmaengine_ref_count == 0, + "client called %s without a reference", __func__); + + rcu_read_lock(); + list_for_each_entry_rcu(device, &dma_device_list, global_node) + list_for_each_entry(chan, &device->channels, device_node) + if (chan->client_count) + device->device_issue_pending(chan); + rcu_read_unlock(); +} +EXPORT_SYMBOL(dma_issue_pending_all); + /** * nth_chan - returns the nth channel of the given capability * @cap: capability to match @@ -490,7 +511,7 @@ void dma_async_client_register(struct dma_client *client) err = dma_chan_get(chan); if (err == -ENODEV) { /* module removed before we could use it */ - list_del_init(&device->global_node); + list_del_rcu(&device->global_node); break; } else if (err) pr_err("dmaengine: failed to get %s: (%d)\n", @@ -635,7 +656,7 @@ int dma_async_device_register(struct dma_device *device) goto err_out; } } - list_add_tail(&device->global_node, &dma_device_list); + list_add_tail_rcu(&device->global_node, &dma_device_list); dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); @@ -677,7 +698,7 @@ void dma_async_device_unregister(struct dma_device *device) struct dma_chan *chan; mutex_lock(&dma_list_mutex); - list_del(&device->global_node); + list_del_rcu(&device->global_node); dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h index 1c816775f135..45f6297821bd 100644 --- a/include/linux/async_tx.h +++ b/include/linux/async_tx.h @@ -59,7 +59,7 @@ enum async_tx_flags { }; #ifdef CONFIG_DMA_ENGINE -void async_tx_issue_pending_all(void); +#define async_tx_issue_pending_all dma_issue_pending_all #ifdef CONFIG_ARCH_HAS_ASYNC_TX_FIND_CHANNEL #include #else diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index b466f02e2433..57a43adfc39e 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -471,6 +471,7 @@ int dma_async_device_register(struct dma_device *device); void dma_async_device_unregister(struct dma_device *device); void dma_run_dependencies(struct dma_async_tx_descriptor *tx); struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type); +void dma_issue_pending_all(void); /* --- Helper iov-locking functions --- */ diff --git a/net/core/dev.c b/net/core/dev.c index 09c66a449da6..e40b0d57f8ff 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2635,14 +2635,7 @@ out: * There may not be any more sk_buffs coming right now, so push * any pending DMA copies to hardware */ - if (!cpus_empty(net_dma.channel_mask)) { - int chan_idx; - for_each_cpu_mask_nr(chan_idx, net_dma.channel_mask) { - struct dma_chan *chan = net_dma.channels[chan_idx]; - if (chan) - dma_async_memcpy_issue_pending(chan); - } - } + dma_issue_pending_all(); #endif return; -- cgit v1.2.3 From f67b45999205164958de4ec0658d51fa4bee066d Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 6 Jan 2009 11:38:15 -0700 Subject: net_dma: convert to dma_find_channel Use the general-purpose channel allocation provided by dmaengine. Reviewed-by: Andrew Morton Signed-off-by: Dan Williams --- include/linux/netdevice.h | 3 --- include/net/netdma.h | 11 ----------- net/core/dev.c | 40 ---------------------------------------- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 7 files changed, 5 insertions(+), 59 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 41e1224651cf..bac2c458d9b8 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1113,9 +1113,6 @@ struct softnet_data struct sk_buff *completion_queue; struct napi_struct backlog; -#ifdef CONFIG_NET_DMA - struct dma_chan *net_dma; -#endif }; DECLARE_PER_CPU(struct softnet_data,softnet_data); diff --git a/include/net/netdma.h b/include/net/netdma.h index cbe2737f4a61..8ba8ce284eeb 100644 --- a/include/net/netdma.h +++ b/include/net/netdma.h @@ -24,17 +24,6 @@ #include #include -static inline struct dma_chan *get_softnet_dma(void) -{ - struct dma_chan *chan; - - rcu_read_lock(); - chan = rcu_dereference(__get_cpu_var(softnet_data).net_dma); - rcu_read_unlock(); - - return chan; -} - int dma_skb_copy_datagram_iovec(struct dma_chan* chan, struct sk_buff *skb, int offset, struct iovec *to, size_t len, struct dma_pinned_list *pinned_list); diff --git a/net/core/dev.c b/net/core/dev.c index e40b0d57f8ff..bbb07dbe1740 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4827,44 +4827,6 @@ static int dev_cpu_callback(struct notifier_block *nfb, } #ifdef CONFIG_NET_DMA -/** - * net_dma_rebalance - try to maintain one DMA channel per CPU - * @net_dma: DMA client and associated data (lock, channels, channel_mask) - * - * This is called when the number of channels allocated to the net_dma client - * changes. The net_dma client tries to have one DMA channel per CPU. - */ - -static void net_dma_rebalance(struct net_dma *net_dma) -{ - unsigned int cpu, i, n, chan_idx; - struct dma_chan *chan; - - if (cpus_empty(net_dma->channel_mask)) { - for_each_online_cpu(cpu) - rcu_assign_pointer(per_cpu(softnet_data, cpu).net_dma, NULL); - return; - } - - i = 0; - cpu = first_cpu(cpu_online_map); - - for_each_cpu_mask_nr(chan_idx, net_dma->channel_mask) { - chan = net_dma->channels[chan_idx]; - - n = ((num_online_cpus() / cpus_weight(net_dma->channel_mask)) - + (i < (num_online_cpus() % - cpus_weight(net_dma->channel_mask)) ? 1 : 0)); - - while(n) { - per_cpu(softnet_data, cpu).net_dma = chan; - cpu = next_cpu(cpu, cpu_online_map); - n--; - } - i++; - } -} - /** * netdev_dma_event - event callback for the net_dma_client * @client: should always be net_dma_client @@ -4894,7 +4856,6 @@ netdev_dma_event(struct dma_client *client, struct dma_chan *chan, ack = DMA_ACK; net_dma->channels[pos] = chan; cpu_set(pos, net_dma->channel_mask); - net_dma_rebalance(net_dma); } break; case DMA_RESOURCE_REMOVED: @@ -4909,7 +4870,6 @@ netdev_dma_event(struct dma_client *client, struct dma_chan *chan, ack = DMA_ACK; cpu_clear(pos, net_dma->channel_mask); net_dma->channels[i] = NULL; - net_dma_rebalance(net_dma); } break; default: diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 75e0e0a2d8db..9b275abc8eb9 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1317,7 +1317,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, if ((available < target) && (len > sysctl_tcp_dma_copybreak) && !(flags & MSG_PEEK) && !sysctl_tcp_low_latency && - __get_cpu_var(softnet_data).net_dma) { + dma_find_channel(DMA_MEMCPY)) { preempt_enable_no_resched(); tp->ucopy.pinned_list = dma_pin_iovec_pages(msg->msg_iov, len); @@ -1527,7 +1527,7 @@ do_prequeue: if (!(flags & MSG_TRUNC)) { #ifdef CONFIG_NET_DMA if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list) - tp->ucopy.dma_chan = get_softnet_dma(); + tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY); if (tp->ucopy.dma_chan) { tp->ucopy.dma_cookie = dma_skb_copy_datagram_iovec( diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 99b7ecbe8893..a6961d75c7ea 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5005,7 +5005,7 @@ static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, return 0; if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list) - tp->ucopy.dma_chan = get_softnet_dma(); + tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY); if (tp->ucopy.dma_chan && skb_csum_unnecessary(skb)) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 9d839fa9331e..19d7b429a262 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1594,7 +1594,7 @@ process: #ifdef CONFIG_NET_DMA struct tcp_sock *tp = tcp_sk(sk); if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list) - tp->ucopy.dma_chan = get_softnet_dma(); + tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY); if (tp->ucopy.dma_chan) ret = tcp_v4_do_rcv(sk, skb); else diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e8b8337a8310..71cd70951d7d 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1640,7 +1640,7 @@ process: #ifdef CONFIG_NET_DMA struct tcp_sock *tp = tcp_sk(sk); if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list) - tp->ucopy.dma_chan = get_softnet_dma(); + tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY); if (tp->ucopy.dma_chan) ret = tcp_v6_do_rcv(sk, skb); else -- cgit v1.2.3 From 209b84a88fe81341b4d8d465acc4a67cb7c3feb3 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 6 Jan 2009 11:38:17 -0700 Subject: dmaengine: replace dma_async_client_register with dmaengine_get Now that clients no longer need to be notified of channel arrival dma_async_client_register can simply increment the dmaengine_ref_count. Reviewed-by: Andrew Morton Signed-off-by: Dan Williams --- crypto/async_tx/async_tx.c | 115 +-------------------------------------------- drivers/dma/dmaengine.c | 22 +++------ include/linux/dmaengine.h | 4 +- net/core/dev.c | 3 +- 4 files changed, 11 insertions(+), 133 deletions(-) (limited to 'net') diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c index 2cdf7a0867b7..f21147f3626a 100644 --- a/crypto/async_tx/async_tx.c +++ b/crypto/async_tx/async_tx.c @@ -28,120 +28,9 @@ #include #ifdef CONFIG_DMA_ENGINE -static enum dma_state_client -dma_channel_add_remove(struct dma_client *client, - struct dma_chan *chan, enum dma_state state); - -static struct dma_client async_tx_dma = { - .event_callback = dma_channel_add_remove, - /* .cap_mask == 0 defaults to all channels */ -}; - -/** - * async_tx_lock - protect modification of async_tx_master_list and serialize - * rebalance operations - */ -static DEFINE_SPINLOCK(async_tx_lock); - -static LIST_HEAD(async_tx_master_list); - -static void -free_dma_chan_ref(struct rcu_head *rcu) -{ - struct dma_chan_ref *ref; - ref = container_of(rcu, struct dma_chan_ref, rcu); - kfree(ref); -} - -static void -init_dma_chan_ref(struct dma_chan_ref *ref, struct dma_chan *chan) -{ - INIT_LIST_HEAD(&ref->node); - INIT_RCU_HEAD(&ref->rcu); - ref->chan = chan; - atomic_set(&ref->count, 0); -} - -static enum dma_state_client -dma_channel_add_remove(struct dma_client *client, - struct dma_chan *chan, enum dma_state state) -{ - unsigned long found, flags; - struct dma_chan_ref *master_ref, *ref; - enum dma_state_client ack = DMA_DUP; /* default: take no action */ - - switch (state) { - case DMA_RESOURCE_AVAILABLE: - found = 0; - rcu_read_lock(); - list_for_each_entry_rcu(ref, &async_tx_master_list, node) - if (ref->chan == chan) { - found = 1; - break; - } - rcu_read_unlock(); - - pr_debug("async_tx: dma resource available [%s]\n", - found ? "old" : "new"); - - if (!found) - ack = DMA_ACK; - else - break; - - /* add the channel to the generic management list */ - master_ref = kmalloc(sizeof(*master_ref), GFP_KERNEL); - if (master_ref) { - init_dma_chan_ref(master_ref, chan); - spin_lock_irqsave(&async_tx_lock, flags); - list_add_tail_rcu(&master_ref->node, - &async_tx_master_list); - spin_unlock_irqrestore(&async_tx_lock, - flags); - } else { - printk(KERN_WARNING "async_tx: unable to create" - " new master entry in response to" - " a DMA_RESOURCE_ADDED event" - " (-ENOMEM)\n"); - return 0; - } - break; - case DMA_RESOURCE_REMOVED: - found = 0; - spin_lock_irqsave(&async_tx_lock, flags); - list_for_each_entry(ref, &async_tx_master_list, node) - if (ref->chan == chan) { - list_del_rcu(&ref->node); - call_rcu(&ref->rcu, free_dma_chan_ref); - found = 1; - break; - } - spin_unlock_irqrestore(&async_tx_lock, flags); - - pr_debug("async_tx: dma resource removed [%s]\n", - found ? "ours" : "not ours"); - - if (found) - ack = DMA_ACK; - else - break; - break; - case DMA_RESOURCE_SUSPEND: - case DMA_RESOURCE_RESUME: - printk(KERN_WARNING "async_tx: does not support dma channel" - " suspend/resume\n"); - break; - default: - BUG(); - } - - return ack; -} - static int __init async_tx_init(void) { - dma_async_client_register(&async_tx_dma); - dma_async_client_chan_request(&async_tx_dma); + dmaengine_get(); printk(KERN_INFO "async_tx: api initialized (async)\n"); @@ -150,7 +39,7 @@ static int __init async_tx_init(void) static void __exit async_tx_exit(void) { - dma_async_client_unregister(&async_tx_dma); + dmaengine_put(); } /** diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 90aca505a1df..3f1849b7f5ef 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -600,10 +600,9 @@ static void dma_clients_notify_available(void) } /** - * dma_async_client_register - register a &dma_client - * @client: ptr to a client structure with valid 'event_callback' and 'cap_mask' + * dmaengine_get - register interest in dma_channels */ -void dma_async_client_register(struct dma_client *client) +void dmaengine_get(void) { struct dma_device *device, *_d; struct dma_chan *chan; @@ -634,25 +633,18 @@ void dma_async_client_register(struct dma_client *client) */ if (dmaengine_ref_count == 1) dma_channel_rebalance(); - list_add_tail(&client->global_node, &dma_client_list); mutex_unlock(&dma_list_mutex); } -EXPORT_SYMBOL(dma_async_client_register); +EXPORT_SYMBOL(dmaengine_get); /** - * dma_async_client_unregister - unregister a client and free the &dma_client - * @client: &dma_client to free - * - * Force frees any allocated DMA channels, frees the &dma_client memory + * dmaengine_put - let dma drivers be removed when ref_count == 0 */ -void dma_async_client_unregister(struct dma_client *client) +void dmaengine_put(void) { struct dma_device *device; struct dma_chan *chan; - if (!client) - return; - mutex_lock(&dma_list_mutex); dmaengine_ref_count--; BUG_ON(dmaengine_ref_count < 0); @@ -663,11 +655,9 @@ void dma_async_client_unregister(struct dma_client *client) list_for_each_entry(chan, &device->channels, device_node) dma_chan_put(chan); } - - list_del(&client->global_node); mutex_unlock(&dma_list_mutex); } -EXPORT_SYMBOL(dma_async_client_unregister); +EXPORT_SYMBOL(dmaengine_put); /** * dma_async_client_chan_request - send all available channels to the diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index d63544cf8a1a..37d95db156d3 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -318,8 +318,8 @@ struct dma_device { /* --- public DMA engine API --- */ -void dma_async_client_register(struct dma_client *client); -void dma_async_client_unregister(struct dma_client *client); +void dmaengine_get(void); +void dmaengine_put(void); void dma_async_client_chan_request(struct dma_client *client); dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest, void *src, size_t len); diff --git a/net/core/dev.c b/net/core/dev.c index bbb07dbe1740..7596fc9403c8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4894,8 +4894,7 @@ static int __init netdev_dma_register(void) } spin_lock_init(&net_dma.lock); dma_cap_set(DMA_MEMCPY, net_dma.client.cap_mask); - dma_async_client_register(&net_dma.client); - dma_async_client_chan_request(&net_dma.client); + dmaengine_get(); return 0; } -- cgit v1.2.3 From aa1e6f1a385eb2b04171ec841f3b760091e4a8ee Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 6 Jan 2009 11:38:17 -0700 Subject: dmaengine: kill struct dma_client and supporting infrastructure All users have been converted to either the general-purpose allocator, dma_find_channel, or dma_request_channel. Reviewed-by: Andrew Morton Signed-off-by: Dan Williams --- drivers/dma/dmaengine.c | 74 ++------------------------------- drivers/dma/dw_dmac.c | 3 +- drivers/dma/fsldma.c | 3 +- drivers/dma/ioat_dma.c | 5 +-- drivers/dma/iop-adma.c | 7 ++-- drivers/dma/mv_xor.c | 7 ++-- drivers/mmc/host/atmel-mci.c | 1 - include/linux/dmaengine.h | 50 +--------------------- net/core/dev.c | 99 ++------------------------------------------ 9 files changed, 17 insertions(+), 232 deletions(-) (limited to 'net') diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 3f1849b7f5ef..9fc91f973a9a 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -31,15 +31,12 @@ * * LOCKING: * - * The subsystem keeps two global lists, dma_device_list and dma_client_list. - * Both of these are protected by a mutex, dma_list_mutex. + * The subsystem keeps a global list of dma_device structs it is protected by a + * mutex, dma_list_mutex. * * Each device has a channels list, which runs unlocked but is never modified * once the device is registered, it's just setup by the driver. * - * Each client is responsible for keeping track of the channels it uses. See - * the definition of dma_event_callback in dmaengine.h. - * * Each device has a kref, which is initialized to 1 when the device is * registered. A kref_get is done for each device registered. When the * device is released, the corresponding kref_put is done in the release @@ -74,7 +71,6 @@ static DEFINE_MUTEX(dma_list_mutex); static LIST_HEAD(dma_device_list); -static LIST_HEAD(dma_client_list); static long dmaengine_ref_count; /* --- sysfs implementation --- */ @@ -189,7 +185,7 @@ static int dma_chan_get(struct dma_chan *chan) /* allocate upon first client reference */ if (chan->client_count == 1 && err == 0) { - int desc_cnt = chan->device->device_alloc_chan_resources(chan, NULL); + int desc_cnt = chan->device->device_alloc_chan_resources(chan); if (desc_cnt < 0) { err = desc_cnt; @@ -218,40 +214,6 @@ static void dma_chan_put(struct dma_chan *chan) chan->device->device_free_chan_resources(chan); } -/** - * dma_client_chan_alloc - try to allocate channels to a client - * @client: &dma_client - * - * Called with dma_list_mutex held. - */ -static void dma_client_chan_alloc(struct dma_client *client) -{ - struct dma_device *device; - struct dma_chan *chan; - enum dma_state_client ack; - - /* Find a channel */ - list_for_each_entry(device, &dma_device_list, global_node) { - if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) - continue; - if (!dma_device_satisfies_mask(device, client->cap_mask)) - continue; - - list_for_each_entry(chan, &device->channels, device_node) { - if (!chan->client_count) - continue; - ack = client->event_callback(client, chan, - DMA_RESOURCE_AVAILABLE); - - /* we are done once this client rejects - * an available resource - */ - if (ack == DMA_NAK) - return; - } - } -} - enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie) { enum dma_status status; @@ -584,21 +546,6 @@ void dma_release_channel(struct dma_chan *chan) } EXPORT_SYMBOL_GPL(dma_release_channel); -/** - * dma_chans_notify_available - broadcast available channels to the clients - */ -static void dma_clients_notify_available(void) -{ - struct dma_client *client; - - mutex_lock(&dma_list_mutex); - - list_for_each_entry(client, &dma_client_list, global_node) - dma_client_chan_alloc(client); - - mutex_unlock(&dma_list_mutex); -} - /** * dmaengine_get - register interest in dma_channels */ @@ -659,19 +606,6 @@ void dmaengine_put(void) } EXPORT_SYMBOL(dmaengine_put); -/** - * dma_async_client_chan_request - send all available channels to the - * client that satisfy the capability mask - * @client - requester - */ -void dma_async_client_chan_request(struct dma_client *client) -{ - mutex_lock(&dma_list_mutex); - dma_client_chan_alloc(client); - mutex_unlock(&dma_list_mutex); -} -EXPORT_SYMBOL(dma_async_client_chan_request); - /** * dma_async_device_register - registers DMA devices found * @device: &dma_device @@ -765,8 +699,6 @@ int dma_async_device_register(struct dma_device *device) dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); - dma_clients_notify_available(); - return 0; err_out: diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c index dbd50804e5d2..a29dda8f801b 100644 --- a/drivers/dma/dw_dmac.c +++ b/drivers/dma/dw_dmac.c @@ -758,8 +758,7 @@ static void dwc_issue_pending(struct dma_chan *chan) spin_unlock_bh(&dwc->lock); } -static int dwc_alloc_chan_resources(struct dma_chan *chan, - struct dma_client *client) +static int dwc_alloc_chan_resources(struct dma_chan *chan) { struct dw_dma_chan *dwc = to_dw_dma_chan(chan); struct dw_dma *dw = to_dw_dma(chan->device); diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c index 0b95dcce447e..46e0128929a0 100644 --- a/drivers/dma/fsldma.c +++ b/drivers/dma/fsldma.c @@ -366,8 +366,7 @@ static struct fsl_desc_sw *fsl_dma_alloc_descriptor( * * Return - The number of descriptors allocated. */ -static int fsl_dma_alloc_chan_resources(struct dma_chan *chan, - struct dma_client *client) +static int fsl_dma_alloc_chan_resources(struct dma_chan *chan) { struct fsl_dma_chan *fsl_chan = to_fsl_chan(chan); diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c index 6607fdd00b1c..e42e1aea0f18 100644 --- a/drivers/dma/ioat_dma.c +++ b/drivers/dma/ioat_dma.c @@ -734,8 +734,7 @@ static void ioat2_dma_massage_chan_desc(struct ioat_dma_chan *ioat_chan) * ioat_dma_alloc_chan_resources - returns the number of allocated descriptors * @chan: the channel to be filled out */ -static int ioat_dma_alloc_chan_resources(struct dma_chan *chan, - struct dma_client *client) +static int ioat_dma_alloc_chan_resources(struct dma_chan *chan) { struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan); struct ioat_desc_sw *desc; @@ -1381,7 +1380,7 @@ static int ioat_dma_self_test(struct ioatdma_device *device) dma_chan = container_of(device->common.channels.next, struct dma_chan, device_node); - if (device->common.device_alloc_chan_resources(dma_chan, NULL) < 1) { + if (device->common.device_alloc_chan_resources(dma_chan) < 1) { dev_err(&device->pdev->dev, "selftest cannot allocate chan resource\n"); err = -ENODEV; diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c index be9ea9f88805..c74ac9eb009a 100644 --- a/drivers/dma/iop-adma.c +++ b/drivers/dma/iop-adma.c @@ -470,8 +470,7 @@ static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan); * greater than 2x the number slots needed to satisfy a device->max_xor * request. * */ -static int iop_adma_alloc_chan_resources(struct dma_chan *chan, - struct dma_client *client) +static int iop_adma_alloc_chan_resources(struct dma_chan *chan) { char *hw_desc; int idx; @@ -865,7 +864,7 @@ static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device) dma_chan = container_of(device->common.channels.next, struct dma_chan, device_node); - if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) { + if (iop_adma_alloc_chan_resources(dma_chan) < 1) { err = -ENODEV; goto out; } @@ -963,7 +962,7 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device) dma_chan = container_of(device->common.channels.next, struct dma_chan, device_node); - if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) { + if (iop_adma_alloc_chan_resources(dma_chan) < 1) { err = -ENODEV; goto out; } diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index 3f46df3390c7..fbaa2f6225e2 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -606,8 +606,7 @@ submit_done: } /* returns the number of allocated descriptors */ -static int mv_xor_alloc_chan_resources(struct dma_chan *chan, - struct dma_client *client) +static int mv_xor_alloc_chan_resources(struct dma_chan *chan) { char *hw_desc; int idx; @@ -957,7 +956,7 @@ static int __devinit mv_xor_memcpy_self_test(struct mv_xor_device *device) dma_chan = container_of(device->common.channels.next, struct dma_chan, device_node); - if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) { + if (mv_xor_alloc_chan_resources(dma_chan) < 1) { err = -ENODEV; goto out; } @@ -1052,7 +1051,7 @@ mv_xor_xor_self_test(struct mv_xor_device *device) dma_chan = container_of(device->common.channels.next, struct dma_chan, device_node); - if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) { + if (mv_xor_alloc_chan_resources(dma_chan) < 1) { err = -ENODEV; goto out; } diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c index 7a34118507db..4b567a0408e1 100644 --- a/drivers/mmc/host/atmel-mci.c +++ b/drivers/mmc/host/atmel-mci.c @@ -55,7 +55,6 @@ enum atmel_mci_state { struct atmel_mci_dma { #ifdef CONFIG_MMC_ATMELMCI_DMA - struct dma_client client; struct dma_chan *chan; struct dma_async_tx_descriptor *data_desc; #endif diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 37d95db156d3..db050e97d2b4 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -28,20 +28,6 @@ #include #include -/** - * enum dma_state - resource PNP/power management state - * @DMA_RESOURCE_SUSPEND: DMA device going into low power state - * @DMA_RESOURCE_RESUME: DMA device returning to full power - * @DMA_RESOURCE_AVAILABLE: DMA device available to the system - * @DMA_RESOURCE_REMOVED: DMA device removed from the system - */ -enum dma_state { - DMA_RESOURCE_SUSPEND, - DMA_RESOURCE_RESUME, - DMA_RESOURCE_AVAILABLE, - DMA_RESOURCE_REMOVED, -}; - /** * enum dma_state_client - state of the channel in the client * @DMA_ACK: client would like to use, or was using this channel @@ -170,23 +156,6 @@ struct dma_chan { void dma_chan_cleanup(struct kref *kref); -/* - * typedef dma_event_callback - function pointer to a DMA event callback - * For each channel added to the system this routine is called for each client. - * If the client would like to use the channel it returns '1' to signal (ack) - * the dmaengine core to take out a reference on the channel and its - * corresponding device. A client must not 'ack' an available channel more - * than once. When a channel is removed all clients are notified. If a client - * is using the channel it must 'ack' the removal. A client must not 'ack' a - * removed channel more than once. - * @client - 'this' pointer for the client context - * @chan - channel to be acted upon - * @state - available or removed - */ -struct dma_client; -typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client, - struct dma_chan *chan, enum dma_state state); - /** * typedef dma_filter_fn - callback filter for dma_request_channel * @chan: channel to be reviewed @@ -199,21 +168,6 @@ typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client, */ typedef enum dma_state_client (*dma_filter_fn)(struct dma_chan *chan, void *filter_param); -/** - * struct dma_client - info on the entity making use of DMA services - * @event_callback: func ptr to call when something happens - * @cap_mask: only return channels that satisfy the requested capabilities - * a value of zero corresponds to any capability - * @slave: data for preparing slave transfer. Must be non-NULL iff the - * DMA_SLAVE capability is requested. - * @global_node: list_head for global dma_client_list - */ -struct dma_client { - dma_event_callback event_callback; - dma_cap_mask_t cap_mask; - struct list_head global_node; -}; - typedef void (*dma_async_tx_callback)(void *dma_async_param); /** * struct dma_async_tx_descriptor - async transaction descriptor @@ -285,8 +239,7 @@ struct dma_device { int dev_id; struct device *dev; - int (*device_alloc_chan_resources)(struct dma_chan *chan, - struct dma_client *client); + int (*device_alloc_chan_resources)(struct dma_chan *chan); void (*device_free_chan_resources)(struct dma_chan *chan); struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)( @@ -320,7 +273,6 @@ struct dma_device { void dmaengine_get(void); void dmaengine_put(void); -void dma_async_client_chan_request(struct dma_client *client); dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest, void *src, size_t len); dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan, diff --git a/net/core/dev.c b/net/core/dev.c index 7596fc9403c8..ac55d84d6255 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -167,25 +167,6 @@ static DEFINE_SPINLOCK(ptype_lock); static struct list_head ptype_base[PTYPE_HASH_SIZE] __read_mostly; static struct list_head ptype_all __read_mostly; /* Taps */ -#ifdef CONFIG_NET_DMA -struct net_dma { - struct dma_client client; - spinlock_t lock; - cpumask_t channel_mask; - struct dma_chan **channels; -}; - -static enum dma_state_client -netdev_dma_event(struct dma_client *client, struct dma_chan *chan, - enum dma_state state); - -static struct net_dma net_dma = { - .client = { - .event_callback = netdev_dma_event, - }, -}; -#endif - /* * The @dev_base_head list is protected by @dev_base_lock and the rtnl * semaphore. @@ -4826,81 +4807,6 @@ static int dev_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#ifdef CONFIG_NET_DMA -/** - * netdev_dma_event - event callback for the net_dma_client - * @client: should always be net_dma_client - * @chan: DMA channel for the event - * @state: DMA state to be handled - */ -static enum dma_state_client -netdev_dma_event(struct dma_client *client, struct dma_chan *chan, - enum dma_state state) -{ - int i, found = 0, pos = -1; - struct net_dma *net_dma = - container_of(client, struct net_dma, client); - enum dma_state_client ack = DMA_DUP; /* default: take no action */ - - spin_lock(&net_dma->lock); - switch (state) { - case DMA_RESOURCE_AVAILABLE: - for (i = 0; i < nr_cpu_ids; i++) - if (net_dma->channels[i] == chan) { - found = 1; - break; - } else if (net_dma->channels[i] == NULL && pos < 0) - pos = i; - - if (!found && pos >= 0) { - ack = DMA_ACK; - net_dma->channels[pos] = chan; - cpu_set(pos, net_dma->channel_mask); - } - break; - case DMA_RESOURCE_REMOVED: - for (i = 0; i < nr_cpu_ids; i++) - if (net_dma->channels[i] == chan) { - found = 1; - pos = i; - break; - } - - if (found) { - ack = DMA_ACK; - cpu_clear(pos, net_dma->channel_mask); - net_dma->channels[i] = NULL; - } - break; - default: - break; - } - spin_unlock(&net_dma->lock); - - return ack; -} - -/** - * netdev_dma_register - register the networking subsystem as a DMA client - */ -static int __init netdev_dma_register(void) -{ - net_dma.channels = kzalloc(nr_cpu_ids * sizeof(struct net_dma), - GFP_KERNEL); - if (unlikely(!net_dma.channels)) { - printk(KERN_NOTICE - "netdev_dma: no memory for net_dma.channels\n"); - return -ENOMEM; - } - spin_lock_init(&net_dma.lock); - dma_cap_set(DMA_MEMCPY, net_dma.client.cap_mask); - dmaengine_get(); - return 0; -} - -#else -static int __init netdev_dma_register(void) { return -ENODEV; } -#endif /* CONFIG_NET_DMA */ /** * netdev_increment_features - increment feature set by one @@ -5120,14 +5026,15 @@ static int __init net_dev_init(void) if (register_pernet_device(&default_device_ops)) goto out; - netdev_dma_register(); - open_softirq(NET_TX_SOFTIRQ, net_tx_action); open_softirq(NET_RX_SOFTIRQ, net_rx_action); hotcpu_notifier(dev_cpu_callback, 0); dst_init(); dev_mcast_init(); + #ifdef CONFIG_NET_DMA + dmaengine_get(); + #endif rc = 0; out: return rc; -- cgit v1.2.3 From 148bc4303f9ba972cbfe5d30dfec93ec0d8ff1e1 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 6 Jan 2009 10:42:24 -0800 Subject: wireless: convert wireless ioctl to net_device_ops Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/wireless/wext.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/wireless/wext.c b/net/wireless/wext.c index e49a2d1ef1e4..cb6a5bb85d80 100644 --- a/net/wireless/wext.c +++ b/net/wireless/wext.c @@ -1055,8 +1055,8 @@ static int wireless_process_ioctl(struct net *net, struct ifreq *ifr, return private(dev, iwr, cmd, info, handler); } /* Old driver API : call driver ioctl handler */ - if (dev->do_ioctl) - return dev->do_ioctl(dev, ifr, cmd); + if (dev->netdev_ops->ndo_do_ioctl) + return dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd); return -EOPNOTSUPP; } -- cgit v1.2.3 From 035da16fb529c0383ac27c712a5bbade5c11cafe Mon Sep 17 00:00:00 2001 From: Mark McLoughlin Date: Mon, 15 Dec 2008 12:58:29 +0000 Subject: s390: remove s390_root_dev_*() Replace s390_root_dev_register() with root_device_register() etc. [Includes fix from Cornelia Huck] Signed-off-by: Mark McLoughlin Cc: Cornelia Huck Signed-off-by: Greg Kroah-Hartman --- arch/s390/include/asm/s390_rdev.h | 15 ------------ drivers/s390/Makefile | 2 +- drivers/s390/block/dcssblk.c | 11 ++++----- drivers/s390/crypto/ap_bus.c | 7 +++--- drivers/s390/kvm/kvm_virtio.c | 5 ++-- drivers/s390/net/cu3088.c | 7 +++--- drivers/s390/net/qeth_core_main.c | 7 +++--- drivers/s390/net/qeth_l2_main.c | 2 -- drivers/s390/net/qeth_l3_main.c | 2 -- drivers/s390/s390_rdev.c | 51 --------------------------------------- net/iucv/iucv.c | 7 +++--- 11 files changed, 20 insertions(+), 96 deletions(-) delete mode 100644 arch/s390/include/asm/s390_rdev.h delete mode 100644 drivers/s390/s390_rdev.c (limited to 'net') diff --git a/arch/s390/include/asm/s390_rdev.h b/arch/s390/include/asm/s390_rdev.h deleted file mode 100644 index 6fa20442a48c..000000000000 --- a/arch/s390/include/asm/s390_rdev.h +++ /dev/null @@ -1,15 +0,0 @@ -/* - * include/asm-s390/ccwdev.h - * - * Copyright (C) 2002,2005 IBM Deutschland Entwicklung GmbH, IBM Corporation - * Author(s): Cornelia Huck - * Carsten Otte - * - * Interface for s390 root device - */ - -#ifndef _S390_RDEV_H_ -#define _S390_RDEV_H_ -extern struct device *s390_root_dev_register(const char *); -extern void s390_root_dev_unregister(struct device *); -#endif /* _S390_RDEV_H_ */ diff --git a/drivers/s390/Makefile b/drivers/s390/Makefile index 4f4e7cf105d4..d0eae59bc366 100644 --- a/drivers/s390/Makefile +++ b/drivers/s390/Makefile @@ -4,7 +4,7 @@ CFLAGS_sysinfo.o += -Iinclude/math-emu -Iarch/s390/math-emu -w -obj-y += s390mach.o sysinfo.o s390_rdev.o +obj-y += s390mach.o sysinfo.o obj-y += cio/ block/ char/ crypto/ net/ scsi/ kvm/ drivers-y += drivers/s390/built-in.o diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c index 26ffc6ab441d..cfdcf1aed33c 100644 --- a/drivers/s390/block/dcssblk.c +++ b/drivers/s390/block/dcssblk.c @@ -18,7 +18,6 @@ #include #include #include -#include #define DCSSBLK_NAME "dcssblk" #define DCSSBLK_MINORS_PER_DISK 1 @@ -946,7 +945,7 @@ dcssblk_check_params(void) static void __exit dcssblk_exit(void) { - s390_root_dev_unregister(dcssblk_root_dev); + root_device_unregister(dcssblk_root_dev); unregister_blkdev(dcssblk_major, DCSSBLK_NAME); } @@ -955,22 +954,22 @@ dcssblk_init(void) { int rc; - dcssblk_root_dev = s390_root_dev_register("dcssblk"); + dcssblk_root_dev = root_device_register("dcssblk"); if (IS_ERR(dcssblk_root_dev)) return PTR_ERR(dcssblk_root_dev); rc = device_create_file(dcssblk_root_dev, &dev_attr_add); if (rc) { - s390_root_dev_unregister(dcssblk_root_dev); + root_device_unregister(dcssblk_root_dev); return rc; } rc = device_create_file(dcssblk_root_dev, &dev_attr_remove); if (rc) { - s390_root_dev_unregister(dcssblk_root_dev); + root_device_unregister(dcssblk_root_dev); return rc; } rc = register_blkdev(0, DCSSBLK_NAME); if (rc < 0) { - s390_root_dev_unregister(dcssblk_root_dev); + root_device_unregister(dcssblk_root_dev); return rc; } dcssblk_major = rc; diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c index 1f5f5d2d87d9..9c148406b980 100644 --- a/drivers/s390/crypto/ap_bus.c +++ b/drivers/s390/crypto/ap_bus.c @@ -36,7 +36,6 @@ #include #include #include -#include #include #include #include @@ -1522,7 +1521,7 @@ int __init ap_module_init(void) } /* Create /sys/devices/ap. */ - ap_root_device = s390_root_dev_register("ap"); + ap_root_device = root_device_register("ap"); rc = IS_ERR(ap_root_device) ? PTR_ERR(ap_root_device) : 0; if (rc) goto out_bus; @@ -1565,7 +1564,7 @@ out_work: hrtimer_cancel(&ap_poll_timer); destroy_workqueue(ap_work_queue); out_root: - s390_root_dev_unregister(ap_root_device); + root_device_unregister(ap_root_device); out_bus: while (i--) bus_remove_file(&ap_bus_type, ap_bus_attrs[i]); @@ -1600,7 +1599,7 @@ void ap_module_exit(void) hrtimer_cancel(&ap_poll_timer); destroy_workqueue(ap_work_queue); tasklet_kill(&ap_tasklet); - s390_root_dev_unregister(ap_root_device); + root_device_unregister(ap_root_device); while ((dev = bus_find_device(&ap_bus_type, NULL, NULL, __ap_match_all))) { diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c index 28c90b89f2b4..cbc8566fab70 100644 --- a/drivers/s390/kvm/kvm_virtio.c +++ b/drivers/s390/kvm/kvm_virtio.c @@ -24,7 +24,6 @@ #include #include #include -#include #define VIRTIO_SUBCODE_64 0x0D00 @@ -335,7 +334,7 @@ static int __init kvm_devices_init(void) if (!MACHINE_IS_KVM) return -ENODEV; - kvm_root = s390_root_dev_register("kvm_s390"); + kvm_root = root_device_register("kvm_s390"); if (IS_ERR(kvm_root)) { rc = PTR_ERR(kvm_root); printk(KERN_ERR "Could not register kvm_s390 root device"); @@ -344,7 +343,7 @@ static int __init kvm_devices_init(void) rc = vmem_add_mapping(real_memory_size, PAGE_SIZE); if (rc) { - s390_root_dev_unregister(kvm_root); + root_device_unregister(kvm_root); return rc; } diff --git a/drivers/s390/net/cu3088.c b/drivers/s390/net/cu3088.c index f4a32375c037..48383459e99b 100644 --- a/drivers/s390/net/cu3088.c +++ b/drivers/s390/net/cu3088.c @@ -25,7 +25,6 @@ #include #include -#include #include #include @@ -120,12 +119,12 @@ cu3088_init (void) { int rc; - cu3088_root_dev = s390_root_dev_register("cu3088"); + cu3088_root_dev = root_device_register("cu3088"); if (IS_ERR(cu3088_root_dev)) return PTR_ERR(cu3088_root_dev); rc = ccw_driver_register(&cu3088_driver); if (rc) - s390_root_dev_unregister(cu3088_root_dev); + root_device_unregister(cu3088_root_dev); return rc; } @@ -134,7 +133,7 @@ static void __exit cu3088_exit (void) { ccw_driver_unregister(&cu3088_driver); - s390_root_dev_unregister(cu3088_root_dev); + root_device_unregister(cu3088_root_dev); } MODULE_DEVICE_TABLE(ccw,cu3088_ids); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 6811dd529f48..d1b5bebea7fb 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -24,7 +24,6 @@ #include #include -#include #include "qeth_core.h" #include "qeth_core_offl.h" @@ -4525,7 +4524,7 @@ static int __init qeth_core_init(void) &driver_attr_group); if (rc) goto driver_err; - qeth_core_root_dev = s390_root_dev_register("qeth"); + qeth_core_root_dev = root_device_register("qeth"); rc = IS_ERR(qeth_core_root_dev) ? PTR_ERR(qeth_core_root_dev) : 0; if (rc) goto register_err; @@ -4539,7 +4538,7 @@ static int __init qeth_core_init(void) return 0; slab_err: - s390_root_dev_unregister(qeth_core_root_dev); + root_device_unregister(qeth_core_root_dev); register_err: driver_remove_file(&qeth_core_ccwgroup_driver.driver, &driver_attr_group); @@ -4557,7 +4556,7 @@ out_err: static void __exit qeth_core_exit(void) { - s390_root_dev_unregister(qeth_core_root_dev); + root_device_unregister(qeth_core_root_dev); driver_remove_file(&qeth_core_ccwgroup_driver.driver, &driver_attr_group); ccwgroup_driver_unregister(&qeth_core_ccwgroup_driver); diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 21627ba3093b..591a2b3ae4cb 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -20,8 +20,6 @@ #include #include -#include - #include "qeth_core.h" #include "qeth_core_offl.h" diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index cfda1ecffdf2..4693ee4e7b98 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -26,8 +26,6 @@ #include #include -#include - #include "qeth_l3.h" #include "qeth_core_offl.h" diff --git a/drivers/s390/s390_rdev.c b/drivers/s390/s390_rdev.c deleted file mode 100644 index 64371c05a3b3..000000000000 --- a/drivers/s390/s390_rdev.c +++ /dev/null @@ -1,51 +0,0 @@ -/* - * drivers/s390/s390_rdev.c - * s390 root device - * - * Copyright (C) 2002, 2005 IBM Deutschland Entwicklung GmbH, - * IBM Corporation - * Author(s): Cornelia Huck (cornelia.huck@de.ibm.com) - * Carsten Otte (cotte@de.ibm.com) - */ - -#include -#include -#include -#include - -static void -s390_root_dev_release(struct device *dev) -{ - kfree(dev); -} - -struct device * -s390_root_dev_register(const char *name) -{ - struct device *dev; - int ret; - - if (!strlen(name)) - return ERR_PTR(-EINVAL); - dev = kzalloc(sizeof(struct device), GFP_KERNEL); - if (!dev) - return ERR_PTR(-ENOMEM); - dev_set_name(dev, name); - dev->release = s390_root_dev_release; - ret = device_register(dev); - if (ret) { - kfree(dev); - return ERR_PTR(ret); - } - return dev; -} - -void -s390_root_dev_unregister(struct device *dev) -{ - if (dev) - device_unregister(dev); -} - -EXPORT_SYMBOL(s390_root_dev_register); -EXPORT_SYMBOL(s390_root_dev_unregister); diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index 032f61e98595..a35240f61ec3 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -50,7 +50,6 @@ #include #include #include -#include #include /* @@ -1696,7 +1695,7 @@ static int __init iucv_init(void) rc = register_external_interrupt(0x4000, iucv_external_interrupt); if (rc) goto out; - iucv_root = s390_root_dev_register("iucv"); + iucv_root = root_device_register("iucv"); if (IS_ERR(iucv_root)) { rc = PTR_ERR(iucv_root); goto out_int; @@ -1740,7 +1739,7 @@ out_free: kfree(iucv_irq_data[cpu]); iucv_irq_data[cpu] = NULL; } - s390_root_dev_unregister(iucv_root); + root_device_unregister(iucv_root); out_int: unregister_external_interrupt(0x4000, iucv_external_interrupt); out: @@ -1770,7 +1769,7 @@ static void __exit iucv_exit(void) kfree(iucv_irq_data[cpu]); iucv_irq_data[cpu] = NULL; } - s390_root_dev_unregister(iucv_root); + root_device_unregister(iucv_root); bus_unregister(&iucv_bus); unregister_external_interrupt(0x4000, iucv_external_interrupt); } -- cgit v1.2.3 From 61294e2e2730f24b833d9a713add5f4f6672c2a4 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 6 Jan 2009 10:45:57 -0800 Subject: sch_teql: convert to net_device_ops Convert this driver to net_device_ops. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/sched/sch_teql.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index cfc8e7caba62..ec697cebb63b 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -289,9 +289,9 @@ restart: do { struct net_device *slave = qdisc_dev(q); - struct netdev_queue *slave_txq; + struct netdev_queue *slave_txq = netdev_get_tx_queue(slave, 0); + const struct net_device_ops *slave_ops = slave->netdev_ops; - slave_txq = netdev_get_tx_queue(slave, 0); if (slave_txq->qdisc_sleeping != q) continue; if (__netif_subqueue_stopped(slave, subq) || @@ -305,7 +305,7 @@ restart: if (__netif_tx_trylock(slave_txq)) { if (!netif_tx_queue_stopped(slave_txq) && !netif_tx_queue_frozen(slave_txq) && - slave->hard_start_xmit(skb, slave) == 0) { + slave_ops->ndo_start_xmit(skb, slave) == 0) { __netif_tx_unlock(slave_txq); master->slaves = NEXT_SLAVE(q); netif_wake_queue(dev); @@ -420,6 +420,14 @@ static int teql_master_mtu(struct net_device *dev, int new_mtu) return 0; } +static const struct net_device_ops teql_netdev_ops = { + .ndo_open = teql_master_open, + .ndo_stop = teql_master_close, + .ndo_start_xmit = teql_master_xmit, + .ndo_get_stats = teql_master_stats, + .ndo_change_mtu = teql_master_mtu, +}; + static __init void teql_master_setup(struct net_device *dev) { struct teql_master *master = netdev_priv(dev); @@ -436,11 +444,7 @@ static __init void teql_master_setup(struct net_device *dev) ops->destroy = teql_destroy; ops->owner = THIS_MODULE; - dev->open = teql_master_open; - dev->hard_start_xmit = teql_master_xmit; - dev->stop = teql_master_close; - dev->get_stats = teql_master_stats; - dev->change_mtu = teql_master_mtu; + dev->netdev_ops = &teql_netdev_ops; dev->type = ARPHRD_VOID; dev->mtu = 1500; dev->tx_queue_len = 100; -- cgit v1.2.3 From 96e93eab20337d063c70d537bd7bc70d90e04fa3 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 6 Jan 2009 10:49:34 -0800 Subject: gro: Add internal interfaces for VLAN Previously GRO's only entry point from the outside is through napi_gro_receive and napi_gro_frags. These interfaces are for device drivers. This patch rearranges things to provide a new set of interfaces for VLANs. These interfaces are for internal use only. The VLAN code itself can then provide a set of entry points for device drivers. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++ net/core/dev.c | 82 ++++++++++++++++++++++++++++++++++------------- 2 files changed, 65 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c28bbba3c23d..114091be8872 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1373,8 +1373,14 @@ extern int netif_rx_ni(struct sk_buff *skb); #define HAVE_NETIF_RECEIVE_SKB 1 extern int netif_receive_skb(struct sk_buff *skb); extern void napi_gro_flush(struct napi_struct *napi); +extern int dev_gro_receive(struct napi_struct *napi, + struct sk_buff *skb); extern int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); +extern void napi_reuse_skb(struct napi_struct *napi, + struct sk_buff *skb); +extern struct sk_buff * napi_fraginfo_skb(struct napi_struct *napi, + struct napi_gro_fraginfo *info); extern int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info); extern void netif_nit_deliver(struct sk_buff *skb); diff --git a/net/core/dev.c b/net/core/dev.c index 382df6c09eec..bab8bcedd62e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2387,7 +2387,7 @@ void napi_gro_flush(struct napi_struct *napi) } EXPORT_SYMBOL(napi_gro_flush); -static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) +int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { struct sk_buff **pp = NULL; struct packet_type *ptype; @@ -2417,11 +2417,14 @@ static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) for (p = napi->gro_list; p; p = p->next) { count++; - NAPI_GRO_CB(p)->same_flow = - p->mac_len == mac_len && - !memcmp(skb_mac_header(p), skb_mac_header(skb), - mac_len); - NAPI_GRO_CB(p)->flush = 0; + + if (!NAPI_GRO_CB(p)->same_flow) + continue; + + if (p->mac_len != mac_len || + memcmp(skb_mac_header(p), skb_mac_header(skb), + mac_len)) + NAPI_GRO_CB(p)->same_flow = 0; } pp = ptype->gro_receive(&napi->gro_list, skb); @@ -2463,6 +2466,19 @@ ok: normal: return -1; } +EXPORT_SYMBOL(dev_gro_receive); + +static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) +{ + struct sk_buff *p; + + for (p = napi->gro_list; p; p = p->next) { + NAPI_GRO_CB(p)->same_flow = 1; + NAPI_GRO_CB(p)->flush = 0; + } + + return dev_gro_receive(napi, skb); +} int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { @@ -2479,11 +2495,26 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) } EXPORT_SYMBOL(napi_gro_receive); -int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) +void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) +{ + skb_shinfo(skb)->nr_frags = 0; + + skb->len -= skb->data_len; + skb->truesize -= skb->data_len; + skb->data_len = 0; + + __skb_pull(skb, skb_headlen(skb)); + skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); + + napi->skb = skb; +} +EXPORT_SYMBOL(napi_reuse_skb); + +struct sk_buff *napi_fraginfo_skb(struct napi_struct *napi, + struct napi_gro_fraginfo *info) { struct net_device *dev = napi->dev; struct sk_buff *skb = napi->skb; - int err = NET_RX_DROP; napi->skb = NULL; @@ -2503,16 +2534,31 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) skb->len += info->len; skb->truesize += info->len; - if (!pskb_may_pull(skb, ETH_HLEN)) - goto reuse; - - err = NET_RX_SUCCESS; + if (!pskb_may_pull(skb, ETH_HLEN)) { + napi_reuse_skb(napi, skb); + goto out; + } skb->protocol = eth_type_trans(skb, dev); skb->ip_summed = info->ip_summed; skb->csum = info->csum; +out: + return skb; +} +EXPORT_SYMBOL(napi_fraginfo_skb); + +int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) +{ + struct sk_buff *skb = napi_fraginfo_skb(napi, info); + int err = NET_RX_DROP; + + if (!skb) + goto out; + + err = NET_RX_SUCCESS; + switch (__napi_gro_receive(napi, skb)) { case -1: return netif_receive_skb(skb); @@ -2521,17 +2567,7 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info) goto out; } -reuse: - skb_shinfo(skb)->nr_frags = 0; - - skb->len -= skb->data_len; - skb->truesize -= skb->data_len; - skb->data_len = 0; - - __skb_pull(skb, skb_headlen(skb)); - skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb)); - - napi->skb = skb; + napi_reuse_skb(napi, skb); out: return err; -- cgit v1.2.3 From e1c096e251e52773afeffbbcb74d0a072be47ea3 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 6 Jan 2009 10:50:09 -0800 Subject: vlan: Add GRO interfaces This patch adds GRO interfaces for hardware-assisted VLAN reception. With this in place we're now at parity with LRO as far as the interface is concerned. That is, you can now take any LRO driver and convert it over to GRO. As the CB memory clashes with GRO's use of CB, I've removed it entirely by storing dev in skb->dev. This is OK because VLAN gets called first thing in netif_receive_skb and skb->dev is not used in between us calling netif_rx and netif_receive_skb getting called. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 19 +++++++++ net/8021q/vlan_core.c | 111 ++++++++++++++++++++++++++++++++++++++---------- 2 files changed, 107 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index a5cb0c3f6dcf..f8ff918c208f 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -115,6 +115,11 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev); extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, u16 vlan_tci, int polling); extern int vlan_hwaccel_do_receive(struct sk_buff *skb); +extern int vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, + unsigned int vlan_tci, struct sk_buff *skb); +extern int vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp, + unsigned int vlan_tci, + struct napi_gro_fraginfo *info); #else static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev) @@ -140,6 +145,20 @@ static inline int vlan_hwaccel_do_receive(struct sk_buff *skb) { return 0; } + +static inline int vlan_gro_receive(struct napi_struct *napi, + struct vlan_group *grp, + unsigned int vlan_tci, struct sk_buff *skb) +{ + return NET_RX_DROP; +} + +static inline int vlan_gro_frags(struct napi_struct *napi, + struct vlan_group *grp, unsigned int vlan_tci, + struct napi_gro_fraginfo *info) +{ + return NET_RX_DROP; +} #endif /** diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index dd86a1dc4cd0..6c1323940263 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -3,46 +3,35 @@ #include #include "vlan.h" -struct vlan_hwaccel_cb { - struct net_device *dev; -}; - -static inline struct vlan_hwaccel_cb *vlan_hwaccel_cb(struct sk_buff *skb) -{ - return (struct vlan_hwaccel_cb *)skb->cb; -} - /* VLAN rx hw acceleration helper. This acts like netif_{rx,receive_skb}(). */ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp, u16 vlan_tci, int polling) { - struct vlan_hwaccel_cb *cb = vlan_hwaccel_cb(skb); - - if (skb_bond_should_drop(skb)) { - dev_kfree_skb_any(skb); - return NET_RX_DROP; - } + if (skb_bond_should_drop(skb)) + goto drop; skb->vlan_tci = vlan_tci; - cb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK); + skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK); + + if (!skb->dev) + goto drop; return (polling ? netif_receive_skb(skb) : netif_rx(skb)); + +drop: + dev_kfree_skb_any(skb); + return NET_RX_DROP; } EXPORT_SYMBOL(__vlan_hwaccel_rx); int vlan_hwaccel_do_receive(struct sk_buff *skb) { - struct vlan_hwaccel_cb *cb = vlan_hwaccel_cb(skb); - struct net_device *dev = cb->dev; + struct net_device *dev = skb->dev; struct net_device_stats *stats; + skb->dev = vlan_dev_info(dev)->real_dev; netif_nit_deliver(skb); - if (dev == NULL) { - kfree_skb(skb); - return -1; - } - skb->dev = dev; skb->priority = vlan_get_ingress_priority(dev, skb->vlan_tci); skb->vlan_tci = 0; @@ -80,3 +69,79 @@ u16 vlan_dev_vlan_id(const struct net_device *dev) return vlan_dev_info(dev)->vlan_id; } EXPORT_SYMBOL_GPL(vlan_dev_vlan_id); + +static int vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp, + unsigned int vlan_tci, struct sk_buff *skb) +{ + struct sk_buff *p; + + if (skb_bond_should_drop(skb)) + goto drop; + + skb->vlan_tci = vlan_tci; + skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK); + + if (!skb->dev) + goto drop; + + for (p = napi->gro_list; p; p = p->next) { + NAPI_GRO_CB(p)->same_flow = p->dev == skb->dev; + NAPI_GRO_CB(p)->flush = 0; + } + + return dev_gro_receive(napi, skb); + +drop: + return 2; +} + +int vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp, + unsigned int vlan_tci, struct sk_buff *skb) +{ + int err = NET_RX_SUCCESS; + + switch (vlan_gro_common(napi, grp, vlan_tci, skb)) { + case -1: + return netif_receive_skb(skb); + + case 2: + err = NET_RX_DROP; + /* fall through */ + + case 1: + kfree_skb(skb); + break; + } + + return err; +} +EXPORT_SYMBOL(vlan_gro_receive); + +int vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp, + unsigned int vlan_tci, struct napi_gro_fraginfo *info) +{ + struct sk_buff *skb = napi_fraginfo_skb(napi, info); + int err = NET_RX_DROP; + + if (!skb) + goto out; + + err = NET_RX_SUCCESS; + + switch (vlan_gro_common(napi, grp, vlan_tci, skb)) { + case -1: + return netif_receive_skb(skb); + + case 2: + err = NET_RX_DROP; + /* fall through */ + + case 1: + napi_reuse_skb(napi, skb); + break; + } + +out: + return err; +} +EXPORT_SYMBOL(vlan_gro_frags); -- cgit v1.2.3 From 1fa17d4ba43d7e5aab5e90777b07da06524f6748 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 6 Jan 2009 11:07:54 -0800 Subject: can: omit unneeded skb_clone() calls The AF_CAN core delivered always cloned sk_buffs to the AF_CAN protocols, although this was _only_ needed by the can-raw protocol. With this (additionally documented) change, the AF_CAN core calls the callback functions of the registered AF_CAN protocols with the original (uncloned) sk_buff pointer and let's the can-raw protocol do the skb_clone() itself which omits all unneeded skb_clone() calls for other AF_CAN protocols. Signed-off-by: Oliver Hartkopp Signed-off-by: Urs Thuermann Signed-off-by: David S. Miller --- include/linux/can/core.h | 2 +- net/can/af_can.c | 15 ++++++++------- net/can/bcm.c | 12 +++++------- net/can/raw.c | 15 ++++++++------- 4 files changed, 22 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/include/linux/can/core.h b/include/linux/can/core.h index f50785ad4781..25085cbadcfc 100644 --- a/include/linux/can/core.h +++ b/include/linux/can/core.h @@ -19,7 +19,7 @@ #include #include -#define CAN_VERSION "20081130" +#define CAN_VERSION "20090105" /* increment this number each time you change some user-space interface */ #define CAN_ABI_VERSION "8" diff --git a/net/can/af_can.c b/net/can/af_can.c index 3dadb338addd..fa417ca6cbe6 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -414,6 +414,12 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, * The filter can be inverted (CAN_INV_FILTER bit set in can_id) or it can * filter for error frames (CAN_ERR_FLAG bit set in mask). * + * The provided pointer to the sk_buff is guaranteed to be valid as long as + * the callback function is running. The callback function must *not* free + * the given sk_buff while processing it's task. When the given sk_buff is + * needed after the end of the callback function it must be cloned inside + * the callback function with skb_clone(). + * * Return: * 0 on success * -ENOMEM on missing cache mem to create subscription entry @@ -569,13 +575,8 @@ EXPORT_SYMBOL(can_rx_unregister); static inline void deliver(struct sk_buff *skb, struct receiver *r) { - struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); - - if (clone) { - clone->sk = skb->sk; - r->func(clone, r->data); - r->matches++; - } + r->func(skb, r->data); + r->matches++; } static int can_rcv_filter(struct dev_rcv_lists *d, struct sk_buff *skb) diff --git a/net/can/bcm.c b/net/can/bcm.c index 6248ae2502c7..1649c8ab2c2f 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -633,7 +633,7 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) hrtimer_cancel(&op->timer); if (op->can_id != rxframe->can_id) - goto rx_freeskb; + return; /* save rx timestamp */ op->rx_stamp = skb->tstamp; @@ -645,19 +645,19 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) if (op->flags & RX_RTR_FRAME) { /* send reply for RTR-request (placed in op->frames[0]) */ bcm_can_tx(op); - goto rx_freeskb; + return; } if (op->flags & RX_FILTER_ID) { /* the easiest case */ bcm_rx_update_and_send(op, &op->last_frames[0], rxframe); - goto rx_freeskb_starttimer; + goto rx_starttimer; } if (op->nframes == 1) { /* simple compare with index 0 */ bcm_rx_cmp_to_index(op, 0, rxframe); - goto rx_freeskb_starttimer; + goto rx_starttimer; } if (op->nframes > 1) { @@ -678,10 +678,8 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) } } -rx_freeskb_starttimer: +rx_starttimer: bcm_rx_starttimer(op); -rx_freeskb: - kfree_skb(skb); } /* diff --git a/net/can/raw.c b/net/can/raw.c index 27aab63df467..0703cba4bf9f 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -99,13 +99,14 @@ static void raw_rcv(struct sk_buff *skb, void *data) struct raw_sock *ro = raw_sk(sk); struct sockaddr_can *addr; - if (!ro->recv_own_msgs) { - /* check the received tx sock reference */ - if (skb->sk == sk) { - kfree_skb(skb); - return; - } - } + /* check the received tx sock reference */ + if (!ro->recv_own_msgs && skb->sk == sk) + return; + + /* clone the given skb to be able to enqueue it into the rcv queue */ + skb = skb_clone(skb, GFP_ATOMIC); + if (!skb) + return; /* * Put the datagram to the queue so that raw_recvmsg() can -- cgit v1.2.3 From d442ad4ab1c86b453e0f44fb3de0932f386ab3e6 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 6 Jan 2009 16:45:26 -0800 Subject: dsa: convert to net_device_ops (v2) Convert this driver to use net_device_ops Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/dsa/slave.c | 51 +++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 41 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/dsa/slave.c b/net/dsa/slave.c index a3a410d20da0..a68fd79e9eca 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -286,6 +286,42 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = { .get_sset_count = dsa_slave_get_sset_count, }; +#ifdef CONFIG_NET_DSA_TAG_DSA +static const struct net_device_ops dsa_netdev_ops = { + .ndo_open = dsa_slave_open, + .ndo_stop = dsa_slave_close, + .ndo_start_xmit = dsa_xmit, + .ndo_change_rx_flags = dsa_slave_change_rx_flags, + .ndo_set_rx_mode = dsa_slave_set_rx_mode, + .ndo_set_multicast_list = dsa_slave_set_rx_mode, + .ndo_set_mac_address = dsa_slave_set_mac_address, + .ndo_do_ioctl = dsa_slave_ioctl, +}; +#endif +#ifdef CONFIG_NET_DSA_TAG_EDSA +static const struct net_device_ops edsa_netdev_ops = { + .ndo_open = dsa_slave_open, + .ndo_stop = dsa_slave_close, + .ndo_start_xmit = edsa_xmit, + .ndo_change_rx_flags = dsa_slave_change_rx_flags, + .ndo_set_rx_mode = dsa_slave_set_rx_mode, + .ndo_set_multicast_list = dsa_slave_set_rx_mode, + .ndo_set_mac_address = dsa_slave_set_mac_address, + .ndo_do_ioctl = dsa_slave_ioctl, +}; +#endif +#ifdef CONFIG_NET_DSA_TAG_TRAILER +static const struct net_device_ops trailer_netdev_ops = { + .ndo_open = dsa_slave_open, + .ndo_stop = dsa_slave_close, + .ndo_start_xmit = trailer_xmit, + .ndo_change_rx_flags = dsa_slave_change_rx_flags, + .ndo_set_rx_mode = dsa_slave_set_rx_mode, + .ndo_set_multicast_list = dsa_slave_set_rx_mode, + .ndo_set_mac_address = dsa_slave_set_mac_address, + .ndo_do_ioctl = dsa_slave_ioctl, +}; +#endif /* slave device setup *******************************************************/ struct net_device * @@ -306,32 +342,27 @@ dsa_slave_create(struct dsa_switch *ds, struct device *parent, SET_ETHTOOL_OPS(slave_dev, &dsa_slave_ethtool_ops); memcpy(slave_dev->dev_addr, master->dev_addr, ETH_ALEN); slave_dev->tx_queue_len = 0; + switch (ds->tag_protocol) { #ifdef CONFIG_NET_DSA_TAG_DSA case htons(ETH_P_DSA): - slave_dev->hard_start_xmit = dsa_xmit; + slave_dev->netdev_ops = &dsa_netdev_ops; break; #endif #ifdef CONFIG_NET_DSA_TAG_EDSA case htons(ETH_P_EDSA): - slave_dev->hard_start_xmit = edsa_xmit; + slave_dev->netdev_ops = &edsa_netdev_ops; break; #endif #ifdef CONFIG_NET_DSA_TAG_TRAILER case htons(ETH_P_TRAILER): - slave_dev->hard_start_xmit = trailer_xmit; + slave_dev->netdev_ops = &trailer_netdev_ops; break; #endif default: BUG(); } - slave_dev->open = dsa_slave_open; - slave_dev->stop = dsa_slave_close; - slave_dev->change_rx_flags = dsa_slave_change_rx_flags; - slave_dev->set_rx_mode = dsa_slave_set_rx_mode; - slave_dev->set_multicast_list = dsa_slave_set_rx_mode; - slave_dev->set_mac_address = dsa_slave_set_mac_address; - slave_dev->do_ioctl = dsa_slave_ioctl; + SET_NETDEV_DEV(slave_dev, parent); slave_dev->vlan_features = master->vlan_features; -- cgit v1.2.3 From 60fa9ca6cfff2be4132ea93b7dd632801ec0c817 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:34 -0800 Subject: wimax: internal API for the kernel space WiMAX stack This file contains declarations and definitions used by the different submodules of the stack. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- net/wimax/wimax-internal.h | 91 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) create mode 100644 net/wimax/wimax-internal.h (limited to 'net') diff --git a/net/wimax/wimax-internal.h b/net/wimax/wimax-internal.h new file mode 100644 index 000000000000..1e743d214856 --- /dev/null +++ b/net/wimax/wimax-internal.h @@ -0,0 +1,91 @@ +/* + * Linux WiMAX + * Internal API for kernel space WiMAX stack + * + * + * Copyright (C) 2007 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * This header file is for declarations and definitions internal to + * the WiMAX stack. For public APIs and documentation, see + * include/net/wimax.h and include/linux/wimax.h. + */ + +#ifndef __WIMAX_INTERNAL_H__ +#define __WIMAX_INTERNAL_H__ +#ifdef __KERNEL__ + +#include +#include + + +/* + * Decide if a (locked) device is ready for use + * + * Before using the device structure, it must be locked + * (wimax_dev->mutex). As well, most operations need to call this + * function to check if the state is the right one. + * + * An error value will be returned if the state is not the right + * one. In that case, the caller should not attempt to use the device + * and just unlock it. + */ +static inline __must_check +int wimax_dev_is_ready(struct wimax_dev *wimax_dev) +{ + if (wimax_dev->state == __WIMAX_ST_NULL) + return -EINVAL; /* Device is not even registered! */ + if (wimax_dev->state == WIMAX_ST_DOWN) + return -ENOMEDIUM; + if (wimax_dev->state == __WIMAX_ST_QUIESCING) + return -ESHUTDOWN; + return 0; +} + + +static inline +void __wimax_state_set(struct wimax_dev *wimax_dev, enum wimax_st state) +{ + wimax_dev->state = state; +} +extern void __wimax_state_change(struct wimax_dev *, enum wimax_st); + +#ifdef CONFIG_DEBUG_FS +extern int wimax_debugfs_add(struct wimax_dev *); +extern void wimax_debugfs_rm(struct wimax_dev *); +#else +static inline int wimax_debugfs_add(struct wimax_dev *wimax_dev) +{ + return 0; +} +static inline void wimax_debugfs_rm(struct wimax_dev *wimax_dev) {} +#endif + +extern void wimax_id_table_add(struct wimax_dev *); +extern struct wimax_dev *wimax_dev_get_by_genl_info(struct genl_info *, int); +extern void wimax_id_table_rm(struct wimax_dev *); +extern void wimax_id_table_release(void); + +extern int wimax_rfkill_add(struct wimax_dev *); +extern void wimax_rfkill_rm(struct wimax_dev *); + +extern struct genl_family wimax_gnl_family; +extern struct genl_multicast_group wimax_gnl_mcg; + +#endif /* #ifdef __KERNEL__ */ +#endif /* #ifndef __WIMAX_INTERNAL_H__ */ -- cgit v1.2.3 From ea912f4e7f264981faf8665cfb63d46d7f948117 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:35 -0800 Subject: wimax: debug macros and debug settings for the WiMAX stack This file contains a simple debug framework that is used in the stack; it allows the debug level to be controlled at compile-time (so the debug code is optimized out) and at run-time (for what wasn't compiled out). This is eventually going to be moved to use dynamic_printk(). Just need to find time to do it. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- include/linux/wimax/debug.h | 453 ++++++++++++++++++++++++++++++++++++++++++++ net/wimax/debug-levels.h | 42 ++++ 2 files changed, 495 insertions(+) create mode 100644 include/linux/wimax/debug.h create mode 100644 net/wimax/debug-levels.h (limited to 'net') diff --git a/include/linux/wimax/debug.h b/include/linux/wimax/debug.h new file mode 100644 index 000000000000..ba0c49399a83 --- /dev/null +++ b/include/linux/wimax/debug.h @@ -0,0 +1,453 @@ +/* + * Linux WiMAX + * Collection of tools to manage debug operations. + * + * + * Copyright (C) 2005-2007 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * Don't #include this file directly, read on! + * + * + * EXECUTING DEBUGGING ACTIONS OR NOT + * + * The main thing this framework provides is decission power to take a + * debug action (like printing a message) if the current debug level + * allows it. + * + * The decission power is at two levels: at compile-time (what does + * not make it is compiled out) and at run-time. The run-time + * selection is done per-submodule (as they are declared by the user + * of the framework). + * + * A call to d_test(L) (L being the target debug level) returns true + * if the action should be taken because the current debug levels + * allow it (both compile and run time). + * + * It follows that a call to d_test() that can be determined to be + * always false at compile time will get the code depending on it + * compiled out by optimization. + * + * + * DEBUG LEVELS + * + * It is up to the caller to define how much a debugging level is. + * + * Convention sets 0 as "no debug" (so an action marked as debug level 0 + * will always be taken). The increasing debug levels are used for + * increased verbosity. + * + * + * USAGE + * + * Group the code in modules and submodules inside each module [which + * in most cases maps to Linux modules and .c files that compose + * those]. + * + * + * For each module, there is: + * + * - a MODULENAME (single word, legal C identifier) + * + * - a debug-levels.h header file that declares the list of + * submodules and that is included by all .c files that use + * the debugging tools. The file name can be anything. + * + * - some (optional) .c code to manipulate the runtime debug levels + * through debugfs. + * + * The debug-levels.h file would look like: + * + * #ifndef __debug_levels__h__ + * #define __debug_levels__h__ + * + * #define D_MODULENAME modulename + * #define D_MASTER 10 + * + * #include + * + * enum d_module { + * D_SUBMODULE_DECLARE(submodule_1), + * D_SUBMODULE_DECLARE(submodule_2), + * ... + * D_SUBMODULE_DECLARE(submodule_N) + * }; + * + * #endif + * + * D_MASTER is the maximum compile-time debug level; any debug actions + * above this will be out. D_MODULENAME is the module name (legal C + * identifier), which has to be unique for each module (to avoid + * namespace collisions during linkage). Note those #defines need to + * be done before #including debug.h + * + * We declare N different submodules whose debug level can be + * independently controlled during runtime. + * + * In a .c file of the module (and only in one of them), define the + * following code: + * + * struct d_level D_LEVEL[] = { + * D_SUBMODULE_DEFINE(submodule_1), + * D_SUBMODULE_DEFINE(submodule_2), + * ... + * D_SUBMODULE_DEFINE(submodule_N), + * }; + * size_t D_LEVEL_SIZE = ARRAY_SIZE(D_LEVEL); + * + * Externs for d_level_MODULENAME and d_level_size_MODULENAME are used + * and declared in this file using the D_LEVEL and D_LEVEL_SIZE macros + * #defined also in this file. + * + * To manipulate from user space the levels, create a debugfs dentry + * and then register each submodule with: + * + * result = d_level_register_debugfs("PREFIX_", submodule_X, parent); + * if (result < 0) + * goto error; + * + * Where PREFIX_ is a name of your chosing. This will create debugfs + * file with a single numeric value that can be use to tweak it. To + * remove the entires, just use debugfs_remove_recursive() on 'parent'. + * + * NOTE: remember that even if this will show attached to some + * particular instance of a device, the settings are *global*. + * + * + * On each submodule (for example, .c files), the debug infrastructure + * should be included like this: + * + * #define D_SUBMODULE submodule_x // matches one in debug-levels.h + * #include "debug-levels.h" + * + * after #including all your include files. + * + * + * Now you can use the d_*() macros below [d_test(), d_fnstart(), + * d_fnend(), d_printf(), d_dump()]. + * + * If their debug level is greater than D_MASTER, they will be + * compiled out. + * + * If their debug level is lower or equal than D_MASTER but greater + * than the current debug level of their submodule, they'll be + * ignored. + * + * Otherwise, the action will be performed. + */ +#ifndef __debug__h__ +#define __debug__h__ + +#include +#include + + +/* Backend stuff */ + +/* + * Debug backend: generate a message header from a 'struct device' + * + * @head: buffer where to place the header + * @head_size: length of @head + * @dev: pointer to device used to generate a header from. If NULL, + * an empty ("") header is generated. + */ +static inline +void __d_head(char *head, size_t head_size, + struct device *dev) +{ + if (dev == NULL) + head[0] = 0; + else if ((unsigned long)dev < 4096) { + printk(KERN_ERR "E: Corrupt dev %p\n", dev); + WARN_ON(1); + } else + snprintf(head, head_size, "%s %s: ", + dev_driver_string(dev), dev->bus_id); +} + + +/* + * Debug backend: log some message if debugging is enabled + * + * @l: intended debug level + * @tag: tag to prefix the message with + * @dev: 'struct device' associated to this message + * @f: printf-like format and arguments + * + * Note this is optimized out if it doesn't pass the compile-time + * check; however, it is *always* compiled. This is useful to make + * sure the printf-like formats and variables are always checked and + * they don't get bit rot if you have all the debugging disabled. + */ +#define _d_printf(l, tag, dev, f, a...) \ +do { \ + char head[64]; \ + if (!d_test(l)) \ + break; \ + __d_head(head, sizeof(head), dev); \ + printk(KERN_ERR "%s%s%s: " f, head, __func__, tag, ##a); \ +} while (0) + + +/* + * CPP sintatic sugar to generate A_B like symbol names when one of + * the arguments is a a preprocessor #define. + */ +#define __D_PASTE__(varname, modulename) varname##_##modulename +#define __D_PASTE(varname, modulename) (__D_PASTE__(varname, modulename)) +#define _D_SUBMODULE_INDEX(_name) (D_SUBMODULE_DECLARE(_name)) + + +/* + * Store a submodule's runtime debug level and name + */ +struct d_level { + u8 level; + const char *name; +}; + + +/* + * List of available submodules and their debug levels + * + * We call them d_level_MODULENAME and d_level_size_MODULENAME; the + * macros D_LEVEL and D_LEVEL_SIZE contain the name already for + * convenience. + * + * This array and the size are defined on some .c file that is part of + * the current module. + */ +#define D_LEVEL __D_PASTE(d_level, D_MODULENAME) +#define D_LEVEL_SIZE __D_PASTE(d_level_size, D_MODULENAME) + +extern struct d_level D_LEVEL[]; +extern size_t D_LEVEL_SIZE; + + +/* + * Frontend stuff + * + * + * Stuff you need to declare prior to using the actual "debug" actions + * (defined below). + */ + +#ifndef D_MODULENAME +#error D_MODULENAME is not defined in your debug-levels.h file +/** + * D_MODULE - Name of the current module + * + * #define in your module's debug-levels.h, making sure it is + * unique. This has to be a legal C identifier. + */ +#define D_MODULENAME undefined_modulename +#endif + + +#ifndef D_MASTER +#warning D_MASTER not defined, but debug.h included! [see docs] +/** + * D_MASTER - Compile time maximum debug level + * + * #define in your debug-levels.h file to the maximum debug level the + * runtime code will be allowed to have. This allows you to provide a + * main knob. + * + * Anything above that level will be optimized out of the compile. + * + * Defaults to zero (no debug code compiled in). + * + * Maximum one definition per module (at the debug-levels.h file). + */ +#define D_MASTER 0 +#endif + +#ifndef D_SUBMODULE +#error D_SUBMODULE not defined, but debug.h included! [see docs] +/** + * D_SUBMODULE - Name of the current submodule + * + * #define in your submodule .c file before #including debug-levels.h + * to the name of the current submodule as previously declared and + * defined with D_SUBMODULE_DECLARE() (in your module's + * debug-levels.h) and D_SUBMODULE_DEFINE(). + * + * This is used to provide runtime-control over the debug levels. + * + * Maximum one per .c file! Can be shared among different .c files + * (meaning they belong to the same submodule categorization). + */ +#define D_SUBMODULE undefined_module +#endif + + +/** + * D_SUBMODULE_DECLARE - Declare a submodule for runtime debug level control + * + * @_name: name of the submodule, restricted to the chars that make up a + * valid C identifier ([a-zA-Z0-9_]). + * + * Declare in the module's debug-levels.h header file as: + * + * enum d_module { + * D_SUBMODULE_DECLARE(submodule_1), + * D_SUBMODULE_DECLARE(submodule_2), + * D_SUBMODULE_DECLARE(submodule_3), + * }; + * + * Some corresponding .c file needs to have a matching + * D_SUBMODULE_DEFINE(). + */ +#define D_SUBMODULE_DECLARE(_name) __D_SUBMODULE_##_name + + +/** + * D_SUBMODULE_DEFINE - Define a submodule for runtime debug level control + * + * @_name: name of the submodule, restricted to the chars that make up a + * valid C identifier ([a-zA-Z0-9_]). + * + * Use once per module (in some .c file) as: + * + * static + * struct d_level d_level_SUBMODULENAME[] = { + * D_SUBMODULE_DEFINE(submodule_1), + * D_SUBMODULE_DEFINE(submodule_2), + * D_SUBMODULE_DEFINE(submodule_3), + * }; + * size_t d_level_size_SUBDMODULENAME = ARRAY_SIZE(d_level_SUBDMODULENAME); + * + * Matching D_SUBMODULE_DECLARE()s have to be present in a + * debug-levels.h header file. + */ +#define D_SUBMODULE_DEFINE(_name) \ +[__D_SUBMODULE_##_name] = { \ + .level = 0, \ + .name = #_name \ +} + + + +/* The actual "debug" operations */ + + +/** + * d_test - Returns true if debugging should be enabled + * + * @l: intended debug level (unsigned) + * + * If the master debug switch is enabled and the current settings are + * higher or equal to the requested level, then debugging + * output/actions should be enabled. + * + * NOTE: + * + * This needs to be coded so that it can be evaluated in compile + * time; this is why the ugly BUG_ON() is placed in there, so the + * D_MASTER evaluation compiles all out if it is compile-time false. + */ +#define d_test(l) \ +({ \ + unsigned __l = l; /* type enforcer */ \ + (D_MASTER) >= __l \ + && ({ \ + BUG_ON(_D_SUBMODULE_INDEX(D_SUBMODULE) >= D_LEVEL_SIZE);\ + D_LEVEL[_D_SUBMODULE_INDEX(D_SUBMODULE)].level >= __l; \ + }); \ +}) + + +/** + * d_fnstart - log message at function start if debugging enabled + * + * @l: intended debug level + * @_dev: 'struct device' pointer, NULL if none (for context) + * @f: printf-like format and arguments + */ +#define d_fnstart(l, _dev, f, a...) _d_printf(l, " FNSTART", _dev, f, ## a) + + +/** + * d_fnend - log message at function end if debugging enabled + * + * @l: intended debug level + * @_dev: 'struct device' pointer, NULL if none (for context) + * @f: printf-like format and arguments + */ +#define d_fnend(l, _dev, f, a...) _d_printf(l, " FNEND", _dev, f, ## a) + + +/** + * d_printf - log message if debugging enabled + * + * @l: intended debug level + * @_dev: 'struct device' pointer, NULL if none (for context) + * @f: printf-like format and arguments + */ +#define d_printf(l, _dev, f, a...) _d_printf(l, "", _dev, f, ## a) + + +/** + * d_dump - log buffer hex dump if debugging enabled + * + * @l: intended debug level + * @_dev: 'struct device' pointer, NULL if none (for context) + * @f: printf-like format and arguments + */ +#define d_dump(l, dev, ptr, size) \ +do { \ + char head[64]; \ + if (!d_test(l)) \ + break; \ + __d_head(head, sizeof(head), dev); \ + print_hex_dump(KERN_ERR, head, 0, 16, 1, \ + ((void *) ptr), (size), 0); \ +} while (0) + + +/** + * Export a submodule's debug level over debugfs as PREFIXSUBMODULE + * + * @prefix: string to prefix the name with + * @submodule: name of submodule (not a string, just the name) + * @dentry: debugfs parent dentry + * + * Returns: 0 if ok, < 0 errno on error. + * + * For removing, just use debugfs_remove_recursive() on the parent. + */ +#define d_level_register_debugfs(prefix, name, parent) \ +({ \ + int rc; \ + struct dentry *fd; \ + struct dentry *verify_parent_type = parent; \ + fd = debugfs_create_u8( \ + prefix #name, 0600, verify_parent_type, \ + &(D_LEVEL[__D_SUBMODULE_ ## name].level)); \ + rc = PTR_ERR(fd); \ + if (IS_ERR(fd) && rc != -ENODEV) \ + printk(KERN_ERR "%s: Can't create debugfs entry %s: " \ + "%d\n", __func__, prefix #name, rc); \ + else \ + rc = 0; \ + rc; \ +}) + + +#endif /* #ifndef __debug__h__ */ diff --git a/net/wimax/debug-levels.h b/net/wimax/debug-levels.h new file mode 100644 index 000000000000..1c29123a3aa9 --- /dev/null +++ b/net/wimax/debug-levels.h @@ -0,0 +1,42 @@ +/* + * Linux WiMAX Stack + * Debug levels control file for the wimax module + * + * + * Copyright (C) 2007-2008 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + */ +#ifndef __debug_levels__h__ +#define __debug_levels__h__ + +/* Maximum compile and run time debug level for all submodules */ +#define D_MODULENAME wimax +#define D_MASTER CONFIG_WIMAX_DEBUG_LEVEL + +#include + +/* List of all the enabled modules */ +enum d_module { + D_SUBMODULE_DECLARE(debugfs), + D_SUBMODULE_DECLARE(id_table), + D_SUBMODULE_DECLARE(op_msg), + D_SUBMODULE_DECLARE(op_reset), + D_SUBMODULE_DECLARE(op_rfkill), + D_SUBMODULE_DECLARE(stack), +}; + +#endif /* #ifndef __debug_levels__h__ */ -- cgit v1.2.3 From 15530dfd330bd19d14e096f88c70355a61fda3f2 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:36 -0800 Subject: wimax: generic device management (registration, deregistration, lookup) Implements the basic life cycles of a 'struct wimax_dev', some common generic netlink functionality for marshalling calls to user space, and the device state machine. For looking up net devices based on their generic netlink family IDs, use a low overhead method that optimizes for the case where most systems have a single WiMAX device, or at most, a very low number of WiMAX adaptors. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- net/wimax/id-table.c | 142 ++++++++++++ net/wimax/stack.c | 599 +++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 741 insertions(+) create mode 100644 net/wimax/id-table.c create mode 100644 net/wimax/stack.c (limited to 'net') diff --git a/net/wimax/id-table.c b/net/wimax/id-table.c new file mode 100644 index 000000000000..d3b88558682c --- /dev/null +++ b/net/wimax/id-table.c @@ -0,0 +1,142 @@ +/* + * Linux WiMAX + * Mappping of generic netlink family IDs to net devices + * + * + * Copyright (C) 2005-2006 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * We assign a single generic netlink family ID to each device (to + * simplify lookup). + * + * We need a way to map family ID to a wimax_dev pointer. + * + * The idea is to use a very simple lookup. Using a netlink attribute + * with (for example) the interface name implies a heavier search over + * all the network devices; seemed kind of a waste given that we know + * we are looking for a WiMAX device and that most systems will have + * just a single WiMAX adapter. + * + * We put all the WiMAX devices in the system in a linked list and + * match the generic link family ID against the list. + * + * By using a linked list, the case of a single adapter in the system + * becomes (almost) no overhead, while still working for many more. If + * it ever goes beyond two, I'll be surprised. + */ +#include +#include +#include +#include +#include +#include "wimax-internal.h" + + +#define D_SUBMODULE id_table +#include "debug-levels.h" + + +static DEFINE_SPINLOCK(wimax_id_table_lock); +static struct list_head wimax_id_table = LIST_HEAD_INIT(wimax_id_table); + + +/* + * wimax_id_table_add - add a gennetlink familiy ID / wimax_dev mapping + * + * @wimax_dev: WiMAX device descriptor to associate to the Generic + * Netlink family ID. + * + * Look for an empty spot in the ID table; if none found, double the + * table's size and get the first spot. + */ +void wimax_id_table_add(struct wimax_dev *wimax_dev) +{ + d_fnstart(3, NULL, "(wimax_dev %p)\n", wimax_dev); + spin_lock(&wimax_id_table_lock); + list_add(&wimax_dev->id_table_node, &wimax_id_table); + spin_unlock(&wimax_id_table_lock); + d_fnend(3, NULL, "(wimax_dev %p)\n", wimax_dev); +} + + +/* + * wimax_get_netdev_by_info - lookup a wimax_dev from the gennetlink info + * + * The generic netlink family ID has been filled out in the + * nlmsghdr->nlmsg_type field, so we pull it from there, look it up in + * the mapping table and reference the wimax_dev. + * + * When done, the reference should be dropped with + * 'dev_put(wimax_dev->net_dev)'. + */ +struct wimax_dev *wimax_dev_get_by_genl_info( + struct genl_info *info, int ifindex) +{ + struct wimax_dev *wimax_dev = NULL; + + d_fnstart(3, NULL, "(info %p ifindex %d)\n", info, ifindex); + spin_lock(&wimax_id_table_lock); + list_for_each_entry(wimax_dev, &wimax_id_table, id_table_node) { + if (wimax_dev->net_dev->ifindex == ifindex) { + dev_hold(wimax_dev->net_dev); + break; + } + } + if (wimax_dev == NULL) + d_printf(1, NULL, "wimax: no devices found with ifindex %d\n", + ifindex); + spin_unlock(&wimax_id_table_lock); + d_fnend(3, NULL, "(info %p ifindex %d) = %p\n", + info, ifindex, wimax_dev); + return wimax_dev; +} + + +/* + * wimax_id_table_rm - Remove a gennetlink familiy ID / wimax_dev mapping + * + * @id: family ID to remove from the table + */ +void wimax_id_table_rm(struct wimax_dev *wimax_dev) +{ + spin_lock(&wimax_id_table_lock); + list_del_init(&wimax_dev->id_table_node); + spin_unlock(&wimax_id_table_lock); +} + + +/* + * Release the gennetlink family id / mapping table + * + * On debug, verify that the table is empty upon removal. + */ +void wimax_id_table_release(void) +{ +#ifndef CONFIG_BUG + return; +#endif + struct wimax_dev *wimax_dev; + + spin_lock(&wimax_id_table_lock); + list_for_each_entry(wimax_dev, &wimax_id_table, id_table_node) { + printk(KERN_ERR "BUG: %s wimax_dev %p ifindex %d not cleared\n", + __func__, wimax_dev, wimax_dev->net_dev->ifindex); + WARN_ON(1); + } + spin_unlock(&wimax_id_table_lock); +} diff --git a/net/wimax/stack.c b/net/wimax/stack.c new file mode 100644 index 000000000000..d4da92f8981a --- /dev/null +++ b/net/wimax/stack.c @@ -0,0 +1,599 @@ +/* + * Linux WiMAX + * Initialization, addition and removal of wimax devices + * + * + * Copyright (C) 2005-2006 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * This implements: + * + * - basic life cycle of 'struct wimax_dev' [wimax_dev_*()]; on + * addition/registration initialize all subfields and allocate + * generic netlink resources for user space communication. On + * removal/unregistration, undo all that. + * + * - device state machine [wimax_state_change()] and support to send + * reports to user space when the state changes + * [wimax_gnl_re_state_change*()]. + * + * See include/net/wimax.h for rationales and design. + * + * ROADMAP + * + * [__]wimax_state_change() Called by drivers to update device's state + * wimax_gnl_re_state_change_alloc() + * wimax_gnl_re_state_change_send() + * + * wimax_dev_init() Init a device + * wimax_dev_add() Register + * wimax_rfkill_add() + * wimax_gnl_add() Register all the generic netlink resources. + * wimax_id_table_add() + * wimax_dev_rm() Unregister + * wimax_id_table_rm() + * wimax_gnl_rm() + * wimax_rfkill_rm() + */ +#include +#include +#include +#include +#include "wimax-internal.h" + + +#define D_SUBMODULE stack +#include "debug-levels.h" + +/* + * Authoritative source for the RE_STATE_CHANGE attribute policy + * + * We don't really use it here, but /me likes to keep the definition + * close to where the data is generated. + */ +/* +static const +struct nla_policy wimax_gnl_re_status_change[WIMAX_GNL_ATTR_MAX + 1] = { + [WIMAX_GNL_STCH_STATE_OLD] = { .type = NLA_U8 }, + [WIMAX_GNL_STCH_STATE_NEW] = { .type = NLA_U8 }, +}; +*/ + + +/* + * Allocate a Report State Change message + * + * @header: save it, you need it for _send() + * + * Creates and fills a basic state change message; different code + * paths can then add more attributes to the message as needed. + * + * Use wimax_gnl_re_state_change_send() to send the returned skb. + * + * Returns: skb with the genl message if ok, IS_ERR() ptr on error + * with an errno code. + */ +static +struct sk_buff *wimax_gnl_re_state_change_alloc( + struct wimax_dev *wimax_dev, + enum wimax_st new_state, enum wimax_st old_state, + void **header) +{ + int result; + struct device *dev = wimax_dev_to_dev(wimax_dev); + void *data; + struct sk_buff *report_skb; + + d_fnstart(3, dev, "(wimax_dev %p new_state %u old_state %u)\n", + wimax_dev, new_state, old_state); + result = -ENOMEM; + report_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (report_skb == NULL) { + dev_err(dev, "RE_STCH: can't create message\n"); + goto error_new; + } + data = genlmsg_put(report_skb, 0, wimax_gnl_mcg.id, &wimax_gnl_family, + 0, WIMAX_GNL_RE_STATE_CHANGE); + if (data == NULL) { + dev_err(dev, "RE_STCH: can't put data into message\n"); + goto error_put; + } + *header = data; + + result = nla_put_u8(report_skb, WIMAX_GNL_STCH_STATE_OLD, old_state); + if (result < 0) { + dev_err(dev, "RE_STCH: Error adding OLD attr: %d\n", result); + goto error_put; + } + result = nla_put_u8(report_skb, WIMAX_GNL_STCH_STATE_NEW, new_state); + if (result < 0) { + dev_err(dev, "RE_STCH: Error adding NEW attr: %d\n", result); + goto error_put; + } + result = nla_put_u32(report_skb, WIMAX_GNL_STCH_IFIDX, + wimax_dev->net_dev->ifindex); + if (result < 0) { + dev_err(dev, "RE_STCH: Error adding IFINDEX attribute\n"); + goto error_put; + } + d_fnend(3, dev, "(wimax_dev %p new_state %u old_state %u) = %p\n", + wimax_dev, new_state, old_state, report_skb); + return report_skb; + +error_put: + nlmsg_free(report_skb); +error_new: + d_fnend(3, dev, "(wimax_dev %p new_state %u old_state %u) = %d\n", + wimax_dev, new_state, old_state, result); + return ERR_PTR(result); +} + + +/* + * Send a Report State Change message (as created with _alloc). + * + * @report_skb: as returned by wimax_gnl_re_state_change_alloc() + * @header: as returned by wimax_gnl_re_state_change_alloc() + * + * Returns: 0 if ok, < 0 errno code on error. + * + * If the message is NULL, pretend it didn't happen. + */ +static +int wimax_gnl_re_state_change_send( + struct wimax_dev *wimax_dev, struct sk_buff *report_skb, + void *header) +{ + int result = 0; + struct device *dev = wimax_dev_to_dev(wimax_dev); + d_fnstart(3, dev, "(wimax_dev %p report_skb %p)\n", + wimax_dev, report_skb); + if (report_skb == NULL) + goto out; + genlmsg_end(report_skb, header); + result = genlmsg_multicast(report_skb, 0, wimax_gnl_mcg.id, GFP_KERNEL); + if (result == -ESRCH) /* Nobody connected, ignore it */ + result = 0; /* btw, the skb is freed already */ + if (result < 0) { + dev_err(dev, "RE_STCH: Error sending: %d\n", result); + nlmsg_free(report_skb); + } +out: + d_fnend(3, dev, "(wimax_dev %p report_skb %p) = %d\n", + wimax_dev, report_skb, result); + return result; +} + + +static +void __check_new_state(enum wimax_st old_state, enum wimax_st new_state, + unsigned allowed_states_bm) +{ + if (WARN_ON(((1 << new_state) & allowed_states_bm) == 0)) { + printk(KERN_ERR "SW BUG! Forbidden state change %u -> %u\n", + old_state, new_state); + } +} + + +/* + * Set the current state of a WiMAX device [unlocking version of + * wimax_state_change(). + */ +void __wimax_state_change(struct wimax_dev *wimax_dev, enum wimax_st new_state) +{ + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_st old_state = wimax_dev->state; + struct sk_buff *stch_skb; + void *header; + + d_fnstart(3, dev, "(wimax_dev %p new_state %u [old %u])\n", + wimax_dev, new_state, old_state); + + if (WARN_ON(new_state >= __WIMAX_ST_INVALID)) { + dev_err(dev, "SW BUG: requesting invalid state %u\n", + new_state); + goto out; + } + if (old_state == new_state) + goto out; + header = NULL; /* gcc complains? can't grok why */ + stch_skb = wimax_gnl_re_state_change_alloc( + wimax_dev, new_state, old_state, &header); + + /* Verify the state transition and do exit-from-state actions */ + switch (old_state) { + case __WIMAX_ST_NULL: + __check_new_state(old_state, new_state, + 1 << WIMAX_ST_DOWN); + break; + case WIMAX_ST_DOWN: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_UNINITIALIZED + | 1 << WIMAX_ST_RADIO_OFF); + break; + case __WIMAX_ST_QUIESCING: + __check_new_state(old_state, new_state, 1 << WIMAX_ST_DOWN); + break; + case WIMAX_ST_UNINITIALIZED: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_RADIO_OFF); + break; + case WIMAX_ST_RADIO_OFF: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_READY); + break; + case WIMAX_ST_READY: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_RADIO_OFF + | 1 << WIMAX_ST_SCANNING + | 1 << WIMAX_ST_CONNECTING + | 1 << WIMAX_ST_CONNECTED); + break; + case WIMAX_ST_SCANNING: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_RADIO_OFF + | 1 << WIMAX_ST_READY + | 1 << WIMAX_ST_CONNECTING + | 1 << WIMAX_ST_CONNECTED); + break; + case WIMAX_ST_CONNECTING: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_RADIO_OFF + | 1 << WIMAX_ST_READY + | 1 << WIMAX_ST_SCANNING + | 1 << WIMAX_ST_CONNECTED); + break; + case WIMAX_ST_CONNECTED: + __check_new_state(old_state, new_state, + 1 << __WIMAX_ST_QUIESCING + | 1 << WIMAX_ST_RADIO_OFF + | 1 << WIMAX_ST_READY); + netif_tx_disable(wimax_dev->net_dev); + netif_carrier_off(wimax_dev->net_dev); + break; + case __WIMAX_ST_INVALID: + default: + dev_err(dev, "SW BUG: wimax_dev %p is in unknown state %u\n", + wimax_dev, wimax_dev->state); + WARN_ON(1); + goto out; + } + + /* Execute the actions of entry to the new state */ + switch (new_state) { + case __WIMAX_ST_NULL: + dev_err(dev, "SW BUG: wimax_dev %p entering NULL state " + "from %u\n", wimax_dev, wimax_dev->state); + WARN_ON(1); /* Nobody can enter this state */ + break; + case WIMAX_ST_DOWN: + break; + case __WIMAX_ST_QUIESCING: + break; + case WIMAX_ST_UNINITIALIZED: + break; + case WIMAX_ST_RADIO_OFF: + break; + case WIMAX_ST_READY: + break; + case WIMAX_ST_SCANNING: + break; + case WIMAX_ST_CONNECTING: + break; + case WIMAX_ST_CONNECTED: + netif_carrier_on(wimax_dev->net_dev); + netif_wake_queue(wimax_dev->net_dev); + break; + case __WIMAX_ST_INVALID: + default: + BUG(); + } + __wimax_state_set(wimax_dev, new_state); + if (stch_skb) + wimax_gnl_re_state_change_send(wimax_dev, stch_skb, header); +out: + d_fnend(3, dev, "(wimax_dev %p new_state %u [old %u]) = void\n", + wimax_dev, new_state, old_state); + return; +} + + +/** + * wimax_state_change - Set the current state of a WiMAX device + * + * @wimax_dev: WiMAX device descriptor (properly referenced) + * @new_state: New state to switch to + * + * This implements the state changes for the wimax devices. It will + * + * - verify that the state transition is legal (for now it'll just + * print a warning if not) according to the table in + * linux/wimax.h's documentation for 'enum wimax_st'. + * + * - perform the actions needed for leaving the current state and + * whichever are needed for entering the new state. + * + * - issue a report to user space indicating the new state (and an + * optional payload with information about the new state). + * + * NOTE: @wimax_dev must be locked + */ +void wimax_state_change(struct wimax_dev *wimax_dev, enum wimax_st new_state) +{ + mutex_lock(&wimax_dev->mutex); + __wimax_state_change(wimax_dev, new_state); + mutex_unlock(&wimax_dev->mutex); + return; +} +EXPORT_SYMBOL_GPL(wimax_state_change); + + +/** + * wimax_state_get() - Return the current state of a WiMAX device + * + * @wimax_dev: WiMAX device descriptor + * + * Returns: Current state of the device according to its driver. + */ +enum wimax_st wimax_state_get(struct wimax_dev *wimax_dev) +{ + enum wimax_st state; + mutex_lock(&wimax_dev->mutex); + state = wimax_dev->state; + mutex_unlock(&wimax_dev->mutex); + return state; +} +EXPORT_SYMBOL_GPL(wimax_state_get); + + +/** + * wimax_dev_init - initialize a newly allocated instance + * + * @wimax_dev: WiMAX device descriptor to initialize. + * + * Initializes fields of a freshly allocated @wimax_dev instance. This + * function assumes that after allocation, the memory occupied by + * @wimax_dev was zeroed. + */ +void wimax_dev_init(struct wimax_dev *wimax_dev) +{ + INIT_LIST_HEAD(&wimax_dev->id_table_node); + __wimax_state_set(wimax_dev, WIMAX_ST_UNINITIALIZED); + mutex_init(&wimax_dev->mutex); + mutex_init(&wimax_dev->mutex_reset); +} +EXPORT_SYMBOL_GPL(wimax_dev_init); + +/* + * This extern is declared here because it's easier to keep track -- + * both declarations are a list of the same + */ +extern struct genl_ops + wimax_gnl_msg_from_user, + wimax_gnl_reset, + wimax_gnl_rfkill; + +static +struct genl_ops *wimax_gnl_ops[] = { + &wimax_gnl_msg_from_user, + &wimax_gnl_reset, + &wimax_gnl_rfkill, +}; + + +static +size_t wimax_addr_scnprint(char *addr_str, size_t addr_str_size, + unsigned char *addr, size_t addr_len) +{ + unsigned cnt, total; + for (total = cnt = 0; cnt < addr_len; cnt++) + total += scnprintf(addr_str + total, addr_str_size - total, + "%02x%c", addr[cnt], + cnt == addr_len - 1 ? '\0' : ':'); + return total; +} + + +/** + * wimax_dev_add - Register a new WiMAX device + * + * @wimax_dev: WiMAX device descriptor (as embedded in your @net_dev's + * priv data). You must have called wimax_dev_init() on it before. + * + * @net_dev: net device the @wimax_dev is associated with. The + * function expects SET_NETDEV_DEV() and register_netdev() were + * already called on it. + * + * Registers the new WiMAX device, sets up the user-kernel control + * interface (generic netlink) and common WiMAX infrastructure. + * + * Note that the parts that will allow interaction with user space are + * setup at the very end, when the rest is in place, as once that + * happens, the driver might get user space control requests via + * netlink or from debugfs that might translate into calls into + * wimax_dev->op_*(). + */ +int wimax_dev_add(struct wimax_dev *wimax_dev, struct net_device *net_dev) +{ + int result; + struct device *dev = net_dev->dev.parent; + char addr_str[32]; + + d_fnstart(3, dev, "(wimax_dev %p net_dev %p)\n", wimax_dev, net_dev); + + /* Do the RFKILL setup before locking, as RFKILL will call + * into our functions. */ + wimax_dev->net_dev = net_dev; + result = wimax_rfkill_add(wimax_dev); + if (result < 0) + goto error_rfkill_add; + + /* Set up user-space interaction */ + mutex_lock(&wimax_dev->mutex); + wimax_id_table_add(wimax_dev); + result = wimax_debugfs_add(wimax_dev); + if (result < 0) { + dev_err(dev, "cannot initialize debugfs: %d\n", + result); + goto error_debugfs_add; + } + + __wimax_state_set(wimax_dev, WIMAX_ST_DOWN); + mutex_unlock(&wimax_dev->mutex); + + wimax_addr_scnprint(addr_str, sizeof(addr_str), + net_dev->dev_addr, net_dev->addr_len); + dev_err(dev, "WiMAX interface %s (%s) ready\n", + net_dev->name, addr_str); + d_fnend(3, dev, "(wimax_dev %p net_dev %p) = 0\n", wimax_dev, net_dev); + return 0; + +error_debugfs_add: + wimax_id_table_rm(wimax_dev); + mutex_unlock(&wimax_dev->mutex); + wimax_rfkill_rm(wimax_dev); +error_rfkill_add: + d_fnend(3, dev, "(wimax_dev %p net_dev %p) = %d\n", + wimax_dev, net_dev, result); + return result; +} +EXPORT_SYMBOL_GPL(wimax_dev_add); + + +/** + * wimax_dev_rm - Unregister an existing WiMAX device + * + * @wimax_dev: WiMAX device descriptor + * + * Unregisters a WiMAX device previously registered for use with + * wimax_add_rm(). + * + * IMPORTANT! Must call before calling unregister_netdev(). + * + * After this function returns, you will not get any more user space + * control requests (via netlink or debugfs) and thus to wimax_dev->ops. + * + * Reentrancy control is ensured by setting the state to + * %__WIMAX_ST_QUIESCING. rfkill operations coming through + * wimax_*rfkill*() will be stopped by the quiescing state; ops coming + * from the rfkill subsystem will be stopped by the support being + * removed by wimax_rfkill_rm(). + */ +void wimax_dev_rm(struct wimax_dev *wimax_dev) +{ + d_fnstart(3, NULL, "(wimax_dev %p)\n", wimax_dev); + + mutex_lock(&wimax_dev->mutex); + __wimax_state_change(wimax_dev, __WIMAX_ST_QUIESCING); + wimax_debugfs_rm(wimax_dev); + wimax_id_table_rm(wimax_dev); + __wimax_state_change(wimax_dev, WIMAX_ST_DOWN); + mutex_unlock(&wimax_dev->mutex); + wimax_rfkill_rm(wimax_dev); + d_fnend(3, NULL, "(wimax_dev %p) = void\n", wimax_dev); +} +EXPORT_SYMBOL_GPL(wimax_dev_rm); + +struct genl_family wimax_gnl_family = { + .id = GENL_ID_GENERATE, + .name = "WiMAX", + .version = WIMAX_GNL_VERSION, + .hdrsize = 0, + .maxattr = WIMAX_GNL_ATTR_MAX, +}; + +struct genl_multicast_group wimax_gnl_mcg = { + .name = "msg", +}; + + + +/* Shutdown the wimax stack */ +static +int __init wimax_subsys_init(void) +{ + int result, cnt; + + d_fnstart(4, NULL, "()\n"); + snprintf(wimax_gnl_family.name, sizeof(wimax_gnl_family.name), + "WiMAX"); + result = genl_register_family(&wimax_gnl_family); + if (unlikely(result < 0)) { + printk(KERN_ERR "cannot register generic netlink family: %d\n", + result); + goto error_register_family; + } + + for (cnt = 0; cnt < ARRAY_SIZE(wimax_gnl_ops); cnt++) { + result = genl_register_ops(&wimax_gnl_family, + wimax_gnl_ops[cnt]); + d_printf(4, NULL, "registering generic netlink op code " + "%u: %d\n", wimax_gnl_ops[cnt]->cmd, result); + if (unlikely(result < 0)) { + printk(KERN_ERR "cannot register generic netlink op " + "code %u: %d\n", + wimax_gnl_ops[cnt]->cmd, result); + goto error_register_ops; + } + } + + result = genl_register_mc_group(&wimax_gnl_family, &wimax_gnl_mcg); + if (result < 0) + goto error_mc_group; + d_fnend(4, NULL, "() = 0\n"); + return 0; + +error_mc_group: +error_register_ops: + for (cnt--; cnt >= 0; cnt--) + genl_unregister_ops(&wimax_gnl_family, + wimax_gnl_ops[cnt]); + genl_unregister_family(&wimax_gnl_family); +error_register_family: + d_fnend(4, NULL, "() = %d\n", result); + return result; + +} +module_init(wimax_subsys_init); + + +/* Shutdown the wimax stack */ +static +void __exit wimax_subsys_exit(void) +{ + int cnt; + wimax_id_table_release(); + genl_unregister_mc_group(&wimax_gnl_family, &wimax_gnl_mcg); + for (cnt = ARRAY_SIZE(wimax_gnl_ops) - 1; cnt >= 0; cnt--) + genl_unregister_ops(&wimax_gnl_family, + wimax_gnl_ops[cnt]); + genl_unregister_family(&wimax_gnl_family); +} +module_exit(wimax_subsys_exit); + +MODULE_AUTHOR("Intel Corporation "); +MODULE_DESCRIPTION("Linux WiMAX stack"); +MODULE_LICENSE("GPL"); + -- cgit v1.2.3 From 3efb40c2c6eea315abcf19239c15d088e2498299 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:37 -0800 Subject: genetlink: export genl_unregister_mc_group() Add an EXPORT_SYMBOL() to genl_unregister_mc_group(), to allow unregistering groups on the run. EXPORT_SYMBOL_GPL() is not used as the rest of the functions exported by this module (eg: genl_register_mc_group) are also not _GPL(). Cleanup is currently done when unregistering a family, but there is no way to unregister a single multicast group due to that function not being exported. Seems to be a mistake as it is documented as for external consumption. This is needed by the WiMAX stack to be able to cleanup unused mc groups. Signed-off-by: Inaky Perez-Gonzalez Acked-by: Johannes Berg Signed-off-by: Greg Kroah-Hartman --- net/netlink/genetlink.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 3e1191cecaf0..1d3dd30099df 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -225,6 +225,7 @@ void genl_unregister_mc_group(struct genl_family *family, __genl_unregister_mc_group(family, grp); genl_unlock(); } +EXPORT_SYMBOL(genl_unregister_mc_group); static void genl_unregister_mc_groups(struct genl_family *family) { -- cgit v1.2.3 From 3e65646bb12be03b14dff53907391095a52d5f49 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:38 -0800 Subject: wimax: basic API: kernel/user messaging, rfkill and reset Implements the three basic operations provided by the stack's control interface to WiMAX devices: - Messaging channel between user space and driver/device This implements a direct communication channel between user space and the driver/device, by which free form messages can be sent back and forth. This is intended for device-specific features, vendor quirks, etc. - RF-kill framework integration Provide most of the RF-Kill integration for WiMAX drivers so that all device drivers have to do is after wimax_dev_add() is call wimax_report_rfkill_{hw,sw}() to update initial state and then every time it changes. Provides wimax_rfkill() for the kernel to call to set software RF-Kill status and/or query current hardware and software switch status. Exports wimax_rfkill() over generic netlink to user space. - Reset a WiMAX device Provides wimax_reset() for the kernel to reset a wimax device as needed and exports it over generic netlink to user space. This API is clearly limited, as it still provides no way to do the basic scan, connect and disconnect in a hardware independent way. The WiMAX case is more complex than WiFi due to the way networks are discovered and provisioned. The next developments are to add the basic operations so they can be offerent by different drivers. However, we'd like to get more vendors to jump in and provide feedback of how the user/kernel API/abstraction layer should be. The user space code for the i2400m, as of now, uses the messaging channel, but that will change as the API evolves. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- net/wimax/op-msg.c | 421 +++++++++++++++++++++++++++++++++++++++ net/wimax/op-reset.c | 143 ++++++++++++++ net/wimax/op-rfkill.c | 532 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 1096 insertions(+) create mode 100644 net/wimax/op-msg.c create mode 100644 net/wimax/op-reset.c create mode 100644 net/wimax/op-rfkill.c (limited to 'net') diff --git a/net/wimax/op-msg.c b/net/wimax/op-msg.c new file mode 100644 index 000000000000..cb3b4ad53683 --- /dev/null +++ b/net/wimax/op-msg.c @@ -0,0 +1,421 @@ +/* + * Linux WiMAX + * Generic messaging interface between userspace and driver/device + * + * + * Copyright (C) 2007-2008 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * This implements a direct communication channel between user space and + * the driver/device, by which free form messages can be sent back and + * forth. + * + * This is intended for device-specific features, vendor quirks, etc. + * + * See include/net/wimax.h + * + * GENERIC NETLINK ENCODING AND CAPACITY + * + * A destination "pipe name" is added to each message; it is up to the + * drivers to assign or use those names (if using them at all). + * + * Messages are encoded as a binary netlink attribute using nla_put() + * using type NLA_UNSPEC (as some versions of libnl still in + * deployment don't yet understand NLA_BINARY). + * + * The maximum capacity of this transport is PAGESIZE per message (so + * the actual payload will be bit smaller depending on the + * netlink/generic netlink attributes and headers). + * + * RECEPTION OF MESSAGES + * + * When a message is received from user space, it is passed verbatim + * to the driver calling wimax_dev->op_msg_from_user(). The return + * value from this function is passed back to user space as an ack + * over the generic netlink protocol. + * + * The stack doesn't do any processing or interpretation of these + * messages. + * + * SENDING MESSAGES + * + * Messages can be sent with wimax_msg(). + * + * If the message delivery needs to happen on a different context to + * that of its creation, wimax_msg_alloc() can be used to get a + * pointer to the message that can be delivered later on with + * wimax_msg_send(). + * + * ROADMAP + * + * wimax_gnl_doit_msg_from_user() Process a message from user space + * wimax_dev_get_by_genl_info() + * wimax_dev->op_msg_from_user() Delivery of message to the driver + * + * wimax_msg() Send a message to user space + * wimax_msg_alloc() + * wimax_msg_send() + */ +#include +#include +#include +#include +#include +#include "wimax-internal.h" + + +#define D_SUBMODULE op_msg +#include "debug-levels.h" + + +/** + * wimax_msg_alloc - Create a new skb for sending a message to userspace + * + * @wimax_dev: WiMAX device descriptor + * @pipe_name: "named pipe" the message will be sent to + * @msg: pointer to the message data to send + * @size: size of the message to send (in bytes), including the header. + * @gfp_flags: flags for memory allocation. + * + * Returns: %0 if ok, negative errno code on error + * + * Description: + * + * Allocates an skb that will contain the message to send to user + * space over the messaging pipe and initializes it, copying the + * payload. + * + * Once this call is done, you can deliver it with + * wimax_msg_send(). + * + * IMPORTANT: + * + * Don't use skb_push()/skb_pull()/skb_reserve() on the skb, as + * wimax_msg_send() depends on skb->data being placed at the + * beginning of the user message. + */ +struct sk_buff *wimax_msg_alloc(struct wimax_dev *wimax_dev, + const char *pipe_name, + const void *msg, size_t size, + gfp_t gfp_flags) +{ + int result; + struct device *dev = wimax_dev->net_dev->dev.parent; + size_t msg_size; + void *genl_msg; + struct sk_buff *skb; + + msg_size = nla_total_size(size) + + nla_total_size(sizeof(u32)) + + (pipe_name ? nla_total_size(strlen(pipe_name)) : 0); + result = -ENOMEM; + skb = genlmsg_new(msg_size, gfp_flags); + if (skb == NULL) + goto error_new; + genl_msg = genlmsg_put(skb, 0, 0, &wimax_gnl_family, + 0, WIMAX_GNL_OP_MSG_TO_USER); + if (genl_msg == NULL) { + dev_err(dev, "no memory to create generic netlink message\n"); + goto error_genlmsg_put; + } + result = nla_put_u32(skb, WIMAX_GNL_MSG_IFIDX, + wimax_dev->net_dev->ifindex); + if (result < 0) { + dev_err(dev, "no memory to add ifindex attribute\n"); + goto error_nla_put; + } + if (pipe_name) { + result = nla_put_string(skb, WIMAX_GNL_MSG_PIPE_NAME, + pipe_name); + if (result < 0) { + dev_err(dev, "no memory to add pipe_name attribute\n"); + goto error_nla_put; + } + } + result = nla_put(skb, WIMAX_GNL_MSG_DATA, size, msg); + if (result < 0) { + dev_err(dev, "no memory to add payload in attribute\n"); + goto error_nla_put; + } + genlmsg_end(skb, genl_msg); + return skb; + +error_nla_put: +error_genlmsg_put: +error_new: + nlmsg_free(skb); + return ERR_PTR(result); + +} +EXPORT_SYMBOL_GPL(wimax_msg_alloc); + + +/** + * wimax_msg_data_len - Return a pointer and size of a message's payload + * + * @msg: Pointer to a message created with wimax_msg_alloc() + * @size: Pointer to where to store the message's size + * + * Returns the pointer to the message data. + */ +const void *wimax_msg_data_len(struct sk_buff *msg, size_t *size) +{ + struct nlmsghdr *nlh = (void *) msg->head; + struct nlattr *nla; + + nla = nlmsg_find_attr(nlh, sizeof(struct genlmsghdr), + WIMAX_GNL_MSG_DATA); + if (nla == NULL) { + printk(KERN_ERR "Cannot find attribute WIMAX_GNL_MSG_DATA\n"); + return NULL; + } + *size = nla_len(nla); + return nla_data(nla); +} +EXPORT_SYMBOL_GPL(wimax_msg_data_len); + + +/** + * wimax_msg_data - Return a pointer to a message's payload + * + * @msg: Pointer to a message created with wimax_msg_alloc() + */ +const void *wimax_msg_data(struct sk_buff *msg) +{ + struct nlmsghdr *nlh = (void *) msg->head; + struct nlattr *nla; + + nla = nlmsg_find_attr(nlh, sizeof(struct genlmsghdr), + WIMAX_GNL_MSG_DATA); + if (nla == NULL) { + printk(KERN_ERR "Cannot find attribute WIMAX_GNL_MSG_DATA\n"); + return NULL; + } + return nla_data(nla); +} +EXPORT_SYMBOL_GPL(wimax_msg_data); + + +/** + * wimax_msg_len - Return a message's payload length + * + * @msg: Pointer to a message created with wimax_msg_alloc() + */ +ssize_t wimax_msg_len(struct sk_buff *msg) +{ + struct nlmsghdr *nlh = (void *) msg->head; + struct nlattr *nla; + + nla = nlmsg_find_attr(nlh, sizeof(struct genlmsghdr), + WIMAX_GNL_MSG_DATA); + if (nla == NULL) { + printk(KERN_ERR "Cannot find attribute WIMAX_GNL_MSG_DATA\n"); + return -EINVAL; + } + return nla_len(nla); +} +EXPORT_SYMBOL_GPL(wimax_msg_len); + + +/** + * wimax_msg_send - Send a pre-allocated message to user space + * + * @wimax_dev: WiMAX device descriptor + * + * @skb: &struct sk_buff returned by wimax_msg_alloc(). Note the + * ownership of @skb is transferred to this function. + * + * Returns: 0 if ok, < 0 errno code on error + * + * Description: + * + * Sends a free-form message that was preallocated with + * wimax_msg_alloc() and filled up. + * + * Assumes that once you pass an skb to this function for sending, it + * owns it and will release it when done (on success). + * + * IMPORTANT: + * + * Don't use skb_push()/skb_pull()/skb_reserve() on the skb, as + * wimax_msg_send() depends on skb->data being placed at the + * beginning of the user message. + */ +int wimax_msg_send(struct wimax_dev *wimax_dev, struct sk_buff *skb) +{ + int result; + struct device *dev = wimax_dev->net_dev->dev.parent; + void *msg = skb->data; + size_t size = skb->len; + might_sleep(); + + d_printf(1, dev, "CTX: wimax msg, %zu bytes\n", size); + d_dump(2, dev, msg, size); + result = genlmsg_multicast(skb, 0, wimax_gnl_mcg.id, GFP_KERNEL); + d_printf(1, dev, "CTX: genl multicast result %d\n", result); + if (result == -ESRCH) /* Nobody connected, ignore it */ + result = 0; /* btw, the skb is freed already */ + return result; +} +EXPORT_SYMBOL_GPL(wimax_msg_send); + + +/** + * wimax_msg - Send a message to user space + * + * @wimax_dev: WiMAX device descriptor (properly referenced) + * @pipe_name: "named pipe" the message will be sent to + * @buf: pointer to the message to send. + * @size: size of the buffer pointed to by @buf (in bytes). + * @gfp_flags: flags for memory allocation. + * + * Returns: %0 if ok, negative errno code on error. + * + * Description: + * + * Sends a free-form message to user space on the device @wimax_dev. + * + * NOTES: + * + * Once the @skb is given to this function, who will own it and will + * release it when done (unless it returns error). + */ +int wimax_msg(struct wimax_dev *wimax_dev, const char *pipe_name, + const void *buf, size_t size, gfp_t gfp_flags) +{ + int result = -ENOMEM; + struct sk_buff *skb; + + skb = wimax_msg_alloc(wimax_dev, pipe_name, buf, size, gfp_flags); + if (skb == NULL) + goto error_msg_new; + result = wimax_msg_send(wimax_dev, skb); +error_msg_new: + return result; +} +EXPORT_SYMBOL_GPL(wimax_msg); + + +static const +struct nla_policy wimax_gnl_msg_policy[WIMAX_GNL_ATTR_MAX + 1] = { + [WIMAX_GNL_MSG_IFIDX] = { + .type = NLA_U32, + }, + [WIMAX_GNL_MSG_DATA] = { + .type = NLA_UNSPEC, /* libnl doesn't grok BINARY yet */ + }, +}; + + +/* + * Relays a message from user space to the driver + * + * The skb is passed to the driver-specific function with the netlink + * and generic netlink headers already stripped. + * + * This call will block while handling/relaying the message. + */ +static +int wimax_gnl_doit_msg_from_user(struct sk_buff *skb, struct genl_info *info) +{ + int result, ifindex; + struct wimax_dev *wimax_dev; + struct device *dev; + struct nlmsghdr *nlh = info->nlhdr; + char *pipe_name; + void *msg_buf; + size_t msg_len; + + might_sleep(); + d_fnstart(3, NULL, "(skb %p info %p)\n", skb, info); + result = -ENODEV; + if (info->attrs[WIMAX_GNL_MSG_IFIDX] == NULL) { + printk(KERN_ERR "WIMAX_GNL_MSG_FROM_USER: can't find IFIDX " + "attribute\n"); + goto error_no_wimax_dev; + } + ifindex = nla_get_u32(info->attrs[WIMAX_GNL_MSG_IFIDX]); + wimax_dev = wimax_dev_get_by_genl_info(info, ifindex); + if (wimax_dev == NULL) + goto error_no_wimax_dev; + dev = wimax_dev_to_dev(wimax_dev); + + /* Unpack arguments */ + result = -EINVAL; + if (info->attrs[WIMAX_GNL_MSG_DATA] == NULL) { + dev_err(dev, "WIMAX_GNL_MSG_FROM_USER: can't find MSG_DATA " + "attribute\n"); + goto error_no_data; + } + msg_buf = nla_data(info->attrs[WIMAX_GNL_MSG_DATA]); + msg_len = nla_len(info->attrs[WIMAX_GNL_MSG_DATA]); + + if (info->attrs[WIMAX_GNL_MSG_PIPE_NAME] == NULL) + pipe_name = NULL; + else { + struct nlattr *attr = info->attrs[WIMAX_GNL_MSG_PIPE_NAME]; + size_t attr_len = nla_len(attr); + /* libnl-1.1 does not yet support NLA_NUL_STRING */ + result = -ENOMEM; + pipe_name = kstrndup(nla_data(attr), attr_len + 1, GFP_KERNEL); + if (pipe_name == NULL) + goto error_alloc; + pipe_name[attr_len] = 0; + } + mutex_lock(&wimax_dev->mutex); + result = wimax_dev_is_ready(wimax_dev); + if (result < 0) + goto error_not_ready; + result = -ENOSYS; + if (wimax_dev->op_msg_from_user == NULL) + goto error_noop; + + d_printf(1, dev, + "CRX: nlmsghdr len %u type %u flags 0x%04x seq 0x%x pid %u\n", + nlh->nlmsg_len, nlh->nlmsg_type, nlh->nlmsg_flags, + nlh->nlmsg_seq, nlh->nlmsg_pid); + d_printf(1, dev, "CRX: wimax message %zu bytes\n", msg_len); + d_dump(2, dev, msg_buf, msg_len); + + result = wimax_dev->op_msg_from_user(wimax_dev, pipe_name, + msg_buf, msg_len, info); +error_noop: +error_not_ready: + mutex_unlock(&wimax_dev->mutex); +error_alloc: + kfree(pipe_name); +error_no_data: + dev_put(wimax_dev->net_dev); +error_no_wimax_dev: + d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result); + return result; +} + + +/* + * Generic Netlink glue + */ + +struct genl_ops wimax_gnl_msg_from_user = { + .cmd = WIMAX_GNL_OP_MSG_FROM_USER, + .flags = GENL_ADMIN_PERM, + .policy = wimax_gnl_msg_policy, + .doit = wimax_gnl_doit_msg_from_user, + .dumpit = NULL, +}; + diff --git a/net/wimax/op-reset.c b/net/wimax/op-reset.c new file mode 100644 index 000000000000..ca269178c4d4 --- /dev/null +++ b/net/wimax/op-reset.c @@ -0,0 +1,143 @@ +/* + * Linux WiMAX + * Implement and export a method for resetting a WiMAX device + * + * + * Copyright (C) 2008 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * This implements a simple synchronous call to reset a WiMAX device. + * + * Resets aim at being warm, keeping the device handles active; + * however, when that fails, it falls back to a cold reset (that will + * disconnect and reconnect the device). + */ + +#include +#include +#include +#include +#include "wimax-internal.h" + +#define D_SUBMODULE op_reset +#include "debug-levels.h" + + +/** + * wimax_reset - Reset a WiMAX device + * + * @wimax_dev: WiMAX device descriptor + * + * Returns: + * + * %0 if ok and a warm reset was done (the device still exists in + * the system). + * + * -%ENODEV if a cold/bus reset had to be done (device has + * disconnected and reconnected, so current handle is not valid + * any more). + * + * -%EINVAL if the device is not even registered. + * + * Any other negative error code shall be considered as + * non-recoverable. + * + * Description: + * + * Called when wanting to reset the device for any reason. Device is + * taken back to power on status. + * + * This call blocks; on succesful return, the device has completed the + * reset process and is ready to operate. + */ +int wimax_reset(struct wimax_dev *wimax_dev) +{ + int result = -EINVAL; + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_st state; + + might_sleep(); + d_fnstart(3, dev, "(wimax_dev %p)\n", wimax_dev); + mutex_lock(&wimax_dev->mutex); + dev_hold(wimax_dev->net_dev); + state = wimax_dev->state; + mutex_unlock(&wimax_dev->mutex); + + if (state >= WIMAX_ST_DOWN) { + mutex_lock(&wimax_dev->mutex_reset); + result = wimax_dev->op_reset(wimax_dev); + mutex_unlock(&wimax_dev->mutex_reset); + } + dev_put(wimax_dev->net_dev); + + d_fnend(3, dev, "(wimax_dev %p) = %d\n", wimax_dev, result); + return result; +} +EXPORT_SYMBOL(wimax_reset); + + +static const +struct nla_policy wimax_gnl_reset_policy[WIMAX_GNL_ATTR_MAX + 1] = { + [WIMAX_GNL_RESET_IFIDX] = { + .type = NLA_U32, + }, +}; + + +/* + * Exporting to user space over generic netlink + * + * Parse the reset command from user space, return error code. + * + * No attributes. + */ +static +int wimax_gnl_doit_reset(struct sk_buff *skb, struct genl_info *info) +{ + int result, ifindex; + struct wimax_dev *wimax_dev; + struct device *dev; + + d_fnstart(3, NULL, "(skb %p info %p)\n", skb, info); + result = -ENODEV; + if (info->attrs[WIMAX_GNL_RESET_IFIDX] == NULL) { + printk(KERN_ERR "WIMAX_GNL_OP_RFKILL: can't find IFIDX " + "attribute\n"); + goto error_no_wimax_dev; + } + ifindex = nla_get_u32(info->attrs[WIMAX_GNL_RESET_IFIDX]); + wimax_dev = wimax_dev_get_by_genl_info(info, ifindex); + if (wimax_dev == NULL) + goto error_no_wimax_dev; + dev = wimax_dev_to_dev(wimax_dev); + /* Execute the operation and send the result back to user space */ + result = wimax_reset(wimax_dev); + dev_put(wimax_dev->net_dev); +error_no_wimax_dev: + d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result); + return result; +} + + +struct genl_ops wimax_gnl_reset = { + .cmd = WIMAX_GNL_OP_RESET, + .flags = GENL_ADMIN_PERM, + .policy = wimax_gnl_reset_policy, + .doit = wimax_gnl_doit_reset, + .dumpit = NULL, +}; diff --git a/net/wimax/op-rfkill.c b/net/wimax/op-rfkill.c new file mode 100644 index 000000000000..8745bac173f1 --- /dev/null +++ b/net/wimax/op-rfkill.c @@ -0,0 +1,532 @@ +/* + * Linux WiMAX + * RF-kill framework integration + * + * + * Copyright (C) 2008 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + * + * + * This integrates into the Linux Kernel rfkill susbystem so that the + * drivers just have to do the bare minimal work, which is providing a + * method to set the software RF-Kill switch and to report changes in + * the software and hardware switch status. + * + * A non-polled generic rfkill device is embedded into the WiMAX + * subsystem's representation of a device. + * + * FIXME: Need polled support? use a timer or add the implementation + * to the stack. + * + * All device drivers have to do is after wimax_dev_init(), call + * wimax_report_rfkill_hw() and wimax_report_rfkill_sw() to update + * initial state and then every time it changes. See wimax.h:struct + * wimax_dev for more information. + * + * ROADMAP + * + * wimax_gnl_doit_rfkill() User space calling wimax_rfkill() + * wimax_rfkill() Kernel calling wimax_rfkill() + * __wimax_rf_toggle_radio() + * + * wimax_rfkill_toggle_radio() RF-Kill subsytem calling + * __wimax_rf_toggle_radio() + * + * __wimax_rf_toggle_radio() + * wimax_dev->op_rfkill_sw_toggle() Driver backend + * __wimax_state_change() + * + * wimax_report_rfkill_sw() Driver reports state change + * __wimax_state_change() + * + * wimax_report_rfkill_hw() Driver reports state change + * __wimax_state_change() + * + * wimax_rfkill_add() Initialize/shutdown rfkill support + * wimax_rfkill_rm() [called by wimax_dev_add/rm()] + */ + +#include +#include +#include +#include +#include +#include +#include "wimax-internal.h" + +#define D_SUBMODULE op_rfkill +#include "debug-levels.h" + +#ifdef CONFIG_RFKILL + + +/** + * wimax_report_rfkill_hw - Reports changes in the hardware RF switch + * + * @wimax_dev: WiMAX device descriptor + * + * @state: New state of the RF Kill switch. %WIMAX_RF_ON radio on, + * %WIMAX_RF_OFF radio off. + * + * When the device detects a change in the state of thehardware RF + * switch, it must call this function to let the WiMAX kernel stack + * know that the state has changed so it can be properly propagated. + * + * The WiMAX stack caches the state (the driver doesn't need to). As + * well, as the change is propagated it will come back as a request to + * change the software state to mirror the hardware state. + * + * If the device doesn't have a hardware kill switch, just report + * it on initialization as always on (%WIMAX_RF_ON, radio on). + */ +void wimax_report_rfkill_hw(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ + int result; + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_st wimax_state; + enum rfkill_state rfkill_state; + + d_fnstart(3, dev, "(wimax_dev %p state %u)\n", wimax_dev, state); + BUG_ON(state == WIMAX_RF_QUERY); + BUG_ON(state != WIMAX_RF_ON && state != WIMAX_RF_OFF); + + mutex_lock(&wimax_dev->mutex); + result = wimax_dev_is_ready(wimax_dev); + if (result < 0) + goto error_not_ready; + + if (state != wimax_dev->rf_hw) { + wimax_dev->rf_hw = state; + rfkill_state = state == WIMAX_RF_ON ? + RFKILL_STATE_OFF : RFKILL_STATE_ON; + if (wimax_dev->rf_hw == WIMAX_RF_ON + && wimax_dev->rf_sw == WIMAX_RF_ON) + wimax_state = WIMAX_ST_READY; + else + wimax_state = WIMAX_ST_RADIO_OFF; + __wimax_state_change(wimax_dev, wimax_state); + input_report_key(wimax_dev->rfkill_input, KEY_WIMAX, + rfkill_state); + } +error_not_ready: + mutex_unlock(&wimax_dev->mutex); + d_fnend(3, dev, "(wimax_dev %p state %u) = void [%d]\n", + wimax_dev, state, result); +} +EXPORT_SYMBOL_GPL(wimax_report_rfkill_hw); + + +/** + * wimax_report_rfkill_sw - Reports changes in the software RF switch + * + * @wimax_dev: WiMAX device descriptor + * + * @state: New state of the RF kill switch. %WIMAX_RF_ON radio on, + * %WIMAX_RF_OFF radio off. + * + * Reports changes in the software RF switch state to the the WiMAX + * stack. + * + * The main use is during initialization, so the driver can query the + * device for its current software radio kill switch state and feed it + * to the system. + * + * On the side, the device does not change the software state by + * itself. In practice, this can happen, as the device might decide to + * switch (in software) the radio off for different reasons. + */ +void wimax_report_rfkill_sw(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ + int result; + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_st wimax_state; + + d_fnstart(3, dev, "(wimax_dev %p state %u)\n", wimax_dev, state); + BUG_ON(state == WIMAX_RF_QUERY); + BUG_ON(state != WIMAX_RF_ON && state != WIMAX_RF_OFF); + + mutex_lock(&wimax_dev->mutex); + result = wimax_dev_is_ready(wimax_dev); + if (result < 0) + goto error_not_ready; + + if (state != wimax_dev->rf_sw) { + wimax_dev->rf_sw = state; + if (wimax_dev->rf_hw == WIMAX_RF_ON + && wimax_dev->rf_sw == WIMAX_RF_ON) + wimax_state = WIMAX_ST_READY; + else + wimax_state = WIMAX_ST_RADIO_OFF; + __wimax_state_change(wimax_dev, wimax_state); + } +error_not_ready: + mutex_unlock(&wimax_dev->mutex); + d_fnend(3, dev, "(wimax_dev %p state %u) = void [%d]\n", + wimax_dev, state, result); +} +EXPORT_SYMBOL_GPL(wimax_report_rfkill_sw); + + +/* + * Callback for the RF Kill toggle operation + * + * This function is called by: + * + * - The rfkill subsystem when the RF-Kill key is pressed in the + * hardware and the driver notifies through + * wimax_report_rfkill_hw(). The rfkill subsystem ends up calling back + * here so the software RF Kill switch state is changed to reflect + * the hardware switch state. + * + * - When the user sets the state through sysfs' rfkill/state file + * + * - When the user calls wimax_rfkill(). + * + * This call blocks! + * + * WARNING! When we call rfkill_unregister(), this will be called with + * state 0! + * + * WARNING: wimax_dev must be locked + */ +static +int __wimax_rf_toggle_radio(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ + int result = 0; + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_st wimax_state; + + might_sleep(); + d_fnstart(3, dev, "(wimax_dev %p state %u)\n", wimax_dev, state); + if (wimax_dev->rf_sw == state) + goto out_no_change; + if (wimax_dev->op_rfkill_sw_toggle != NULL) + result = wimax_dev->op_rfkill_sw_toggle(wimax_dev, state); + else if (state == WIMAX_RF_OFF) /* No op? can't turn off */ + result = -ENXIO; + else /* No op? can turn on */ + result = 0; /* should never happen tho */ + if (result >= 0) { + result = 0; + wimax_dev->rf_sw = state; + wimax_state = state == WIMAX_RF_ON ? + WIMAX_ST_READY : WIMAX_ST_RADIO_OFF; + __wimax_state_change(wimax_dev, wimax_state); + } +out_no_change: + d_fnend(3, dev, "(wimax_dev %p state %u) = %d\n", + wimax_dev, state, result); + return result; +} + + +/* + * Translate from rfkill state to wimax state + * + * NOTE: Special state handling rules here + * + * Just pretend the call didn't happen if we are in a state where + * we know for sure it cannot be handled (WIMAX_ST_DOWN or + * __WIMAX_ST_QUIESCING). rfkill() needs it to register and + * unregister, as it will run this path. + * + * NOTE: This call will block until the operation is completed. + */ +static +int wimax_rfkill_toggle_radio(void *data, enum rfkill_state state) +{ + int result; + struct wimax_dev *wimax_dev = data; + struct device *dev = wimax_dev_to_dev(wimax_dev); + enum wimax_rf_state rf_state; + + d_fnstart(3, dev, "(wimax_dev %p state %u)\n", wimax_dev, state); + switch (state) { + case RFKILL_STATE_ON: + rf_state = WIMAX_RF_OFF; + break; + case RFKILL_STATE_OFF: + rf_state = WIMAX_RF_ON; + break; + default: + BUG(); + } + mutex_lock(&wimax_dev->mutex); + if (wimax_dev->state <= __WIMAX_ST_QUIESCING) + result = 0; /* just pretend it didn't happen */ + else + result = __wimax_rf_toggle_radio(wimax_dev, rf_state); + mutex_unlock(&wimax_dev->mutex); + d_fnend(3, dev, "(wimax_dev %p state %u) = %d\n", + wimax_dev, state, result); + return result; +} + + +/** + * wimax_rfkill - Set the software RF switch state for a WiMAX device + * + * @wimax_dev: WiMAX device descriptor + * + * @state: New RF state. + * + * Returns: + * + * >= 0 toggle state if ok, < 0 errno code on error. The toggle state + * is returned as a bitmap, bit 0 being the hardware RF state, bit 1 + * the software RF state. + * + * 0 means disabled (%WIMAX_RF_ON, radio on), 1 means enabled radio + * off (%WIMAX_RF_OFF). + * + * Description: + * + * Called by the user when he wants to request the WiMAX radio to be + * switched on (%WIMAX_RF_ON) or off (%WIMAX_RF_OFF). With + * %WIMAX_RF_QUERY, just the current state is returned. + * + * NOTE: + * + * This call will block until the operation is complete. + */ +int wimax_rfkill(struct wimax_dev *wimax_dev, enum wimax_rf_state state) +{ + int result; + struct device *dev = wimax_dev_to_dev(wimax_dev); + + d_fnstart(3, dev, "(wimax_dev %p state %u)\n", wimax_dev, state); + mutex_lock(&wimax_dev->mutex); + result = wimax_dev_is_ready(wimax_dev); + if (result < 0) + goto error_not_ready; + switch (state) { + case WIMAX_RF_ON: + case WIMAX_RF_OFF: + result = __wimax_rf_toggle_radio(wimax_dev, state); + if (result < 0) + goto error; + break; + case WIMAX_RF_QUERY: + break; + default: + result = -EINVAL; + goto error; + } + result = wimax_dev->rf_sw << 1 | wimax_dev->rf_hw; +error: +error_not_ready: + mutex_unlock(&wimax_dev->mutex); + d_fnend(3, dev, "(wimax_dev %p state %u) = %d\n", + wimax_dev, state, result); + return result; +} +EXPORT_SYMBOL(wimax_rfkill); + + +/* + * Register a new WiMAX device's RF Kill support + * + * WARNING: wimax_dev->mutex must be unlocked + */ +int wimax_rfkill_add(struct wimax_dev *wimax_dev) +{ + int result; + struct rfkill *rfkill; + struct input_dev *input_dev; + struct device *dev = wimax_dev_to_dev(wimax_dev); + + d_fnstart(3, dev, "(wimax_dev %p)\n", wimax_dev); + /* Initialize RF Kill */ + result = -ENOMEM; + rfkill = rfkill_allocate(dev, RFKILL_TYPE_WIMAX); + if (rfkill == NULL) + goto error_rfkill_allocate; + wimax_dev->rfkill = rfkill; + + rfkill->name = wimax_dev->name; + rfkill->state = RFKILL_STATE_OFF; + rfkill->data = wimax_dev; + rfkill->toggle_radio = wimax_rfkill_toggle_radio; + rfkill->user_claim_unsupported = 1; + + /* Initialize the input device for the hw key */ + input_dev = input_allocate_device(); + if (input_dev == NULL) + goto error_input_allocate; + wimax_dev->rfkill_input = input_dev; + d_printf(1, dev, "rfkill %p input %p\n", rfkill, input_dev); + + input_dev->name = wimax_dev->name; + /* FIXME: get a real device bus ID and stuff? do we care? */ + input_dev->id.bustype = BUS_HOST; + input_dev->id.vendor = 0xffff; + input_dev->evbit[0] = BIT(EV_KEY); + set_bit(KEY_WIMAX, input_dev->keybit); + + /* Register both */ + result = input_register_device(wimax_dev->rfkill_input); + if (result < 0) + goto error_input_register; + result = rfkill_register(wimax_dev->rfkill); + if (result < 0) + goto error_rfkill_register; + + /* If there is no SW toggle op, SW RFKill is always on */ + if (wimax_dev->op_rfkill_sw_toggle == NULL) + wimax_dev->rf_sw = WIMAX_RF_ON; + + d_fnend(3, dev, "(wimax_dev %p) = 0\n", wimax_dev); + return 0; + + /* if rfkill_register() suceeds, can't use rfkill_free() any + * more, only rfkill_unregister() [it owns the refcount]; with + * the input device we have the same issue--hence the if. */ +error_rfkill_register: + input_unregister_device(wimax_dev->rfkill_input); + wimax_dev->rfkill_input = NULL; +error_input_register: + if (wimax_dev->rfkill_input) + input_free_device(wimax_dev->rfkill_input); +error_input_allocate: + rfkill_free(wimax_dev->rfkill); +error_rfkill_allocate: + d_fnend(3, dev, "(wimax_dev %p) = %d\n", wimax_dev, result); + return result; +} + + +/* + * Deregister a WiMAX device's RF Kill support + * + * Ick, we can't call rfkill_free() after rfkill_unregister()...oh + * well. + * + * WARNING: wimax_dev->mutex must be unlocked + */ +void wimax_rfkill_rm(struct wimax_dev *wimax_dev) +{ + struct device *dev = wimax_dev_to_dev(wimax_dev); + d_fnstart(3, dev, "(wimax_dev %p)\n", wimax_dev); + rfkill_unregister(wimax_dev->rfkill); /* frees */ + input_unregister_device(wimax_dev->rfkill_input); + d_fnend(3, dev, "(wimax_dev %p)\n", wimax_dev); +} + + +#else /* #ifdef CONFIG_RFKILL */ + +void wimax_report_rfkill_hw(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ +} +EXPORT_SYMBOL_GPL(wimax_report_rfkill_hw); + +void wimax_report_rfkill_sw(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ +} +EXPORT_SYMBOL_GPL(wimax_report_rfkill_sw); + +int wimax_rfkill(struct wimax_dev *wimax_dev, + enum wimax_rf_state state) +{ + return WIMAX_RF_ON << 1 | WIMAX_RF_ON; +} +EXPORT_SYMBOL_GPL(wimax_rfkill); + +int wimax_rfkill_add(struct wimax_dev *wimax_dev) +{ + return 0; +} + +void wimax_rfkill_rm(struct wimax_dev *wimax_dev) +{ +} + +#endif /* #ifdef CONFIG_RFKILL */ + + +/* + * Exporting to user space over generic netlink + * + * Parse the rfkill command from user space, return a combination + * value that describe the states of the different toggles. + * + * Only one attribute: the new state requested (on, off or no change, + * just query). + */ + +static const +struct nla_policy wimax_gnl_rfkill_policy[WIMAX_GNL_ATTR_MAX + 1] = { + [WIMAX_GNL_RFKILL_IFIDX] = { + .type = NLA_U32, + }, + [WIMAX_GNL_RFKILL_STATE] = { + .type = NLA_U32 /* enum wimax_rf_state */ + }, +}; + + +static +int wimax_gnl_doit_rfkill(struct sk_buff *skb, struct genl_info *info) +{ + int result, ifindex; + struct wimax_dev *wimax_dev; + struct device *dev; + enum wimax_rf_state new_state; + + d_fnstart(3, NULL, "(skb %p info %p)\n", skb, info); + result = -ENODEV; + if (info->attrs[WIMAX_GNL_RFKILL_IFIDX] == NULL) { + printk(KERN_ERR "WIMAX_GNL_OP_RFKILL: can't find IFIDX " + "attribute\n"); + goto error_no_wimax_dev; + } + ifindex = nla_get_u32(info->attrs[WIMAX_GNL_RFKILL_IFIDX]); + wimax_dev = wimax_dev_get_by_genl_info(info, ifindex); + if (wimax_dev == NULL) + goto error_no_wimax_dev; + dev = wimax_dev_to_dev(wimax_dev); + result = -EINVAL; + if (info->attrs[WIMAX_GNL_RFKILL_STATE] == NULL) { + dev_err(dev, "WIMAX_GNL_RFKILL: can't find RFKILL_STATE " + "attribute\n"); + goto error_no_pid; + } + new_state = nla_get_u32(info->attrs[WIMAX_GNL_RFKILL_STATE]); + + /* Execute the operation and send the result back to user space */ + result = wimax_rfkill(wimax_dev, new_state); +error_no_pid: + dev_put(wimax_dev->net_dev); +error_no_wimax_dev: + d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result); + return result; +} + + +struct genl_ops wimax_gnl_rfkill = { + .cmd = WIMAX_GNL_OP_RFKILL, + .flags = GENL_ADMIN_PERM, + .policy = wimax_gnl_rfkill_policy, + .doit = wimax_gnl_doit_rfkill, + .dumpit = NULL, +}; + -- cgit v1.2.3 From 617209ccf73b571953cf4c76519c65a5e52d15fd Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Sat, 20 Dec 2008 16:57:40 -0800 Subject: wimax: debugfs controls Expose knobs to control the stack's debug output. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- net/wimax/debugfs.c | 90 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 90 insertions(+) create mode 100644 net/wimax/debugfs.c (limited to 'net') diff --git a/net/wimax/debugfs.c b/net/wimax/debugfs.c new file mode 100644 index 000000000000..87cf4430079c --- /dev/null +++ b/net/wimax/debugfs.c @@ -0,0 +1,90 @@ +/* + * Linux WiMAX + * Debugfs support + * + * + * Copyright (C) 2005-2006 Intel Corporation + * Inaky Perez-Gonzalez + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA. + */ +#include +#include +#include "wimax-internal.h" + +#define D_SUBMODULE debugfs +#include "debug-levels.h" + + +/* Debug framework control of debug levels */ +struct d_level D_LEVEL[] = { + D_SUBMODULE_DEFINE(debugfs), + D_SUBMODULE_DEFINE(id_table), + D_SUBMODULE_DEFINE(op_msg), + D_SUBMODULE_DEFINE(op_reset), + D_SUBMODULE_DEFINE(op_rfkill), + D_SUBMODULE_DEFINE(stack), +}; +size_t D_LEVEL_SIZE = ARRAY_SIZE(D_LEVEL); + +#define __debugfs_register(prefix, name, parent) \ +do { \ + result = d_level_register_debugfs(prefix, name, parent); \ + if (result < 0) \ + goto error; \ +} while (0) + + +int wimax_debugfs_add(struct wimax_dev *wimax_dev) +{ + int result; + struct net_device *net_dev = wimax_dev->net_dev; + struct device *dev = net_dev->dev.parent; + struct dentry *dentry; + char buf[128]; + + snprintf(buf, sizeof(buf), "wimax:%s", net_dev->name); + dentry = debugfs_create_dir(buf, NULL); + result = PTR_ERR(dentry); + if (IS_ERR(dentry)) { + if (result == -ENODEV) + result = 0; /* No debugfs support */ + else + dev_err(dev, "Can't create debugfs dentry: %d\n", + result); + goto out; + } + wimax_dev->debugfs_dentry = dentry; + __debugfs_register("wimax_dl_", debugfs, dentry); + __debugfs_register("wimax_dl_", id_table, dentry); + __debugfs_register("wimax_dl_", op_msg, dentry); + __debugfs_register("wimax_dl_", op_reset, dentry); + __debugfs_register("wimax_dl_", op_rfkill, dentry); + __debugfs_register("wimax_dl_", stack, dentry); + result = 0; +out: + return result; + +error: + debugfs_remove_recursive(wimax_dev->debugfs_dentry); + return result; +} + +void wimax_debugfs_rm(struct wimax_dev *wimax_dev) +{ + debugfs_remove_recursive(wimax_dev->debugfs_dentry); +} + + -- cgit v1.2.3 From b0c83ae1de01880075955c7224e751440688ec74 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Tue, 23 Dec 2008 16:18:24 -0800 Subject: wimax: Makefile, Kconfig and docbook linkage for the stack This patch provides Makefile and KConfig for the WiMAX stack, integrating them into the networking stack's Makefile, Kconfig and doc-book templates. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- Documentation/DocBook/networking.tmpl | 8 ++++++++ net/Kconfig | 2 ++ net/Makefile | 1 + net/wimax/Kconfig | 38 +++++++++++++++++++++++++++++++++++ net/wimax/Makefile | 13 ++++++++++++ 5 files changed, 62 insertions(+) create mode 100644 net/wimax/Kconfig create mode 100644 net/wimax/Makefile (limited to 'net') diff --git a/Documentation/DocBook/networking.tmpl b/Documentation/DocBook/networking.tmpl index 627707a3cb9d..59ad69a9d777 100644 --- a/Documentation/DocBook/networking.tmpl +++ b/Documentation/DocBook/networking.tmpl @@ -74,6 +74,14 @@ !Enet/sunrpc/rpcb_clnt.c !Enet/sunrpc/clnt.c + WiMAX +!Enet/wimax/op-msg.c +!Enet/wimax/op-reset.c +!Enet/wimax/op-rfkill.c +!Enet/wimax/stack.c +!Iinclude/net/wimax.h +!Iinclude/linux/wimax.h + diff --git a/net/Kconfig b/net/Kconfig index 6ec2cce7c167..bf2776018f71 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -254,6 +254,8 @@ source "net/mac80211/Kconfig" endif # WIRELESS +source "net/wimax/Kconfig" + source "net/rfkill/Kconfig" source "net/9p/Kconfig" diff --git a/net/Makefile b/net/Makefile index ba4460432b7c..0fcce89d7169 100644 --- a/net/Makefile +++ b/net/Makefile @@ -63,3 +63,4 @@ endif ifeq ($(CONFIG_NET),y) obj-$(CONFIG_SYSCTL) += sysctl_net.o endif +obj-$(CONFIG_WIMAX) += wimax/ diff --git a/net/wimax/Kconfig b/net/wimax/Kconfig new file mode 100644 index 000000000000..0bdbb6928205 --- /dev/null +++ b/net/wimax/Kconfig @@ -0,0 +1,38 @@ +# +# WiMAX LAN device configuration +# + +menuconfig WIMAX + tristate "WiMAX Wireless Broadband support" + help + + Select to configure support for devices that provide + wireless broadband connectivity using the WiMAX protocol + (IEEE 802.16). + + Please note that most of these devices require signing up + for a service plan with a provider. + + The different WiMAX drivers can be enabled in the menu entry + + Device Drivers > Network device support > WiMAX Wireless + Broadband devices + + If unsure, it is safe to select M (module). + +config WIMAX_DEBUG_LEVEL + int "WiMAX debug level" + depends on WIMAX + default 8 + help + + Select the maximum debug verbosity level to be compiled into + the WiMAX stack code. + + By default, debug messages are disabled at runtime and can + be selectively enabled for different parts of the code using + the sysfs debug-levels file. + + If set at zero, this will compile out all the debug code. + + It is recommended that it is left at 8. diff --git a/net/wimax/Makefile b/net/wimax/Makefile new file mode 100644 index 000000000000..5b80b941c2c9 --- /dev/null +++ b/net/wimax/Makefile @@ -0,0 +1,13 @@ + +obj-$(CONFIG_WIMAX) += wimax.o + +wimax-y := \ + id-table.o \ + op-msg.o \ + op-reset.o \ + op-rfkill.o \ + stack.o + +wimax-$(CONFIG_DEBUG_FS) += debugfs.o + + -- cgit v1.2.3 From 2779e3ae39645515cb6c1126634f47c28c9e7190 Mon Sep 17 00:00:00 2001 From: Tom Tucker Date: Mon, 5 Jan 2009 11:12:52 -0600 Subject: svc: Move kfree of deferral record to common code The rqstp structure has a pointer to a svc_deferred_req record that is allocated when requests are deferred. This record is common to all transports and can be freed in common code. Move the kfree of the rq_deferred to the common svc_xprt_release function. This also fixes a memory leak in the RDMA transport which does not kfree the dr structure in it's version of the xpo_release_rqst callback. Signed-off-by: Tom Tucker Signed-off-by: J. Bruce Fields --- net/sunrpc/svc_xprt.c | 3 +++ net/sunrpc/svcsock.c | 5 ----- 2 files changed, 3 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 3fe4f1004278..29619612b9f1 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -448,6 +448,9 @@ static void svc_xprt_release(struct svc_rqst *rqstp) rqstp->rq_xprt->xpt_ops->xpo_release_rqst(rqstp); + kfree(rqstp->rq_deferred); + rqstp->rq_deferred = NULL; + svc_free_res_pages(rqstp); rqstp->rq_res.page_len = 0; rqstp->rq_res.page_base = 0; diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index cccfa7deb9af..3c103404501d 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -103,7 +103,6 @@ static void svc_reclassify_socket(struct socket *sock) static void svc_release_skb(struct svc_rqst *rqstp) { struct sk_buff *skb = rqstp->rq_xprt_ctxt; - struct svc_deferred_req *dr = rqstp->rq_deferred; if (skb) { struct svc_sock *svsk = @@ -113,10 +112,6 @@ static void svc_release_skb(struct svc_rqst *rqstp) dprintk("svc: service %p, releasing skb %p\n", rqstp, skb); skb_free_datagram(svsk->sk_sk, skb); } - if (dr) { - rqstp->rq_deferred = NULL; - kfree(dr); - } } union svc_pktinfo_u { -- cgit v1.2.3 From 22945e4a1c7454c97f5d8aee1ef526c83fef3223 Mon Sep 17 00:00:00 2001 From: Tom Tucker Date: Mon, 5 Jan 2009 15:21:19 -0600 Subject: svc: Clean up deferred requests on transport destruction A race between svc_revisit and svc_delete_xprt can result in deferred requests holding references on a transport that can never be recovered because dead transports are not enqueued for subsequent processing. Check for XPT_DEAD in revisit to clean up completing deferrals on a dead transport and sweep a transport's deferred queue to do the same for queued but unprocessed deferrals. Signed-off-by: Tom Tucker Signed-off-by: J. Bruce Fields --- net/sunrpc/svc_xprt.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 29619612b9f1..a78b87937c73 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -850,6 +850,11 @@ static void svc_age_temp_xprts(unsigned long closure) void svc_delete_xprt(struct svc_xprt *xprt) { struct svc_serv *serv = xprt->xpt_server; + struct svc_deferred_req *dr; + + /* Only do this once */ + if (test_and_set_bit(XPT_DEAD, &xprt->xpt_flags)) + return; dprintk("svc: svc_delete_xprt(%p)\n", xprt); xprt->xpt_ops->xpo_detach(xprt); @@ -864,12 +869,16 @@ void svc_delete_xprt(struct svc_xprt *xprt) * while still attached to a queue, the queue itself * is about to be destroyed (in svc_destroy). */ - if (!test_and_set_bit(XPT_DEAD, &xprt->xpt_flags)) { - BUG_ON(atomic_read(&xprt->xpt_ref.refcount) < 2); - if (test_bit(XPT_TEMP, &xprt->xpt_flags)) - serv->sv_tmpcnt--; + if (test_bit(XPT_TEMP, &xprt->xpt_flags)) + serv->sv_tmpcnt--; + + for (dr = svc_deferred_dequeue(xprt); dr; + dr = svc_deferred_dequeue(xprt)) { svc_xprt_put(xprt); + kfree(dr); } + + svc_xprt_put(xprt); spin_unlock_bh(&serv->sv_lock); } @@ -915,17 +924,19 @@ static void svc_revisit(struct cache_deferred_req *dreq, int too_many) container_of(dreq, struct svc_deferred_req, handle); struct svc_xprt *xprt = dr->xprt; - if (too_many) { + spin_lock(&xprt->xpt_lock); + set_bit(XPT_DEFERRED, &xprt->xpt_flags); + if (too_many || test_bit(XPT_DEAD, &xprt->xpt_flags)) { + spin_unlock(&xprt->xpt_lock); + dprintk("revisit canceled\n"); svc_xprt_put(xprt); kfree(dr); return; } dprintk("revisit queued\n"); dr->xprt = NULL; - spin_lock(&xprt->xpt_lock); list_add(&dr->handle.recent, &xprt->xpt_deferred); spin_unlock(&xprt->xpt_lock); - set_bit(XPT_DEFERRED, &xprt->xpt_flags); svc_xprt_enqueue(xprt); svc_xprt_put(xprt); } -- cgit v1.2.3 From 24c3767e41a6a59d32bb45abe899eb194e6bf1b8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 23 Dec 2008 16:30:12 -0500 Subject: SUNRPC: The sunrpc server code should not be used by out-of-tree modules Signed-off-by: Trond Myklebust Signed-off-by: J. Bruce Fields --- net/sunrpc/cache.c | 20 ++++++++++---------- net/sunrpc/stats.c | 6 +++--- net/sunrpc/svc.c | 14 +++++++------- net/sunrpc/svc_xprt.c | 8 ++++---- net/sunrpc/svcauth.c | 14 +++++++------- net/sunrpc/svcauth_unix.c | 12 ++++++------ net/sunrpc/svcsock.c | 4 ++-- 7 files changed, 39 insertions(+), 39 deletions(-) (limited to 'net') diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index c9966713282a..4735caad26ed 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -98,7 +98,7 @@ struct cache_head *sunrpc_cache_lookup(struct cache_detail *detail, return new; } -EXPORT_SYMBOL(sunrpc_cache_lookup); +EXPORT_SYMBOL_GPL(sunrpc_cache_lookup); static void queue_loose(struct cache_detail *detail, struct cache_head *ch); @@ -173,7 +173,7 @@ struct cache_head *sunrpc_cache_update(struct cache_detail *detail, cache_put(old, detail); return tmp; } -EXPORT_SYMBOL(sunrpc_cache_update); +EXPORT_SYMBOL_GPL(sunrpc_cache_update); static int cache_make_upcall(struct cache_detail *detail, struct cache_head *h); /* @@ -245,7 +245,7 @@ int cache_check(struct cache_detail *detail, cache_put(h, detail); return rv; } -EXPORT_SYMBOL(cache_check); +EXPORT_SYMBOL_GPL(cache_check); /* * caches need to be periodically cleaned. @@ -373,7 +373,7 @@ int cache_register(struct cache_detail *cd) schedule_delayed_work(&cache_cleaner, 0); return 0; } -EXPORT_SYMBOL(cache_register); +EXPORT_SYMBOL_GPL(cache_register); void cache_unregister(struct cache_detail *cd) { @@ -399,7 +399,7 @@ void cache_unregister(struct cache_detail *cd) out: printk(KERN_ERR "nfsd: failed to unregister %s cache\n", cd->name); } -EXPORT_SYMBOL(cache_unregister); +EXPORT_SYMBOL_GPL(cache_unregister); /* clean cache tries to find something to clean * and cleans it. @@ -514,7 +514,7 @@ void cache_flush(void) while (cache_clean() != -1) cond_resched(); } -EXPORT_SYMBOL(cache_flush); +EXPORT_SYMBOL_GPL(cache_flush); void cache_purge(struct cache_detail *detail) { @@ -523,7 +523,7 @@ void cache_purge(struct cache_detail *detail) cache_flush(); detail->flush_time = 1; } -EXPORT_SYMBOL(cache_purge); +EXPORT_SYMBOL_GPL(cache_purge); /* @@ -988,7 +988,7 @@ void qword_add(char **bpp, int *lp, char *str) *bpp = bp; *lp = len; } -EXPORT_SYMBOL(qword_add); +EXPORT_SYMBOL_GPL(qword_add); void qword_addhex(char **bpp, int *lp, char *buf, int blen) { @@ -1017,7 +1017,7 @@ void qword_addhex(char **bpp, int *lp, char *buf, int blen) *bpp = bp; *lp = len; } -EXPORT_SYMBOL(qword_addhex); +EXPORT_SYMBOL_GPL(qword_addhex); static void warn_no_listener(struct cache_detail *detail) { @@ -1140,7 +1140,7 @@ int qword_get(char **bpp, char *dest, int bufsize) *dest = '\0'; return len; } -EXPORT_SYMBOL(qword_get); +EXPORT_SYMBOL_GPL(qword_get); /* diff --git a/net/sunrpc/stats.c b/net/sunrpc/stats.c index 50b049c6598a..085372ef4feb 100644 --- a/net/sunrpc/stats.c +++ b/net/sunrpc/stats.c @@ -106,7 +106,7 @@ void svc_seq_show(struct seq_file *seq, const struct svc_stat *statp) { seq_putc(seq, '\n'); } } -EXPORT_SYMBOL(svc_seq_show); +EXPORT_SYMBOL_GPL(svc_seq_show); /** * rpc_alloc_iostats - allocate an rpc_iostats structure @@ -249,14 +249,14 @@ svc_proc_register(struct svc_stat *statp, const struct file_operations *fops) { return do_register(statp->program->pg_name, statp, fops); } -EXPORT_SYMBOL(svc_proc_register); +EXPORT_SYMBOL_GPL(svc_proc_register); void svc_proc_unregister(const char *name) { remove_proc_entry(name, proc_net_rpc); } -EXPORT_SYMBOL(svc_proc_unregister); +EXPORT_SYMBOL_GPL(svc_proc_unregister); void rpc_proc_init(void) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 54c98d876847..c51fed4d1af1 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -431,7 +431,7 @@ svc_create(struct svc_program *prog, unsigned int bufsize, { return __svc_create(prog, bufsize, /*npools*/1, family, shutdown); } -EXPORT_SYMBOL(svc_create); +EXPORT_SYMBOL_GPL(svc_create); struct svc_serv * svc_create_pooled(struct svc_program *prog, unsigned int bufsize, @@ -450,7 +450,7 @@ svc_create_pooled(struct svc_program *prog, unsigned int bufsize, return serv; } -EXPORT_SYMBOL(svc_create_pooled); +EXPORT_SYMBOL_GPL(svc_create_pooled); /* * Destroy an RPC service. Should be called with appropriate locking to @@ -492,7 +492,7 @@ svc_destroy(struct svc_serv *serv) kfree(serv->sv_pools); kfree(serv); } -EXPORT_SYMBOL(svc_destroy); +EXPORT_SYMBOL_GPL(svc_destroy); /* * Allocate an RPC server's buffer space. @@ -567,7 +567,7 @@ out_thread: out_enomem: return ERR_PTR(-ENOMEM); } -EXPORT_SYMBOL(svc_prepare_thread); +EXPORT_SYMBOL_GPL(svc_prepare_thread); /* * Choose a pool in which to create a new thread, for svc_set_num_threads @@ -689,7 +689,7 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) return error; } -EXPORT_SYMBOL(svc_set_num_threads); +EXPORT_SYMBOL_GPL(svc_set_num_threads); /* * Called from a server thread as it's exiting. Caller must hold the BKL or @@ -717,7 +717,7 @@ svc_exit_thread(struct svc_rqst *rqstp) if (serv) svc_destroy(serv); } -EXPORT_SYMBOL(svc_exit_thread); +EXPORT_SYMBOL_GPL(svc_exit_thread); #ifdef CONFIG_SUNRPC_REGISTER_V4 @@ -1231,7 +1231,7 @@ err_bad: svc_putnl(resv, ntohl(rpc_stat)); goto sendit; } -EXPORT_SYMBOL(svc_process); +EXPORT_SYMBOL_GPL(svc_process); /* * Return (transport-specific) limit on the rpc payload. diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index a78b87937c73..e588df5d6b34 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -440,7 +440,7 @@ void svc_reserve(struct svc_rqst *rqstp, int space) svc_xprt_enqueue(xprt); } } -EXPORT_SYMBOL(svc_reserve); +EXPORT_SYMBOL_GPL(svc_reserve); static void svc_xprt_release(struct svc_rqst *rqstp) { @@ -501,7 +501,7 @@ void svc_wake_up(struct svc_serv *serv) spin_unlock_bh(&pool->sp_lock); } } -EXPORT_SYMBOL(svc_wake_up); +EXPORT_SYMBOL_GPL(svc_wake_up); int svc_port_is_privileged(struct sockaddr *sin) { @@ -743,7 +743,7 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) serv->sv_stats->netcnt++; return len; } -EXPORT_SYMBOL(svc_recv); +EXPORT_SYMBOL_GPL(svc_recv); /* * Drop request @@ -753,7 +753,7 @@ void svc_drop(struct svc_rqst *rqstp) dprintk("svc: xprt %p dropped request\n", rqstp->rq_xprt); svc_xprt_release(rqstp); } -EXPORT_SYMBOL(svc_drop); +EXPORT_SYMBOL_GPL(svc_drop); /* * Return reply to client. diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index 8a73cbb16052..e64109b02aee 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -57,13 +57,13 @@ svc_authenticate(struct svc_rqst *rqstp, __be32 *authp) rqstp->rq_authop = aops; return aops->accept(rqstp, authp); } -EXPORT_SYMBOL(svc_authenticate); +EXPORT_SYMBOL_GPL(svc_authenticate); int svc_set_client(struct svc_rqst *rqstp) { return rqstp->rq_authop->set_client(rqstp); } -EXPORT_SYMBOL(svc_set_client); +EXPORT_SYMBOL_GPL(svc_set_client); /* A request, which was authenticated, has now executed. * Time to finalise the credentials and verifier @@ -95,7 +95,7 @@ svc_auth_register(rpc_authflavor_t flavor, struct auth_ops *aops) spin_unlock(&authtab_lock); return rv; } -EXPORT_SYMBOL(svc_auth_register); +EXPORT_SYMBOL_GPL(svc_auth_register); void svc_auth_unregister(rpc_authflavor_t flavor) @@ -105,7 +105,7 @@ svc_auth_unregister(rpc_authflavor_t flavor) authtab[flavor] = NULL; spin_unlock(&authtab_lock); } -EXPORT_SYMBOL(svc_auth_unregister); +EXPORT_SYMBOL_GPL(svc_auth_unregister); /************************************************** * 'auth_domains' are stored in a hash table indexed by name. @@ -132,7 +132,7 @@ void auth_domain_put(struct auth_domain *dom) spin_unlock(&auth_domain_lock); } } -EXPORT_SYMBOL(auth_domain_put); +EXPORT_SYMBOL_GPL(auth_domain_put); struct auth_domain * auth_domain_lookup(char *name, struct auth_domain *new) @@ -157,10 +157,10 @@ auth_domain_lookup(char *name, struct auth_domain *new) spin_unlock(&auth_domain_lock); return new; } -EXPORT_SYMBOL(auth_domain_lookup); +EXPORT_SYMBOL_GPL(auth_domain_lookup); struct auth_domain *auth_domain_find(char *name) { return auth_domain_lookup(name, NULL); } -EXPORT_SYMBOL(auth_domain_find); +EXPORT_SYMBOL_GPL(auth_domain_find); diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 82240e6127b2..5c865e2d299e 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -64,7 +64,7 @@ struct auth_domain *unix_domain_find(char *name) rv = auth_domain_lookup(name, &new->h); } } -EXPORT_SYMBOL(unix_domain_find); +EXPORT_SYMBOL_GPL(unix_domain_find); static void svcauth_unix_domain_release(struct auth_domain *dom) { @@ -358,7 +358,7 @@ int auth_unix_add_addr(struct in6_addr *addr, struct auth_domain *dom) else return -ENOMEM; } -EXPORT_SYMBOL(auth_unix_add_addr); +EXPORT_SYMBOL_GPL(auth_unix_add_addr); int auth_unix_forget_old(struct auth_domain *dom) { @@ -370,7 +370,7 @@ int auth_unix_forget_old(struct auth_domain *dom) udom->addr_changes++; return 0; } -EXPORT_SYMBOL(auth_unix_forget_old); +EXPORT_SYMBOL_GPL(auth_unix_forget_old); struct auth_domain *auth_unix_lookup(struct in6_addr *addr) { @@ -395,13 +395,13 @@ struct auth_domain *auth_unix_lookup(struct in6_addr *addr) cache_put(&ipm->h, &ip_map_cache); return rv; } -EXPORT_SYMBOL(auth_unix_lookup); +EXPORT_SYMBOL_GPL(auth_unix_lookup); void svcauth_unix_purge(void) { cache_purge(&ip_map_cache); } -EXPORT_SYMBOL(svcauth_unix_purge); +EXPORT_SYMBOL_GPL(svcauth_unix_purge); static inline struct ip_map * ip_map_cached_get(struct svc_rqst *rqstp) @@ -714,7 +714,7 @@ svcauth_unix_set_client(struct svc_rqst *rqstp) return SVC_OK; } -EXPORT_SYMBOL(svcauth_unix_set_client); +EXPORT_SYMBOL_GPL(svcauth_unix_set_client); static int svcauth_null_accept(struct svc_rqst *rqstp, __be32 *authp) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 3c103404501d..5763e6460fea 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -285,7 +285,7 @@ svc_sock_names(char *buf, struct svc_serv *serv, char *toclose) return -ENOENT; return len; } -EXPORT_SYMBOL(svc_sock_names); +EXPORT_SYMBOL_GPL(svc_sock_names); /* * Check input queue length @@ -1097,7 +1097,7 @@ void svc_sock_update_bufs(struct svc_serv *serv) } spin_unlock_bh(&serv->sv_lock); } -EXPORT_SYMBOL(svc_sock_update_bufs); +EXPORT_SYMBOL_GPL(svc_sock_update_bufs); /* * Initialize socket for RPC use and create svc_sock struct -- cgit v1.2.3 From 03b35ccb7c41ccc256631ff33e6887b7be88137b Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 7 Jan 2009 17:21:44 -0800 Subject: appletalk: convert aarp to net_device_ops Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/appletalk/aarp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index b03ff58e9308..89f99d3beb60 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -443,13 +443,14 @@ static void aarp_send_probe_phase1(struct atalk_iface *iface) { struct ifreq atreq; struct sockaddr_at *sa = (struct sockaddr_at *)&atreq.ifr_addr; + const struct net_device_ops *ops = iface->dev->netdev_ops; sa->sat_addr.s_node = iface->address.s_node; sa->sat_addr.s_net = ntohs(iface->address.s_net); /* We pass the Net:Node to the drivers/cards by a Device ioctl. */ - if (!(iface->dev->do_ioctl(iface->dev, &atreq, SIOCSIFADDR))) { - (void)iface->dev->do_ioctl(iface->dev, &atreq, SIOCGIFADDR); + if (!(ops->ndo_do_ioctl(iface->dev, &atreq, SIOCSIFADDR))) { + ops->ndo_do_ioctl(iface->dev, &atreq, SIOCGIFADDR); if (iface->address.s_net != htons(sa->sat_addr.s_net) || iface->address.s_node != sa->sat_addr.s_node) iface->status |= ATIF_PROBE_FAIL; -- cgit v1.2.3 From b4d7f0a46bc0e30514b1779caff0fce6e424c4b5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 7 Jan 2009 17:23:17 -0800 Subject: bluetooth: driver API update Convert to net_device_ops and use internal net_device_stats in bnep device. Note: no need for bnep_net_ioctl since if ioctl is not set, then dev_ifsioc handles it by returning -EOPNOTSUPP Signed-off-by: Stephen Hemminger Acked-by: Marcel Holtmann Signed-off-by: David S. Miller --- net/bluetooth/bnep/bnep.h | 1 - net/bluetooth/bnep/core.c | 12 ++++++------ net/bluetooth/bnep/netdev.c | 33 +++++++++++++-------------------- 3 files changed, 19 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/net/bluetooth/bnep/bnep.h b/net/bluetooth/bnep/bnep.h index d20f8a40f36e..0d9e506f5d5a 100644 --- a/net/bluetooth/bnep/bnep.h +++ b/net/bluetooth/bnep/bnep.h @@ -165,7 +165,6 @@ struct bnep_session { struct socket *sock; struct net_device *dev; - struct net_device_stats stats; }; void bnep_net_setup(struct net_device *dev); diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index 70fea8bdb4e5..52a6ce0d772b 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -306,7 +306,7 @@ static inline int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) struct sk_buff *nskb; u8 type; - s->stats.rx_bytes += skb->len; + dev->stats.rx_bytes += skb->len; type = *(u8 *) skb->data; skb_pull(skb, 1); @@ -343,7 +343,7 @@ static inline int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) * may not be modified and because of the alignment requirements. */ nskb = alloc_skb(2 + ETH_HLEN + skb->len, GFP_KERNEL); if (!nskb) { - s->stats.rx_dropped++; + dev->stats.rx_dropped++; kfree_skb(skb); return -ENOMEM; } @@ -378,14 +378,14 @@ static inline int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) skb_copy_from_linear_data(skb, __skb_put(nskb, skb->len), skb->len); kfree_skb(skb); - s->stats.rx_packets++; + dev->stats.rx_packets++; nskb->ip_summed = CHECKSUM_NONE; nskb->protocol = eth_type_trans(nskb, dev); netif_rx_ni(nskb); return 0; badframe: - s->stats.rx_errors++; + dev->stats.rx_errors++; kfree_skb(skb); return 0; } @@ -448,8 +448,8 @@ send: kfree_skb(skb); if (len > 0) { - s->stats.tx_bytes += len; - s->stats.tx_packets++; + s->dev->stats.tx_bytes += len; + s->dev->stats.tx_packets++; return 0; } diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c index f897da6e0444..d7a0e9722def 100644 --- a/net/bluetooth/bnep/netdev.c +++ b/net/bluetooth/bnep/netdev.c @@ -55,12 +55,6 @@ static int bnep_net_close(struct net_device *dev) return 0; } -static struct net_device_stats *bnep_net_get_stats(struct net_device *dev) -{ - struct bnep_session *s = netdev_priv(dev); - return &s->stats; -} - static void bnep_net_set_mc_list(struct net_device *dev) { #ifdef CONFIG_BT_BNEP_MC_FILTER @@ -128,11 +122,6 @@ static void bnep_net_timeout(struct net_device *dev) netif_wake_queue(dev); } -static int bnep_net_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - return -EINVAL; -} - #ifdef CONFIG_BT_BNEP_MC_FILTER static inline int bnep_net_mc_filter(struct sk_buff *skb, struct bnep_session *s) { @@ -217,6 +206,18 @@ static int bnep_net_xmit(struct sk_buff *skb, struct net_device *dev) return 0; } +static const struct net_device_ops bnep_netdev_ops = { + .ndo_open = bnep_net_open, + .ndo_stop = bnep_net_close, + .ndo_start_xmit = bnep_net_xmit, + .ndo_validate_addr = eth_validate_addr, + .ndo_set_multicast_list = bnep_net_set_mc_list, + .ndo_set_mac_address = bnep_net_set_mac_addr, + .ndo_tx_timeout = bnep_net_timeout, + .ndo_change_mtu = eth_change_mtu, + +}; + void bnep_net_setup(struct net_device *dev) { @@ -224,15 +225,7 @@ void bnep_net_setup(struct net_device *dev) dev->addr_len = ETH_ALEN; ether_setup(dev); - - dev->open = bnep_net_open; - dev->stop = bnep_net_close; - dev->hard_start_xmit = bnep_net_xmit; - dev->get_stats = bnep_net_get_stats; - dev->do_ioctl = bnep_net_ioctl; - dev->set_mac_address = bnep_net_set_mac_addr; - dev->set_multicast_list = bnep_net_set_mc_list; + dev->netdev_ops = &bnep_netdev_ops; dev->watchdog_timeo = HZ * 2; - dev->tx_timeout = bnep_net_timeout; } -- cgit v1.2.3 From ab638e69ff7d4882ac152ada17eab340c93df080 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 7 Jan 2009 17:24:34 -0800 Subject: phonet: update to net_device_ops Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- drivers/usb/gadget/f_phonet.c | 12 ++++++++---- net/phonet/pep-gprs.c | 12 ++++++++---- 2 files changed, 16 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/drivers/usb/gadget/f_phonet.c b/drivers/usb/gadget/f_phonet.c index d8fc9b32fe36..c0916c7b217e 100644 --- a/drivers/usb/gadget/f_phonet.c +++ b/drivers/usb/gadget/f_phonet.c @@ -279,6 +279,13 @@ static int pn_net_mtu(struct net_device *dev, int new_mtu) return err; } +static const struct net_device_ops pn_netdev_ops = { + .ndo_open = pn_net_open, + .ndo_stop = pn_net_close, + .ndo_start_xmit = pn_net_xmit, + .ndo_change_mtu = pn_net_mtu, +}; + static void pn_net_setup(struct net_device *dev) { dev->features = 0; @@ -290,12 +297,9 @@ static void pn_net_setup(struct net_device *dev) dev->addr_len = 1; dev->tx_queue_len = 1; + dev->netdev_ops = &pn_netdev_ops; dev->destructor = free_netdev; dev->header_ops = &phonet_header_ops; - dev->open = pn_net_open; - dev->stop = pn_net_close; - dev->hard_start_xmit = pn_net_xmit; /* mandatory */ - dev->change_mtu = pn_net_mtu; } /*-------------------------------------------------------------------------*/ diff --git a/net/phonet/pep-gprs.c b/net/phonet/pep-gprs.c index b0ceac2d6cd1..6a91a32a80c1 100644 --- a/net/phonet/pep-gprs.c +++ b/net/phonet/pep-gprs.c @@ -227,6 +227,13 @@ static int gprs_set_mtu(struct net_device *dev, int new_mtu) return 0; } +static const struct net_device_ops gprs_netdev_ops = { + .ndo_open = gprs_open, + .ndo_stop = gprs_close, + .ndo_start_xmit = gprs_xmit, + .ndo_change_mtu = gprs_set_mtu, +}; + static void gprs_setup(struct net_device *dev) { dev->features = NETIF_F_FRAGLIST; @@ -237,11 +244,8 @@ static void gprs_setup(struct net_device *dev) dev->addr_len = 0; dev->tx_queue_len = 10; + dev->netdev_ops = &gprs_netdev_ops; dev->destructor = free_netdev; - dev->open = gprs_open; - dev->stop = gprs_close; - dev->hard_start_xmit = gprs_xmit; /* mandatory */ - dev->change_mtu = gprs_set_mtu; } /* -- cgit v1.2.3 From c19a28e1193a6c854738d609ae9b2fe2f6e6bea4 Mon Sep 17 00:00:00 2001 From: Fernando Carrijo Date: Wed, 7 Jan 2009 18:09:08 -0800 Subject: remove lots of double-semicolons Cc: Ingo Molnar Cc: Thomas Gleixner Acked-by: Theodore Ts'o Acked-by: Mark Fasheh Acked-by: David S. Miller Cc: James Morris Acked-by: Casey Schaufler Acked-by: Takashi Iwai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/kernel/cpu/cpufreq/longhaul.c | 2 +- fs/ocfs2/alloc.c | 2 +- fs/ocfs2/file.c | 2 +- net/ipv6/route.c | 2 +- net/ipv6/sysctl_net_ipv6.c | 2 +- net/sched/sch_sfq.c | 2 +- security/smack/smackfs.c | 2 +- sound/soc/au1x/dbdma2.c | 2 +- sound/soc/davinci/davinci-pcm.c | 2 +- 9 files changed, 9 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.c b/arch/x86/kernel/cpu/cpufreq/longhaul.c index b0461856acfb..a4cff5d6e380 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.c @@ -982,7 +982,7 @@ static int __init longhaul_init(void) case 10: printk(KERN_ERR PFX "Use acpi-cpufreq driver for VIA C7\n"); default: - ;; + ; } return -ENODEV; diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 54ff4c77aaa3..d861096c9d81 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -3868,7 +3868,7 @@ static void ocfs2_split_record(struct inode *inode, struct ocfs2_extent_list *left_el = NULL, *right_el, *insert_el, *el; struct ocfs2_extent_rec *rec, *tmprec; - right_el = path_leaf_el(right_path);; + right_el = path_leaf_el(right_path); if (left_path) left_el = path_leaf_el(left_path); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index e8f795f978aa..a5887df2cd8a 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1605,7 +1605,7 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, struct ocfs2_space_resv *sr) { struct inode *inode = file->f_path.dentry->d_inode; - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if ((cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) && !ocfs2_writes_unwritten_extents(osb)) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 76f06b94ab9f..c4a59824ac2c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2752,7 +2752,7 @@ int __init ip6_route_init(void) kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0, SLAB_HWCACHE_ALIGN, NULL); if (!ip6_dst_ops_template.kmem_cachep) - goto out;; + goto out; ret = register_pernet_subsys(&ip6_route_net_ops); if (ret) diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 9048fe7e7ea7..a031034720b4 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -128,7 +128,7 @@ static struct ctl_table_header *ip6_header; int ipv6_sysctl_register(void) { - int err = -ENOMEM;; + int err = -ENOMEM; ip6_header = register_net_sysctl_rotable(net_ipv6_ctl_path, ipv6_table); if (ip6_header == NULL) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index f3965df00559..33133d27b539 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -435,7 +435,7 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt) int i; q->perturb_timer.function = sfq_perturbation; - q->perturb_timer.data = (unsigned long)sch;; + q->perturb_timer.data = (unsigned long)sch; init_timer_deferrable(&q->perturb_timer); for (i = 0; i < SFQ_HASH_DIVISOR; i++) diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index bf107a389ac1..71e2b914363e 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -569,7 +569,7 @@ static ssize_t smk_write_cipso(struct file *file, const char __user *buf, if (skp == NULL) goto out; - rule += SMK_LABELLEN;; + rule += SMK_LABELLEN; ret = sscanf(rule, "%d", &maplevel); if (ret != 1 || maplevel > SMACK_CIPSO_MAXLEVEL) goto out; diff --git a/sound/soc/au1x/dbdma2.c b/sound/soc/au1x/dbdma2.c index 74c823d60f91..bc8d654576c0 100644 --- a/sound/soc/au1x/dbdma2.c +++ b/sound/soc/au1x/dbdma2.c @@ -187,7 +187,7 @@ static int au1x_pcm_dbdma_realloc(struct au1xpsc_audio_dmadata *pcd, au1x_pcm_dmatx_cb, (void *)pcd); if (!pcd->ddma_chan) - return -ENOMEM;; + return -ENOMEM; au1xxx_dbdma_set_devwidth(pcd->ddma_chan, msbits); au1xxx_dbdma_ring_alloc(pcd->ddma_chan, 2); diff --git a/sound/soc/davinci/davinci-pcm.c b/sound/soc/davinci/davinci-pcm.c index 74abc9b4f1cc..366049d8578c 100644 --- a/sound/soc/davinci/davinci-pcm.c +++ b/sound/soc/davinci/davinci-pcm.c @@ -212,7 +212,7 @@ davinci_pcm_pointer(struct snd_pcm_substream *substream) if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) count = src - runtime->dma_addr; else - count = dst - runtime->dma_addr;; + count = dst - runtime->dma_addr; spin_unlock(&prtd->lock); -- cgit v1.2.3 From 73ac36ea14fd18ea3dc057e41b16ff31a3c0bd5a Mon Sep 17 00:00:00 2001 From: Coly Li Date: Wed, 7 Jan 2009 18:09:16 -0800 Subject: fix similar typos to successfull When I review ocfs2 code, find there are 2 typos to "successfull". After doing grep "successfull " in kernel tree, 22 typos found totally -- great minds always think alike :) This patch fixes all the similar typos. Thanks for Randy's ack and comments. Signed-off-by: Coly Li Acked-by: Randy Dunlap Acked-by: Roland Dreier Cc: Jeremy Kerr Cc: Jeff Garzik Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: Theodore Ts'o Cc: Mark Fasheh Cc: Vlad Yasevich Cc: Sridhar Samudrala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/hwmon/abituguru-datasheet | 4 ++-- Documentation/scsi/scsi_fc_transport.txt | 4 ++-- arch/powerpc/platforms/cell/spufs/spufs.h | 2 +- drivers/infiniband/hw/nes/nes_cm.c | 2 +- drivers/isdn/hardware/eicon/debuglib.h | 2 +- drivers/isdn/hardware/eicon/os_4bri.c | 2 +- drivers/isdn/hardware/eicon/os_bri.c | 2 +- drivers/isdn/hardware/eicon/os_pri.c | 2 +- drivers/mtd/ubi/kapi.c | 2 +- drivers/net/wireless/ath5k/dma.c | 2 +- drivers/net/wireless/zd1211rw/zd_mac.c | 2 +- drivers/s390/block/dasd_3990_erp.c | 2 +- drivers/s390/block/dasd_int.h | 2 +- drivers/s390/char/tape_3590.c | 2 +- drivers/s390/cio/cio.c | 2 +- drivers/s390/cio/qdio_main.c | 2 +- fs/ext4/extents.c | 2 +- fs/ocfs2/dlmglue.c | 4 ++-- net/sctp/auth.c | 2 +- 19 files changed, 22 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/Documentation/hwmon/abituguru-datasheet b/Documentation/hwmon/abituguru-datasheet index 4d184f2db0ea..d9251efdcec7 100644 --- a/Documentation/hwmon/abituguru-datasheet +++ b/Documentation/hwmon/abituguru-datasheet @@ -121,7 +121,7 @@ Once all bytes have been read data will hold 0x09, but there is no reason to test for this. Notice that the number of bytes is bank address dependent see above and below. -After completing a successfull read it is advised to put the uGuru back in +After completing a successful read it is advised to put the uGuru back in ready mode, so that it is ready for the next read / write cycle. This way if your program / driver is unloaded and later loaded again the detection algorithm described above will still work. @@ -141,7 +141,7 @@ don't ask why this is the way it is. Once DATA holds 0x01 read CMD it should hold 0xAC now. -After completing a successfull write it is advised to put the uGuru back in +After completing a successful write it is advised to put the uGuru back in ready mode, so that it is ready for the next read / write cycle. This way if your program / driver is unloaded and later loaded again the detection algorithm described above will still work. diff --git a/Documentation/scsi/scsi_fc_transport.txt b/Documentation/scsi/scsi_fc_transport.txt index 38d324d62b25..e5b071d46619 100644 --- a/Documentation/scsi/scsi_fc_transport.txt +++ b/Documentation/scsi/scsi_fc_transport.txt @@ -191,7 +191,7 @@ Vport States: This is equivalent to a driver "attach" on an adapter, which is independent of the adapter's link state. - Instantiation of the vport on the FC link via ELS traffic, etc. - This is equivalent to a "link up" and successfull link initialization. + This is equivalent to a "link up" and successful link initialization. Further information can be found in the interfaces section below for Vport Creation. @@ -320,7 +320,7 @@ Vport Creation: This is equivalent to a driver "attach" on an adapter, which is independent of the adapter's link state. - Instantiation of the vport on the FC link via ELS traffic, etc. - This is equivalent to a "link up" and successfull link initialization. + This is equivalent to a "link up" and successful link initialization. The LLDD's vport_create() function will not synchronously wait for both parts to be fully completed before returning. It must validate that the diff --git a/arch/powerpc/platforms/cell/spufs/spufs.h b/arch/powerpc/platforms/cell/spufs/spufs.h index 15c62d3ca129..3bf908e2873a 100644 --- a/arch/powerpc/platforms/cell/spufs/spufs.h +++ b/arch/powerpc/platforms/cell/spufs/spufs.h @@ -314,7 +314,7 @@ extern char *isolated_loader; * we need to call spu_release(ctx) before sleeping, and * then spu_acquire(ctx) when awoken. * - * Returns with state_mutex re-acquired when successfull or + * Returns with state_mutex re-acquired when successful or * with -ERESTARTSYS and the state_mutex dropped when interrupted. */ diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index a812db243477..6ba57e91d7ab 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -2705,7 +2705,7 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) sizeof(struct ietf_mpa_frame)); - /* notify OF layer that accept event was successfull */ + /* notify OF layer that accept event was successful */ cm_id->add_ref(cm_id); cm_event.event = IW_CM_EVENT_ESTABLISHED; diff --git a/drivers/isdn/hardware/eicon/debuglib.h b/drivers/isdn/hardware/eicon/debuglib.h index 016410cf2273..8ea587783e14 100644 --- a/drivers/isdn/hardware/eicon/debuglib.h +++ b/drivers/isdn/hardware/eicon/debuglib.h @@ -235,7 +235,7 @@ typedef void ( * DbgOld) (unsigned short, char *, va_list) ; typedef void ( * DbgEv) (unsigned short, unsigned long, va_list) ; typedef void ( * DbgIrq) (unsigned short, int, char *, va_list) ; typedef struct _DbgHandle_ -{ char Registered ; /* driver successfull registered */ +{ char Registered ; /* driver successfully registered */ #define DBG_HANDLE_REG_NEW 0x01 /* this (new) structure */ #define DBG_HANDLE_REG_OLD 0x7f /* old structure (see below) */ char Version; /* version of this structure */ diff --git a/drivers/isdn/hardware/eicon/os_4bri.c b/drivers/isdn/hardware/eicon/os_4bri.c index 7b4ec3f60dbf..c964b8d91ada 100644 --- a/drivers/isdn/hardware/eicon/os_4bri.c +++ b/drivers/isdn/hardware/eicon/os_4bri.c @@ -997,7 +997,7 @@ diva_4bri_start_adapter(PISDN_ADAPTER IoAdapter, diva_xdi_display_adapter_features(IoAdapter->ANum); for (i = 0; i < IoAdapter->tasks; i++) { - DBG_LOG(("A(%d) %s adapter successfull started", + DBG_LOG(("A(%d) %s adapter successfully started", IoAdapter->QuadroList->QuadroAdapter[i]->ANum, (IoAdapter->tasks == 1) ? "BRI 2.0" : "4BRI")) diva_xdi_didd_register_adapter(IoAdapter->QuadroList->QuadroAdapter[i]->ANum); diff --git a/drivers/isdn/hardware/eicon/os_bri.c b/drivers/isdn/hardware/eicon/os_bri.c index f31bba5b16ff..08f01993f46b 100644 --- a/drivers/isdn/hardware/eicon/os_bri.c +++ b/drivers/isdn/hardware/eicon/os_bri.c @@ -736,7 +736,7 @@ diva_bri_start_adapter(PISDN_ADAPTER IoAdapter, IoAdapter->Properties.Features = (word) features; diva_xdi_display_adapter_features(IoAdapter->ANum); - DBG_LOG(("A(%d) BRI adapter successfull started", IoAdapter->ANum)) + DBG_LOG(("A(%d) BRI adapter successfully started", IoAdapter->ANum)) /* Register with DIDD */ diff --git a/drivers/isdn/hardware/eicon/os_pri.c b/drivers/isdn/hardware/eicon/os_pri.c index 903356547b79..5d65405c75f4 100644 --- a/drivers/isdn/hardware/eicon/os_pri.c +++ b/drivers/isdn/hardware/eicon/os_pri.c @@ -513,7 +513,7 @@ diva_pri_start_adapter(PISDN_ADAPTER IoAdapter, diva_xdi_display_adapter_features(IoAdapter->ANum); - DBG_LOG(("A(%d) PRI adapter successfull started", IoAdapter->ANum)) + DBG_LOG(("A(%d) PRI adapter successfully started", IoAdapter->ANum)) /* Register with DIDD */ diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c index 5d9bcf109c13..4abbe573fa40 100644 --- a/drivers/mtd/ubi/kapi.c +++ b/drivers/mtd/ubi/kapi.c @@ -564,7 +564,7 @@ EXPORT_SYMBOL_GPL(ubi_leb_unmap); * @dtype: expected data type * * This function maps an un-mapped logical eraseblock @lnum to a physical - * eraseblock. This means, that after a successfull invocation of this + * eraseblock. This means, that after a successful invocation of this * function the logical eraseblock @lnum will be empty (contain only %0xFF * bytes) and be mapped to a physical eraseblock, even if an unclean reboot * happens. diff --git a/drivers/net/wireless/ath5k/dma.c b/drivers/net/wireless/ath5k/dma.c index 7e2b1a67e5da..b65b4feb2d28 100644 --- a/drivers/net/wireless/ath5k/dma.c +++ b/drivers/net/wireless/ath5k/dma.c @@ -594,7 +594,7 @@ int ath5k_hw_get_isr(struct ath5k_hw *ah, enum ath5k_int *interrupt_mask) * XXX: BMISS interrupts may occur after association. * I found this on 5210 code but it needs testing. If this is * true we should disable them before assoc and re-enable them - * after a successfull assoc + some jiffies. + * after a successful assoc + some jiffies. interrupt_mask &= ~AR5K_INT_BMISS; */ } diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 9caa96a13586..a611ad857983 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -287,7 +287,7 @@ static void zd_op_stop(struct ieee80211_hw *hw) * @skb - a sk-buffer * @flags: extra flags to set in the TX status info * @ackssi: ACK signal strength - * @success - True for successfull transmission of the frame + * @success - True for successful transmission of the frame * * This information calls ieee80211_tx_status_irqsafe() if required by the * control information. It copies the control information into the status diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c index b8f9c00633f3..d82aad5224f0 100644 --- a/drivers/s390/block/dasd_3990_erp.c +++ b/drivers/s390/block/dasd_3990_erp.c @@ -2621,7 +2621,7 @@ dasd_3990_erp_action(struct dasd_ccw_req * cqr) } } - /* double-check if current erp/cqr was successfull */ + /* double-check if current erp/cqr was successful */ if ((cqr->irb.scsw.cmd.cstat == 0x00) && (cqr->irb.scsw.cmd.dstat == (DEV_STAT_CHN_END | DEV_STAT_DEV_END))) { diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 05a14536c369..4a39084d9c95 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -199,7 +199,7 @@ struct dasd_ccw_req { #define DASD_CQR_ERROR 0x82 /* request is completed with error */ #define DASD_CQR_CLEAR_PENDING 0x83 /* request is clear pending */ #define DASD_CQR_CLEARED 0x84 /* request was cleared */ -#define DASD_CQR_SUCCESS 0x85 /* request was successfull */ +#define DASD_CQR_SUCCESS 0x85 /* request was successful */ /* per dasd_ccw_req flags */ diff --git a/drivers/s390/char/tape_3590.c b/drivers/s390/char/tape_3590.c index 4005c44a404c..71605a179d65 100644 --- a/drivers/s390/char/tape_3590.c +++ b/drivers/s390/char/tape_3590.c @@ -801,7 +801,7 @@ tape_3590_done(struct tape_device *device, struct tape_request *request) static inline int tape_3590_erp_succeded(struct tape_device *device, struct tape_request *request) { - DBF_EVENT(3, "Error Recovery successfull for %s\n", + DBF_EVENT(3, "Error Recovery successful for %s\n", tape_op_verbose[request->op]); return tape_3590_done(device, request); } diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 06b71823f399..659f8a791656 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -379,7 +379,7 @@ int cio_commit_config(struct subchannel *sch) if (ccode < 0) /* -EIO if msch gets a program check. */ return ccode; switch (ccode) { - case 0: /* successfull */ + case 0: /* successful */ if (stsch(sch->schid, &schib) || !css_sch_is_valid(&schib)) return -ENODEV; diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c index 744f928a59ea..10cb0f8726e5 100644 --- a/drivers/s390/cio/qdio_main.c +++ b/drivers/s390/cio/qdio_main.c @@ -114,7 +114,7 @@ static inline int qdio_check_ccq(struct qdio_q *q, unsigned int ccq) * @count: count of buffers to examine * @auto_ack: automatically acknowledge buffers * - * Returns the number of successfull extracted equal buffer states. + * Returns the number of successfully extracted equal buffer states. * Stops processing if a state is different from the last buffers state. */ static int qdio_do_eqbs(struct qdio_q *q, unsigned char *state, diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index ea2ce3c0ae66..3f54db31cdc2 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2536,7 +2536,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, */ newdepth = ext_depth(inode); /* - * update the extent length after successfull insert of the + * update the extent length after successful insert of the * split extent */ orig_ex.ee_len = cpu_to_le16(ee_len - diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index f731ab491795..b0c4cadd4c45 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -1324,7 +1324,7 @@ again: goto out; } - mlog(0, "lock %s, successfull return from ocfs2_dlm_lock\n", + mlog(0, "lock %s, successful return from ocfs2_dlm_lock\n", lockres->l_name); /* At this point we've gone inside the dlm and need to @@ -2951,7 +2951,7 @@ static int ocfs2_drop_lock(struct ocfs2_super *osb, ocfs2_dlm_dump_lksb(&lockres->l_lksb); BUG(); } - mlog(0, "lock %s, successfull return from ocfs2_dlm_unlock\n", + mlog(0, "lock %s, successful return from ocfs2_dlm_unlock\n", lockres->l_name); ocfs2_wait_on_busy_lock(lockres); diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 20c576f530fa..56935bbc1496 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -489,7 +489,7 @@ int sctp_auth_init_hmacs(struct sctp_endpoint *ep, gfp_t gfp) return 0; out_err: - /* Clean up any successfull allocations */ + /* Clean up any successful allocations */ sctp_auth_destroy_hmacs(ep->auth_hmacs); return -ENOMEM; } -- cgit v1.2.3 From 787e9208360117835101f513f7db593dc2525cf8 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 8 Jan 2009 10:40:57 -0800 Subject: ipv6: Add GRO support This patch adds GRO support for IPv6. IPv6 GRO supports extension headers in the same way as GSO (by using the same infrastructure). It's also simpler compared to IPv4 since we no longer have to worry about fragmentation attributes or header checksums. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/net/protocol.h | 3 ++ net/ipv6/af_inet6.c | 107 ++++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 105 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/net/protocol.h b/include/net/protocol.h index cb2965aa1b62..ffa5b8b1f1df 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -59,6 +59,9 @@ struct inet6_protocol int (*gso_send_check)(struct sk_buff *skb); struct sk_buff *(*gso_segment)(struct sk_buff *skb, int features); + struct sk_buff **(*gro_receive)(struct sk_buff **head, + struct sk_buff *skb); + int (*gro_complete)(struct sk_buff *skb); unsigned int flags; /* INET6_PROTO_xxx */ }; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 437b750b98fd..94f74f5b0cbf 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -672,8 +672,7 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb) EXPORT_SYMBOL_GPL(ipv6_opt_accepted); -static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb, - int proto) +static int ipv6_gso_pull_exthdrs(struct sk_buff *skb, int proto) { struct inet6_protocol *ops = NULL; @@ -704,7 +703,7 @@ static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb, __skb_pull(skb, len); } - return ops; + return proto; } static int ipv6_gso_send_check(struct sk_buff *skb) @@ -721,7 +720,9 @@ static int ipv6_gso_send_check(struct sk_buff *skb) err = -EPROTONOSUPPORT; rcu_read_lock(); - ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); + ops = rcu_dereference(inet6_protos[ + ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr)]); + if (likely(ops && ops->gso_send_check)) { skb_reset_transport_header(skb); err = ops->gso_send_check(skb); @@ -757,7 +758,9 @@ static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features) segs = ERR_PTR(-EPROTONOSUPPORT); rcu_read_lock(); - ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); + ops = rcu_dereference(inet6_protos[ + ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr)]); + if (likely(ops && ops->gso_segment)) { skb_reset_transport_header(skb); segs = ops->gso_segment(skb, features); @@ -777,11 +780,105 @@ out: return segs; } +struct ipv6_gro_cb { + struct napi_gro_cb napi; + int proto; +}; + +#define IPV6_GRO_CB(skb) ((struct ipv6_gro_cb *)(skb)->cb) + +static struct sk_buff **ipv6_gro_receive(struct sk_buff **head, + struct sk_buff *skb) +{ + struct inet6_protocol *ops; + struct sk_buff **pp = NULL; + struct sk_buff *p; + struct ipv6hdr *iph; + unsigned int nlen; + int flush = 1; + int proto; + + if (unlikely(!pskb_may_pull(skb, sizeof(*iph)))) + goto out; + + iph = ipv6_hdr(skb); + __skb_pull(skb, sizeof(*iph)); + + flush += ntohs(iph->payload_len) != skb->len; + + rcu_read_lock(); + proto = ipv6_gso_pull_exthdrs(skb, iph->nexthdr); + IPV6_GRO_CB(skb)->proto = proto; + ops = rcu_dereference(inet6_protos[proto]); + if (!ops || !ops->gro_receive) + goto out_unlock; + + flush--; + skb_reset_transport_header(skb); + nlen = skb_network_header_len(skb); + + for (p = *head; p; p = p->next) { + struct ipv6hdr *iph2; + + if (!NAPI_GRO_CB(p)->same_flow) + continue; + + iph2 = ipv6_hdr(p); + + /* All fields must match except length. */ + if (nlen != skb_network_header_len(p) || + memcmp(iph, iph2, offsetof(struct ipv6hdr, payload_len)) || + memcmp(&iph->nexthdr, &iph2->nexthdr, + nlen - offsetof(struct ipv6hdr, nexthdr))) { + NAPI_GRO_CB(p)->same_flow = 0; + continue; + } + + NAPI_GRO_CB(p)->flush |= flush; + } + + NAPI_GRO_CB(skb)->flush |= flush; + + pp = ops->gro_receive(head, skb); + +out_unlock: + rcu_read_unlock(); + +out: + NAPI_GRO_CB(skb)->flush |= flush; + + return pp; +} + +static int ipv6_gro_complete(struct sk_buff *skb) +{ + struct inet6_protocol *ops; + struct ipv6hdr *iph = ipv6_hdr(skb); + int err = -ENOSYS; + + iph->payload_len = htons(skb->len - skb_network_offset(skb) - + sizeof(*iph)); + + rcu_read_lock(); + ops = rcu_dereference(inet6_protos[IPV6_GRO_CB(skb)->proto]); + if (WARN_ON(!ops || !ops->gro_complete)) + goto out_unlock; + + err = ops->gro_complete(skb); + +out_unlock: + rcu_read_unlock(); + + return err; +} + static struct packet_type ipv6_packet_type = { .type = __constant_htons(ETH_P_IPV6), .func = ipv6_rcv, .gso_send_check = ipv6_gso_send_check, .gso_segment = ipv6_gso_segment, + .gro_receive = ipv6_gro_receive, + .gro_complete = ipv6_gro_complete, }; static int __init ipv6_packet_init(void) -- cgit v1.2.3 From 684f2176015b313ab59cecf574117969cf638f28 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 8 Jan 2009 10:41:23 -0800 Subject: tcp6: Add GRO support This patch adds GRO support for TCP over IPv6. The code is exactly the same as the IPv4 version except for the pseudo-header checksum computation. Note that I've removed the unused tcphdr argument from tcp_v6_check rather than invent a bogus value for GRO. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 2 ++ net/ipv6/tcp_ipv6.c | 45 +++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 43 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 35bcddf8a932..bd6ff907d9e4 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2542,6 +2542,7 @@ out: return pp; } +EXPORT_SYMBOL(tcp_gro_receive); int tcp_gro_complete(struct sk_buff *skb) { @@ -2558,6 +2559,7 @@ int tcp_gro_complete(struct sk_buff *skb) return 0; } +EXPORT_SYMBOL(tcp_gro_complete); #ifdef CONFIG_TCP_MD5SIG static unsigned long tcp_md5sig_users; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e8b8337a8310..1297306d729c 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -101,7 +101,7 @@ static void tcp_v6_hash(struct sock *sk) } } -static __inline__ __sum16 tcp_v6_check(struct tcphdr *th, int len, +static __inline__ __sum16 tcp_v6_check(int len, struct in6_addr *saddr, struct in6_addr *daddr, __wsum base) @@ -501,7 +501,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req) if (skb) { struct tcphdr *th = tcp_hdr(skb); - th->check = tcp_v6_check(th, skb->len, + th->check = tcp_v6_check(skb->len, &treq->loc_addr, &treq->rmt_addr, csum_partial(th, skb->len, skb->csum)); @@ -942,6 +942,41 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb) return 0; } +struct sk_buff **tcp6_gro_receive(struct sk_buff **head, struct sk_buff *skb) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + + switch (skb->ip_summed) { + case CHECKSUM_COMPLETE: + if (!tcp_v6_check(skb->len, &iph->saddr, &iph->daddr, + skb->csum)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + break; + } + + /* fall through */ + case CHECKSUM_NONE: + NAPI_GRO_CB(skb)->flush = 1; + return NULL; + } + + return tcp_gro_receive(head, skb); +} +EXPORT_SYMBOL(tcp6_gro_receive); + +int tcp6_gro_complete(struct sk_buff *skb) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + struct tcphdr *th = tcp_hdr(skb); + + th->check = ~tcp_v6_check(skb->len - skb_transport_offset(skb), + &iph->saddr, &iph->daddr, 0); + skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6; + + return tcp_gro_complete(skb); +} +EXPORT_SYMBOL(tcp6_gro_complete); + static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts, struct tcp_md5sig_key *key, int rst) { @@ -1429,14 +1464,14 @@ out: static __sum16 tcp_v6_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!tcp_v6_check(tcp_hdr(skb), skb->len, &ipv6_hdr(skb)->saddr, + if (!tcp_v6_check(skb->len, &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->csum)) { skb->ip_summed = CHECKSUM_UNNECESSARY; return 0; } } - skb->csum = ~csum_unfold(tcp_v6_check(tcp_hdr(skb), skb->len, + skb->csum = ~csum_unfold(tcp_v6_check(skb->len, &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, 0)); @@ -2062,6 +2097,8 @@ static struct inet6_protocol tcpv6_protocol = { .err_handler = tcp_v6_err, .gso_send_check = tcp_v6_gso_send_check, .gso_segment = tcp_tso_segment, + .gro_receive = tcp6_gro_receive, + .gro_complete = tcp6_gro_complete, .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -- cgit v1.2.3 From cc883d16c3b7434c7da2c45b54a49c2a99e83db7 Mon Sep 17 00:00:00 2001 From: Frank Blaschka Date: Thu, 8 Jan 2009 10:50:20 -0800 Subject: vlan: add neigh_setup In case the real device has a neigh_setup function, this neigh_setup function should be used for the vlan device. Signed-off-by: Frank Blaschka Signed-off-by: David S. Miller --- net/8021q/vlan_dev.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'net') diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 89a3bbdfca3f..4a19acd3a32b 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -546,6 +546,18 @@ static int vlan_dev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return err; } +static int vlan_dev_neigh_setup(struct net_device *dev, struct neigh_parms *pa) +{ + struct net_device *real_dev = vlan_dev_info(dev)->real_dev; + const struct net_device_ops *ops = real_dev->netdev_ops; + int err = 0; + + if (netif_device_present(real_dev) && ops->ndo_neigh_setup) + err = ops->ndo_neigh_setup(dev, pa); + + return err; +} + static void vlan_dev_change_rx_flags(struct net_device *dev, int change) { struct net_device *real_dev = vlan_dev_info(dev)->real_dev; @@ -713,6 +725,7 @@ static const struct net_device_ops vlan_netdev_ops = { .ndo_set_multicast_list = vlan_dev_set_rx_mode, .ndo_change_rx_flags = vlan_dev_change_rx_flags, .ndo_do_ioctl = vlan_dev_ioctl, + .ndo_neigh_setup = vlan_dev_neigh_setup, }; static const struct net_device_ops vlan_netdev_accel_ops = { @@ -728,6 +741,7 @@ static const struct net_device_ops vlan_netdev_accel_ops = { .ndo_set_multicast_list = vlan_dev_set_rx_mode, .ndo_change_rx_flags = vlan_dev_change_rx_flags, .ndo_do_ioctl = vlan_dev_ioctl, + .ndo_neigh_setup = vlan_dev_neigh_setup, }; void vlan_setup(struct net_device *dev) -- cgit v1.2.3 From d48e470f76887d0befe025049158aeb6c1219d71 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Thu, 8 Jan 2009 11:06:48 -0800 Subject: wimax: fix '#ifndef CONFIG_BUG' layout to avoid warning Reported by Randy Dunlap: > Also, this warning needs to be fixed: > > linux-next-20090106/net/wimax/id-table.c:133: warning: ISO C90 > forbids mixed declarations and code Move the return on #defined(CONFIG_BUG) below the variable declarations so it doesn't violate ISO C90. On wimax_id_table_release() we want to do a debug check if CONFIG_BUG is enabled. However, we also want the debug code to be always compiled to ensure there is no bitrot. It will be optimized out by the compiler when CONFIG_BUG is disabled. Added a note to the function header stating this. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: David S. Miller --- net/wimax/id-table.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/wimax/id-table.c b/net/wimax/id-table.c index d3b88558682c..5e685f7eda90 100644 --- a/net/wimax/id-table.c +++ b/net/wimax/id-table.c @@ -123,15 +123,17 @@ void wimax_id_table_rm(struct wimax_dev *wimax_dev) /* * Release the gennetlink family id / mapping table * - * On debug, verify that the table is empty upon removal. + * On debug, verify that the table is empty upon removal. We want the + * code always compiled, to ensure it doesn't bit rot. It will be + * compiled out if CONFIG_BUG is disabled. */ void wimax_id_table_release(void) { + struct wimax_dev *wimax_dev; + #ifndef CONFIG_BUG return; #endif - struct wimax_dev *wimax_dev; - spin_lock(&wimax_id_table_lock); list_for_each_entry(wimax_dev, &wimax_id_table, id_table_node) { printk(KERN_ERR "BUG: %s wimax_dev %p ifindex %d not cleared\n", -- cgit v1.2.3 From 71f78afd67ca51d3656ba45aea293d6e2a27c8bc Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Thu, 8 Jan 2009 11:07:36 -0800 Subject: wimax: fix kconfig interactions with rfkill and input layers WiMAX can work without RFKILL, but it was missing a check to make sure RFKILL is not being made a module with wimax compiled into the kernel. This caused failed builds in s390, where CONFIG_INPUT is always off. When RFKILL is enabled, the code uses the input layer to report hardware switch changes; thus, if RFKILL is enabled, INPUT has to be too. It also needs to display some message when INPUT is disabled that explains why WiMAX is not selectable. (issues found by Randy Dunlap in the linux-next tree). Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: David S. Miller --- net/wimax/Kconfig | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'net') diff --git a/net/wimax/Kconfig b/net/wimax/Kconfig index 0bdbb6928205..18495cdcd10d 100644 --- a/net/wimax/Kconfig +++ b/net/wimax/Kconfig @@ -1,9 +1,23 @@ # # WiMAX LAN device configuration # +# Note the ugly 'depends on' on WIMAX: that disallows RFKILL to be a +# module if WIMAX is to be linked in. The WiMAX code is done in such a +# way that it doesn't require and explicit dependency on RFKILL in +# case an embedded system wants to rip it out. +# +# As well, enablement of the RFKILL code means we need the INPUT layer +# support to inject events coming from hw rfkill switches. That +# dependency could be killed if input.h provided appropiate means to +# work when input is disabled. + +comment "WiMAX Wireless Broadband support requires CONFIG_INPUT enabled" + depends on INPUT = n && RFKILL != n menuconfig WIMAX tristate "WiMAX Wireless Broadband support" + depends on (y && RFKILL != m) || m + depends on (INPUT && RFKILL != n) || RFKILL = n help Select to configure support for devices that provide -- cgit v1.2.3 From a2e9da4b09c99857080afd2e8143f70bc03ebe9b Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Thu, 8 Jan 2009 11:08:01 -0800 Subject: wimax: testing for rfkill support should also test for CONFIG_RFKILL_MODULE Current WiMAX rfkill code is missing the case where rfkill is compiled in as modules and works only when rfkill is compiled in. This is not correct. Fixed to test for CONFIG_RFKILL or CONFIG_RKILL_MODULE. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: David S. Miller --- net/wimax/op-rfkill.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/wimax/op-rfkill.c b/net/wimax/op-rfkill.c index 8745bac173f1..2b75aee04217 100644 --- a/net/wimax/op-rfkill.c +++ b/net/wimax/op-rfkill.c @@ -71,7 +71,7 @@ #define D_SUBMODULE op_rfkill #include "debug-levels.h" -#ifdef CONFIG_RFKILL +#if defined(CONFIG_RFKILL) || defined(CONFIG_RFKILL_MODULE) /** -- cgit v1.2.3