diff options
Diffstat (limited to 'drivers/s390/net/qeth_l3_main.c')
-rw-r--r-- | drivers/s390/net/qeth_l3_main.c | 3391 |
1 files changed, 3391 insertions, 0 deletions
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c new file mode 100644 index 00000000000..a856cb47fc7 --- /dev/null +++ b/drivers/s390/net/qeth_l3_main.c @@ -0,0 +1,3391 @@ +/* + * drivers/s390/net/qeth_l3_main.c + * + * Copyright IBM Corp. 2007 + * Author(s): Utz Bacher <utz.bacher@de.ibm.com>, + * Frank Pavlic <fpavlic@de.ibm.com>, + * Thomas Spatzier <tspat@de.ibm.com>, + * Frank Blaschka <frank.blaschka@de.ibm.com> + */ + +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/etherdevice.h> +#include <linux/mii.h> +#include <linux/ip.h> +#include <linux/reboot.h> +#include <linux/inetdevice.h> +#include <linux/igmp.h> + +#include <net/ip.h> +#include <net/arp.h> + +#include <asm/s390_rdev.h> + +#include "qeth_l3.h" +#include "qeth_core_offl.h" + +DEFINE_PER_CPU(char[256], qeth_l3_dbf_txt_buf); + +static int qeth_l3_set_offline(struct ccwgroup_device *); +static int qeth_l3_recover(void *); +static int qeth_l3_stop(struct net_device *); +static void qeth_l3_set_multicast_list(struct net_device *); +static int qeth_l3_neigh_setup(struct net_device *, struct neigh_parms *); +static int qeth_l3_register_addr_entry(struct qeth_card *, + struct qeth_ipaddr *); +static int qeth_l3_deregister_addr_entry(struct qeth_card *, + struct qeth_ipaddr *); +static int __qeth_l3_set_online(struct ccwgroup_device *, int); +static int __qeth_l3_set_offline(struct ccwgroup_device *, int); + + +static int qeth_l3_isxdigit(char *buf) +{ + while (*buf) { + if (!isxdigit(*buf++)) + return 0; + } + return 1; +} + +void qeth_l3_ipaddr4_to_string(const __u8 *addr, char *buf) +{ + sprintf(buf, "%i.%i.%i.%i", addr[0], addr[1], addr[2], addr[3]); +} + +int qeth_l3_string_to_ipaddr4(const char *buf, __u8 *addr) +{ + int count = 0, rc = 0; + int in[4]; + char c; + + rc = sscanf(buf, "%u.%u.%u.%u%c", + &in[0], &in[1], &in[2], &in[3], &c); + if (rc != 4 && (rc != 5 || c != '\n')) + return -EINVAL; + for (count = 0; count < 4; count++) { + if (in[count] > 255) + return -EINVAL; + addr[count] = in[count]; + } + return 0; +} + +void qeth_l3_ipaddr6_to_string(const __u8 *addr, char *buf) +{ + sprintf(buf, "%02x%02x:%02x%02x:%02x%02x:%02x%02x" + ":%02x%02x:%02x%02x:%02x%02x:%02x%02x", + addr[0], addr[1], addr[2], addr[3], + addr[4], addr[5], addr[6], addr[7], + addr[8], addr[9], addr[10], addr[11], + addr[12], addr[13], addr[14], addr[15]); +} + +int qeth_l3_string_to_ipaddr6(const char *buf, __u8 *addr) +{ + const char *end, *end_tmp, *start; + __u16 *in; + char num[5]; + int num2, cnt, out, found, save_cnt; + unsigned short in_tmp[8] = {0, }; + + cnt = out = found = save_cnt = num2 = 0; + end = start = buf; + in = (__u16 *) addr; + memset(in, 0, 16); + while (*end) { + end = strchr(start, ':'); + if (end == NULL) { + end = buf + strlen(buf); + end_tmp = strchr(start, '\n'); + if (end_tmp != NULL) + end = end_tmp; + out = 1; + } + if ((end - start)) { + memset(num, 0, 5); + if ((end - start) > 4) + return -EINVAL; + memcpy(num, start, end - start); + if (!qeth_l3_isxdigit(num)) + return -EINVAL; + sscanf(start, "%x", &num2); + if (found) + in_tmp[save_cnt++] = num2; + else + in[cnt++] = num2; + if (out) + break; + } else { + if (found) + return -EINVAL; + found = 1; + } + start = ++end; + } + if (cnt + save_cnt > 8) + return -EINVAL; + cnt = 7; + while (save_cnt) + in[cnt--] = in_tmp[--save_cnt]; + return 0; +} + +void qeth_l3_ipaddr_to_string(enum qeth_prot_versions proto, const __u8 *addr, + char *buf) +{ + if (proto == QETH_PROT_IPV4) + qeth_l3_ipaddr4_to_string(addr, buf); + else if (proto == QETH_PROT_IPV6) + qeth_l3_ipaddr6_to_string(addr, buf); +} + +int qeth_l3_string_to_ipaddr(const char *buf, enum qeth_prot_versions proto, + __u8 *addr) +{ + if (proto == QETH_PROT_IPV4) + return qeth_l3_string_to_ipaddr4(buf, addr); + else if (proto == QETH_PROT_IPV6) + return qeth_l3_string_to_ipaddr6(buf, addr); + else + return -EINVAL; +} + +static void qeth_l3_convert_addr_to_bits(u8 *addr, u8 *bits, int len) +{ + int i, j; + u8 octet; + + for (i = 0; i < len; ++i) { + octet = addr[i]; + for (j = 7; j >= 0; --j) { + bits[i*8 + j] = octet & 1; + octet >>= 1; + } + } +} + +static int qeth_l3_is_addr_covered_by_ipato(struct qeth_card *card, + struct qeth_ipaddr *addr) +{ + struct qeth_ipato_entry *ipatoe; + u8 addr_bits[128] = {0, }; + u8 ipatoe_bits[128] = {0, }; + int rc = 0; + + if (!card->ipato.enabled) + return 0; + + qeth_l3_convert_addr_to_bits((u8 *) &addr->u, addr_bits, + (addr->proto == QETH_PROT_IPV4)? 4:16); + list_for_each_entry(ipatoe, &card->ipato.entries, entry) { + if (addr->proto != ipatoe->proto) + continue; + qeth_l3_convert_addr_to_bits(ipatoe->addr, ipatoe_bits, + (ipatoe->proto == QETH_PROT_IPV4) ? + 4 : 16); + if (addr->proto == QETH_PROT_IPV4) + rc = !memcmp(addr_bits, ipatoe_bits, + min(32, ipatoe->mask_bits)); + else + rc = !memcmp(addr_bits, ipatoe_bits, + min(128, ipatoe->mask_bits)); + if (rc) + break; + } + /* invert? */ + if ((addr->proto == QETH_PROT_IPV4) && card->ipato.invert4) + rc = !rc; + else if ((addr->proto == QETH_PROT_IPV6) && card->ipato.invert6) + rc = !rc; + + return rc; +} + +/* + * Add IP to be added to todo list. If there is already an "add todo" + * in this list we just incremenent the reference count. + * Returns 0 if we just incremented reference count. + */ +static int __qeth_l3_insert_ip_todo(struct qeth_card *card, + struct qeth_ipaddr *addr, int add) +{ + struct qeth_ipaddr *tmp, *t; + int found = 0; + + list_for_each_entry_safe(tmp, t, card->ip_tbd_list, entry) { + if ((addr->type == QETH_IP_TYPE_DEL_ALL_MC) && + (tmp->type == QETH_IP_TYPE_DEL_ALL_MC)) + return 0; + if ((tmp->proto == QETH_PROT_IPV4) && + (addr->proto == QETH_PROT_IPV4) && + (tmp->type == addr->type) && + (tmp->is_multicast == addr->is_multicast) && + (tmp->u.a4.addr == addr->u.a4.addr) && + (tmp->u.a4.mask == addr->u.a4.mask)) { + found = 1; + break; + } + if ((tmp->proto == QETH_PROT_IPV6) && + (addr->proto == QETH_PROT_IPV6) && + (tmp->type == addr->type) && + (tmp->is_multicast == addr->is_multicast) && + (tmp->u.a6.pfxlen == addr->u.a6.pfxlen) && + (memcmp(&tmp->u.a6.addr, &addr->u.a6.addr, + sizeof(struct in6_addr)) == 0)) { + found = 1; + break; + } + } + if (found) { + if (addr->users != 0) + tmp->users += addr->users; + else + tmp->users += add ? 1 : -1; + if (tmp->users == 0) { + list_del(&tmp->entry); + kfree(tmp); + } + return 0; + } else { + if (addr->type == QETH_IP_TYPE_DEL_ALL_MC) + list_add(&addr->entry, card->ip_tbd_list); + else { + if (addr->users == 0) + addr->users += add ? 1 : -1; + if (add && (addr->type == QETH_IP_TYPE_NORMAL) && + qeth_l3_is_addr_covered_by_ipato(card, addr)) { + QETH_DBF_TEXT(trace, 2, "tkovaddr"); + addr->set_flags |= QETH_IPA_SETIP_TAKEOVER_FLAG; + } + list_add_tail(&addr->entry, card->ip_tbd_list); + } + return 1; + } +} + +static int qeth_l3_delete_ip(struct qeth_card *card, struct qeth_ipaddr *addr) +{ + unsigned long flags; + int rc = 0; + + QETH_DBF_TEXT(trace, 4, "delip"); + + if (addr->proto == QETH_PROT_IPV4) + QETH_DBF_HEX(trace, 4, &addr->u.a4.addr, 4); + else { + QETH_DBF_HEX(trace, 4, &addr->u.a6.addr, 8); + QETH_DBF_HEX(trace, 4, ((char *)&addr->u.a6.addr) + 8, 8); + } + spin_lock_irqsave(&card->ip_lock, flags); + rc = __qeth_l3_insert_ip_todo(card, addr, 0); + spin_unlock_irqrestore(&card->ip_lock, flags); + return rc; +} + +static int qeth_l3_add_ip(struct qeth_card *card, struct qeth_ipaddr *addr) +{ + unsigned long flags; + int rc = 0; + + QETH_DBF_TEXT(trace, 4, "addip"); + if (addr->proto == QETH_PROT_IPV4) + QETH_DBF_HEX(trace, 4, &addr->u.a4.addr, 4); + else { + QETH_DBF_HEX(trace, 4, &addr->u.a6.addr, 8); + QETH_DBF_HEX(trace, 4, ((char *)&addr->u.a6.addr) + 8, 8); + } + spin_lock_irqsave(&card->ip_lock, flags); + rc = __qeth_l3_insert_ip_todo(card, addr, 1); + spin_unlock_irqrestore(&card->ip_lock, flags); + return rc; +} + + +static struct qeth_ipaddr *qeth_l3_get_addr_buffer( + enum qeth_prot_versions prot) +{ + struct qeth_ipaddr *addr; + + addr = kzalloc(sizeof(struct qeth_ipaddr), GFP_ATOMIC); + if (addr == NULL) { + PRINT_WARN("Not enough memory to add address\n"); + return NULL; + } + addr->type = QETH_IP_TYPE_NORMAL; + addr->proto = prot; + return addr; +} + +static void qeth_l3_delete_mc_addresses(struct qeth_card *card) +{ + struct qeth_ipaddr *iptodo; + unsigned long flags; + + QETH_DBF_TEXT(trace, 4, "delmc"); + iptodo = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (!iptodo) { + QETH_DBF_TEXT(trace, 2, "dmcnomem"); + return; + } + iptodo->type = QETH_IP_TYPE_DEL_ALL_MC; + spin_lock_irqsave(&card->ip_lock, flags); + if (!__qeth_l3_insert_ip_todo(card, iptodo, 0)) + kfree(iptodo); + spin_unlock_irqrestore(&card->ip_lock, flags); +} + +/* + * Add/remove address to/from card's ip list, i.e. try to add or remove + * reference to/from an IP address that is already registered on the card. + * Returns: + * 0 address was on card and its reference count has been adjusted, + * but is still > 0, so nothing has to be done + * also returns 0 if card was not on card and the todo was to delete + * the address -> there is also nothing to be done + * 1 address was not on card and the todo is to add it to the card's ip + * list + * -1 address was on card and its reference count has been decremented + * to <= 0 by the todo -> address must be removed from card + */ +static int __qeth_l3_ref_ip_on_card(struct qeth_card *card, + struct qeth_ipaddr *todo, struct qeth_ipaddr **__addr) +{ + struct qeth_ipaddr *addr; + int found = 0; + + list_for_each_entry(addr, &card->ip_list, entry) { + if ((addr->proto == QETH_PROT_IPV4) && + (todo->proto == QETH_PROT_IPV4) && + (addr->type == todo->type) && + (addr->u.a4.addr == todo->u.a4.addr) && + (addr->u.a4.mask == todo->u.a4.mask)) { + found = 1; + break; + } + if ((addr->proto == QETH_PROT_IPV6) && + (todo->proto == QETH_PROT_IPV6) && + (addr->type == todo->type) && + (addr->u.a6.pfxlen == todo->u.a6.pfxlen) && + (memcmp(&addr->u.a6.addr, &todo->u.a6.addr, + sizeof(struct in6_addr)) == 0)) { + found = 1; + break; + } + } + if (found) { + addr->users += todo->users; + if (addr->users <= 0) { + *__addr = addr; + return -1; + } else { + /* for VIPA and RXIP limit refcount to 1 */ + if (addr->type != QETH_IP_TYPE_NORMAL) + addr->users = 1; + return 0; + } + } + if (todo->users > 0) { + /* for VIPA and RXIP limit refcount to 1 */ + if (todo->type != QETH_IP_TYPE_NORMAL) + todo->users = 1; + return 1; + } else + return 0; +} + +static void __qeth_l3_delete_all_mc(struct qeth_card *card, + unsigned long *flags) +{ + struct qeth_ipaddr *addr, *tmp; + int rc; +again: + list_for_each_entry_safe(addr, tmp, &card->ip_list, entry) { + if (addr->is_multicast) { + list_del(&addr->entry); + spin_unlock_irqrestore(&card->ip_lock, *flags); + rc = qeth_l3_deregister_addr_entry(card, addr); + spin_lock_irqsave(&card->ip_lock, *flags); + if (!rc) { + kfree(addr); + goto again; + } else + list_add(&addr->entry, &card->ip_list); + } + } +} + +static void qeth_l3_set_ip_addr_list(struct qeth_card *card) +{ + struct list_head *tbd_list; + struct qeth_ipaddr *todo, *addr; + unsigned long flags; + int rc; + + QETH_DBF_TEXT(trace, 2, "sdiplist"); + QETH_DBF_HEX(trace, 2, &card, sizeof(void *)); + + spin_lock_irqsave(&card->ip_lock, flags); + tbd_list = card->ip_tbd_list; + card->ip_tbd_list = kmalloc(sizeof(struct list_head), GFP_ATOMIC); + if (!card->ip_tbd_list) { + QETH_DBF_TEXT(trace, 0, "silnomem"); + card->ip_tbd_list = tbd_list; + spin_unlock_irqrestore(&card->ip_lock, flags); + return; + } else + INIT_LIST_HEAD(card->ip_tbd_list); + + while (!list_empty(tbd_list)) { + todo = list_entry(tbd_list->next, struct qeth_ipaddr, entry); + list_del(&todo->entry); + if (todo->type == QETH_IP_TYPE_DEL_ALL_MC) { + __qeth_l3_delete_all_mc(card, &flags); + kfree(todo); + continue; + } + rc = __qeth_l3_ref_ip_on_card(card, todo, &addr); + if (rc == 0) { + /* nothing to be done; only adjusted refcount */ + kfree(todo); + } else if (rc == 1) { + /* new entry to be added to on-card list */ + spin_unlock_irqrestore(&card->ip_lock, flags); + rc = qeth_l3_register_addr_entry(card, todo); + spin_lock_irqsave(&card->ip_lock, flags); + if (!rc) + list_add_tail(&todo->entry, &card->ip_list); + else + kfree(todo); + } else if (rc == -1) { + /* on-card entry to be removed */ + list_del_init(&addr->entry); + spin_unlock_irqrestore(&card->ip_lock, flags); + rc = qeth_l3_deregister_addr_entry(card, addr); + spin_lock_irqsave(&card->ip_lock, flags); + if (!rc) + kfree(addr); + else + list_add_tail(&addr->entry, &card->ip_list); + kfree(todo); + } + } + spin_unlock_irqrestore(&card->ip_lock, flags); + kfree(tbd_list); +} + +static void qeth_l3_clear_ip_list(struct qeth_card *card, int clean, + int recover) +{ + struct qeth_ipaddr *addr, *tmp; + unsigned long flags; + + QETH_DBF_TEXT(trace, 4, "clearip"); + spin_lock_irqsave(&card->ip_lock, flags); + /* clear todo list */ + list_for_each_entry_safe(addr, tmp, card->ip_tbd_list, entry) { + list_del(&addr->entry); + kfree(addr); + } + + while (!list_empty(&card->ip_list)) { + addr = list_entry(card->ip_list.next, + struct qeth_ipaddr, entry); + list_del_init(&addr->entry); + if (clean) { + spin_unlock_irqrestore(&card->ip_lock, flags); + qeth_l3_deregister_addr_entry(card, addr); + spin_lock_irqsave(&card->ip_lock, flags); + } + if (!recover || addr->is_multicast) { + kfree(addr); + continue; + } + list_add_tail(&addr->entry, card->ip_tbd_list); + } + spin_unlock_irqrestore(&card->ip_lock, flags); +} + +static int qeth_l3_address_exists_in_list(struct list_head *list, + struct qeth_ipaddr *addr, int same_type) +{ + struct qeth_ipaddr *tmp; + + list_for_each_entry(tmp, list, entry) { + if ((tmp->proto == QETH_PROT_IPV4) && + (addr->proto == QETH_PROT_IPV4) && + ((same_type && (tmp->type == addr->type)) || + (!same_type && (tmp->type != addr->type))) && + (tmp->u.a4.addr == addr->u.a4.addr)) + return 1; + + if ((tmp->proto == QETH_PROT_IPV6) && + (addr->proto == QETH_PROT_IPV6) && + ((same_type && (tmp->type == addr->type)) || + (!same_type && (tmp->type != addr->type))) && + (memcmp(&tmp->u.a6.addr, &addr->u.a6.addr, + sizeof(struct in6_addr)) == 0)) + return 1; + + } + return 0; +} + +static int qeth_l3_send_setdelmc(struct qeth_card *card, + struct qeth_ipaddr *addr, int ipacmd) +{ + int rc; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 4, "setdelmc"); + + iob = qeth_get_ipacmd_buffer(card, ipacmd, addr->proto); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + memcpy(&cmd->data.setdelipm.mac, addr->mac, OSA_ADDR_LEN); + if (addr->proto == QETH_PROT_IPV6) + memcpy(cmd->data.setdelipm.ip6, &addr->u.a6.addr, + sizeof(struct in6_addr)); + else + memcpy(&cmd->data.setdelipm.ip4, &addr->u.a4.addr, 4); + + rc = qeth_send_ipa_cmd(card, iob, NULL, NULL); + + return rc; +} + +static void qeth_l3_fill_netmask(u8 *netmask, unsigned int len) +{ + int i, j; + for (i = 0; i < 16; i++) { + j = (len) - (i * 8); + if (j >= 8) + netmask[i] = 0xff; + else if (j > 0) + netmask[i] = (u8)(0xFF00 >> j); + else + netmask[i] = 0; + } +} + +static int qeth_l3_send_setdelip(struct qeth_card *card, + struct qeth_ipaddr *addr, int ipacmd, unsigned int flags) +{ + int rc; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + __u8 netmask[16]; + + QETH_DBF_TEXT(trace, 4, "setdelip"); + QETH_DBF_TEXT_(trace, 4, "flags%02X", flags); + + iob = qeth_get_ipacmd_buffer(card, ipacmd, addr->proto); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + if (addr->proto == QETH_PROT_IPV6) { + memcpy(cmd->data.setdelip6.ip_addr, &addr->u.a6.addr, + sizeof(struct in6_addr)); + qeth_l3_fill_netmask(netmask, addr->u.a6.pfxlen); + memcpy(cmd->data.setdelip6.mask, netmask, + sizeof(struct in6_addr)); + cmd->data.setdelip6.flags = flags; + } else { + memcpy(cmd->data.setdelip4.ip_addr, &addr->u.a4.addr, 4); + memcpy(cmd->data.setdelip4.mask, &addr->u.a4.mask, 4); + cmd->data.setdelip4.flags = flags; + } + + rc = qeth_send_ipa_cmd(card, iob, NULL, NULL); + + return rc; +} + +static int qeth_l3_send_setrouting(struct qeth_card *card, + enum qeth_routing_types type, enum qeth_prot_versions prot) +{ + int rc; + struct qeth_ipa_cmd *cmd; + struct qeth_cmd_buffer *iob; + + QETH_DBF_TEXT(trace, 4, "setroutg"); + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_SETRTG, prot); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + cmd->data.setrtg.type = (type); + rc = qeth_send_ipa_cmd(card, iob, NULL, NULL); + + return rc; +} + +static void qeth_l3_correct_routing_type(struct qeth_card *card, + enum qeth_routing_types *type, enum qeth_prot_versions prot) +{ + if (card->info.type == QETH_CARD_TYPE_IQD) { + switch (*type) { + case NO_ROUTER: + case PRIMARY_CONNECTOR: + case SECONDARY_CONNECTOR: + case MULTICAST_ROUTER: + return; + default: + goto out_inval; + } + } else { + switch (*type) { + case NO_ROUTER: + case PRIMARY_ROUTER: + case SECONDARY_ROUTER: + return; + case MULTICAST_ROUTER: + if (qeth_is_ipafunc_supported(card, prot, + IPA_OSA_MC_ROUTER)) + return; + default: + goto out_inval; + } + } +out_inval: + PRINT_WARN("Routing type '%s' not supported for interface %s.\n" + "Router status set to 'no router'.\n", + ((*type == PRIMARY_ROUTER)? "primary router" : + (*type == SECONDARY_ROUTER)? "secondary router" : + (*type == PRIMARY_CONNECTOR)? "primary connector" : + (*type == SECONDARY_CONNECTOR)? "secondary connector" : + (*type == MULTICAST_ROUTER)? "multicast router" : + "unknown"), + card->dev->name); + *type = NO_ROUTER; +} + +int qeth_l3_setrouting_v4(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "setrtg4"); + + qeth_l3_correct_routing_type(card, &card->options.route4.type, + QETH_PROT_IPV4); + + rc = qeth_l3_send_setrouting(card, card->options.route4.type, + QETH_PROT_IPV4); + if (rc) { + card->options.route4.type = NO_ROUTER; + PRINT_WARN("Error (0x%04x) while setting routing type on %s. " + "Type set to 'no router'.\n", + rc, QETH_CARD_IFNAME(card)); + } + return rc; +} + +int qeth_l3_setrouting_v6(struct qeth_card *card) +{ + int rc = 0; + + QETH_DBF_TEXT(trace, 3, "setrtg6"); +#ifdef CONFIG_QETH_IPV6 + + if (!qeth_is_supported(card, IPA_IPV6)) + return 0; + qeth_l3_correct_routing_type(card, &card->options.route6.type, + QETH_PROT_IPV6); + + rc = qeth_l3_send_setrouting(card, card->options.route6.type, + QETH_PROT_IPV6); + if (rc) { + card->options.route6.type = NO_ROUTER; + PRINT_WARN("Error (0x%04x) while setting routing type on %s. " + "Type set to 'no router'.\n", + rc, QETH_CARD_IFNAME(card)); + } +#endif + return rc; +} + +/* + * IP address takeover related functions + */ +static void qeth_l3_clear_ipato_list(struct qeth_card *card) +{ + + struct qeth_ipato_entry *ipatoe, *tmp; + unsigned long flags; + + spin_lock_irqsave(&card->ip_lock, flags); + list_for_each_entry_safe(ipatoe, tmp, &card->ipato.entries, entry) { + list_del(&ipatoe->entry); + kfree(ipatoe); + } + spin_unlock_irqrestore(&card->ip_lock, flags); +} + +int qeth_l3_add_ipato_entry(struct qeth_card *card, + struct qeth_ipato_entry *new) +{ + struct qeth_ipato_entry *ipatoe; + unsigned long flags; + int rc = 0; + + QETH_DBF_TEXT(trace, 2, "addipato"); + spin_lock_irqsave(&card->ip_lock, flags); + list_for_each_entry(ipatoe, &card->ipato.entries, entry) { + if (ipatoe->proto != new->proto) + continue; + if (!memcmp(ipatoe->addr, new->addr, + (ipatoe->proto == QETH_PROT_IPV4)? 4:16) && + (ipatoe->mask_bits == new->mask_bits)) { + PRINT_WARN("ipato entry already exists!\n"); + rc = -EEXIST; + break; + } + } + if (!rc) + list_add_tail(&new->entry, &card->ipato.entries); + + spin_unlock_irqrestore(&card->ip_lock, flags); + return rc; +} + +void qeth_l3_del_ipato_entry(struct qeth_card *card, + enum qeth_prot_versions proto, u8 *addr, int mask_bits) +{ + struct qeth_ipato_entry *ipatoe, *tmp; + unsigned long flags; + + QETH_DBF_TEXT(trace, 2, "delipato"); + spin_lock_irqsave(&card->ip_lock, flags); + list_for_each_entry_safe(ipatoe, tmp, &card->ipato.entries, entry) { + if (ipatoe->proto != proto) + continue; + if (!memcmp(ipatoe->addr, addr, + (proto == QETH_PROT_IPV4)? 4:16) && + (ipatoe->mask_bits == mask_bits)) { + list_del(&ipatoe->entry); + kfree(ipatoe); + } + } + spin_unlock_irqrestore(&card->ip_lock, flags); +} + +/* + * VIPA related functions + */ +int qeth_l3_add_vipa(struct qeth_card *card, enum qeth_prot_versions proto, + const u8 *addr) +{ + struct qeth_ipaddr *ipaddr; + unsigned long flags; + int rc = 0; + + ipaddr = qeth_l3_get_addr_buffer(proto); + if (ipaddr) { + if (proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "addvipa4"); + memcpy(&ipaddr->u.a4.addr, addr, 4); + ipaddr->u.a4.mask = 0; + } else if (proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "addvipa6"); + memcpy(&ipaddr->u.a6.addr, addr, 16); + ipaddr->u.a6.pfxlen = 0; + } + ipaddr->type = QETH_IP_TYPE_VIPA; + ipaddr->set_flags = QETH_IPA_SETIP_VIPA_FLAG; + ipaddr->del_flags = QETH_IPA_DELIP_VIPA_FLAG; + } else + return -ENOMEM; + spin_lock_irqsave(&card->ip_lock, flags); + if (qeth_l3_address_exists_in_list(&card->ip_list, ipaddr, 0) || + qeth_l3_address_exists_in_list(card->ip_tbd_list, ipaddr, 0)) + rc = -EEXIST; + spin_unlock_irqrestore(&card->ip_lock, flags); + if (rc) { + PRINT_WARN("Cannot add VIPA. Address already exists!\n"); + return rc; + } + if (!qeth_l3_add_ip(card, ipaddr)) + kfree(ipaddr); + qeth_l3_set_ip_addr_list(card); + return rc; +} + +void qeth_l3_del_vipa(struct qeth_card *card, enum qeth_prot_versions proto, + const u8 *addr) +{ + struct qeth_ipaddr *ipaddr; + + ipaddr = qeth_l3_get_addr_buffer(proto); + if (ipaddr) { + if (proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "delvipa4"); + memcpy(&ipaddr->u.a4.addr, addr, 4); + ipaddr->u.a4.mask = 0; + } else if (proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "delvipa6"); + memcpy(&ipaddr->u.a6.addr, addr, 16); + ipaddr->u.a6.pfxlen = 0; + } + ipaddr->type = QETH_IP_TYPE_VIPA; + } else + return; + if (!qeth_l3_delete_ip(card, ipaddr)) + kfree(ipaddr); + qeth_l3_set_ip_addr_list(card); +} + +/* + * proxy ARP related functions + */ +int qeth_l3_add_rxip(struct qeth_card *card, enum qeth_prot_versions proto, + const u8 *addr) +{ + struct qeth_ipaddr *ipaddr; + unsigned long flags; + int rc = 0; + + ipaddr = qeth_l3_get_addr_buffer(proto); + if (ipaddr) { + if (proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "addrxip4"); + memcpy(&ipaddr->u.a4.addr, addr, 4); + ipaddr->u.a4.mask = 0; + } else if (proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "addrxip6"); + memcpy(&ipaddr->u.a6.addr, addr, 16); + ipaddr->u.a6.pfxlen = 0; + } + ipaddr->type = QETH_IP_TYPE_RXIP; + ipaddr->set_flags = QETH_IPA_SETIP_TAKEOVER_FLAG; + ipaddr->del_flags = 0; + } else + return -ENOMEM; + spin_lock_irqsave(&card->ip_lock, flags); + if (qeth_l3_address_exists_in_list(&card->ip_list, ipaddr, 0) || + qeth_l3_address_exists_in_list(card->ip_tbd_list, ipaddr, 0)) + rc = -EEXIST; + spin_unlock_irqrestore(&card->ip_lock, flags); + if (rc) { + PRINT_WARN("Cannot add RXIP. Address already exists!\n"); + return rc; + } + if (!qeth_l3_add_ip(card, ipaddr)) + kfree(ipaddr); + qeth_l3_set_ip_addr_list(card); + return 0; +} + +void qeth_l3_del_rxip(struct qeth_card *card, enum qeth_prot_versions proto, + const u8 *addr) +{ + struct qeth_ipaddr *ipaddr; + + ipaddr = qeth_l3_get_addr_buffer(proto); + if (ipaddr) { + if (proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "addrxip4"); + memcpy(&ipaddr->u.a4.addr, addr, 4); + ipaddr->u.a4.mask = 0; + } else if (proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "addrxip6"); + memcpy(&ipaddr->u.a6.addr, addr, 16); + ipaddr->u.a6.pfxlen = 0; + } + ipaddr->type = QETH_IP_TYPE_RXIP; + } else + return; + if (!qeth_l3_delete_ip(card, ipaddr)) + kfree(ipaddr); + qeth_l3_set_ip_addr_list(card); +} + +static int qeth_l3_register_addr_entry(struct qeth_card *card, + struct qeth_ipaddr *addr) +{ + char buf[50]; + int rc = 0; + int cnt = 3; + + if (addr->proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "setaddr4"); + QETH_DBF_HEX(trace, 3, &addr->u.a4.addr, sizeof(int)); + } else if (addr->proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "setaddr6"); + QETH_DBF_HEX(trace, 3, &addr->u.a6.addr, 8); + QETH_DBF_HEX(trace, 3, ((char *)&addr->u.a6.addr) + 8, 8); + } else { + QETH_DBF_TEXT(trace, 2, "setaddr?"); + QETH_DBF_HEX(trace, 3, addr, sizeof(struct qeth_ipaddr)); + } + do { + if (addr->is_multicast) + rc = qeth_l3_send_setdelmc(card, addr, IPA_CMD_SETIPM); + else + rc = qeth_l3_send_setdelip(card, addr, IPA_CMD_SETIP, + addr->set_flags); + if (rc) + QETH_DBF_TEXT(trace, 2, "failed"); + } while ((--cnt > 0) && rc); + if (rc) { + QETH_DBF_TEXT(trace, 2, "FAILED"); + qeth_l3_ipaddr_to_string(addr->proto, (u8 *)&addr->u, buf); + PRINT_WARN("Could not register IP address %s (rc=0x%x/%d)\n", + buf, rc, rc); + } + return rc; +} + +static int qeth_l3_deregister_addr_entry(struct qeth_card *card, + struct qeth_ipaddr *addr) +{ + int rc = 0; + + if (addr->proto == QETH_PROT_IPV4) { + QETH_DBF_TEXT(trace, 2, "deladdr4"); + QETH_DBF_HEX(trace, 3, &addr->u.a4.addr, sizeof(int)); + } else if (addr->proto == QETH_PROT_IPV6) { + QETH_DBF_TEXT(trace, 2, "deladdr6"); + QETH_DBF_HEX(trace, 3, &addr->u.a6.addr, 8); + QETH_DBF_HEX(trace, 3, ((char *)&addr->u.a6.addr) + 8, 8); + } else { + QETH_DBF_TEXT(trace, 2, "deladdr?"); + QETH_DBF_HEX(trace, 3, addr, sizeof(struct qeth_ipaddr)); + } + if (addr->is_multicast) + rc = qeth_l3_send_setdelmc(card, addr, IPA_CMD_DELIPM); + else + rc = qeth_l3_send_setdelip(card, addr, IPA_CMD_DELIP, + addr->del_flags); + if (rc) { + QETH_DBF_TEXT(trace, 2, "failed"); + /* TODO: re-activate this warning as soon as we have a + * clean mirco code + qeth_ipaddr_to_string(addr->proto, (u8 *)&addr->u, buf); + PRINT_WARN("Could not deregister IP address %s (rc=%x)\n", + buf, rc); + */ + } + + return rc; +} + +static inline u8 qeth_l3_get_qeth_hdr_flags4(int cast_type) +{ + if (cast_type == RTN_MULTICAST) + return QETH_CAST_MULTICAST; + if (cast_type == RTN_BROADCAST) + return QETH_CAST_BROADCAST; + return QETH_CAST_UNICAST; +} + +static inline u8 qeth_l3_get_qeth_hdr_flags6(int cast_type) +{ + u8 ct = QETH_HDR_PASSTHRU | QETH_HDR_IPV6; + if (cast_type == RTN_MULTICAST) + return ct | QETH_CAST_MULTICAST; + if (cast_type == RTN_ANYCAST) + return ct | QETH_CAST_ANYCAST; + if (cast_type == RTN_BROADCAST) + return ct | QETH_CAST_BROADCAST; + return ct | QETH_CAST_UNICAST; +} + +static int qeth_l3_send_setadp_mode(struct qeth_card *card, __u32 command, + __u32 mode) +{ + int rc; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 4, "adpmode"); + + iob = qeth_get_adapter_cmd(card, command, + sizeof(struct qeth_ipacmd_setadpparms)); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + cmd->data.setadapterparms.data.mode = mode; + rc = qeth_send_ipa_cmd(card, iob, qeth_default_setadapterparms_cb, + NULL); + return rc; +} + +static int qeth_l3_setadapter_hstr(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 4, "adphstr"); + + if (qeth_adp_supported(card, IPA_SETADP_SET_BROADCAST_MODE)) { + rc = qeth_l3_send_setadp_mode(card, + IPA_SETADP_SET_BROADCAST_MODE, + card->options.broadcast_mode); + if (rc) + PRINT_WARN("couldn't set broadcast mode on " + "device %s: x%x\n", + CARD_BUS_ID(card), rc); + rc = qeth_l3_send_setadp_mode(card, + IPA_SETADP_ALTER_MAC_ADDRESS, + card->options.macaddr_mode); + if (rc) + PRINT_WARN("couldn't set macaddr mode on " + "device %s: x%x\n", CARD_BUS_ID(card), rc); + return rc; + } + if (card->options.broadcast_mode == QETH_TR_BROADCAST_LOCAL) + PRINT_WARN("set adapter parameters not available " + "to set broadcast mode, using ALLRINGS " + "on device %s:\n", CARD_BUS_ID(card)); + if (card->options.macaddr_mode == QETH_TR_MACADDR_CANONICAL) + PRINT_WARN("set adapter parameters not available " + "to set macaddr mode, using NONCANONICAL " + "on device %s:\n", CARD_BUS_ID(card)); + return 0; +} + +static int qeth_l3_setadapter_parms(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(setup, 2, "setadprm"); + + if (!qeth_is_supported(card, IPA_SETADAPTERPARMS)) { + PRINT_WARN("set adapter parameters not supported " + "on device %s.\n", + CARD_BUS_ID(card)); + QETH_DBF_TEXT(setup, 2, " notsupp"); + return 0; + } + rc = qeth_query_setadapterparms(card); + if (rc) { + PRINT_WARN("couldn't set adapter parameters on device %s: " + "x%x\n", CARD_BUS_ID(card), rc); + return rc; + } + if (qeth_adp_supported(card, IPA_SETADP_ALTER_MAC_ADDRESS)) { + rc = qeth_setadpparms_change_macaddr(card); + if (rc) + PRINT_WARN("couldn't get MAC address on " + "device %s: x%x\n", + CARD_BUS_ID(card), rc); + } + + if ((card->info.link_type == QETH_LINK_TYPE_HSTR) || + (card->info.link_type == QETH_LINK_TYPE_LANE_TR)) + rc = qeth_l3_setadapter_hstr(card); + + return rc; +} + +static int qeth_l3_default_setassparms_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) +{ + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 4, "defadpcb"); + + cmd = (struct qeth_ipa_cmd *) data; + if (cmd->hdr.return_code == 0) { + cmd->hdr.return_code = cmd->data.setassparms.hdr.return_code; + if (cmd->hdr.prot_version == QETH_PROT_IPV4) + card->options.ipa4.enabled_funcs = cmd->hdr.ipa_enabled; + if (cmd->hdr.prot_version == QETH_PROT_IPV6) + card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled; + } + if (cmd->data.setassparms.hdr.assist_no == IPA_INBOUND_CHECKSUM && + cmd->data.setassparms.hdr.command_code == IPA_CMD_ASS_START) { + card->info.csum_mask = cmd->data.setassparms.data.flags_32bit; + QETH_DBF_TEXT_(trace, 3, "csum:%d", card->info.csum_mask); + } + return 0; +} + +static struct qeth_cmd_buffer *qeth_l3_get_setassparms_cmd( + struct qeth_card *card, enum qeth_ipa_funcs ipa_func, __u16 cmd_code, + __u16 len, enum qeth_prot_versions prot) +{ + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 4, "getasscm"); + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_SETASSPARMS, prot); + + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + cmd->data.setassparms.hdr.assist_no = ipa_func; + cmd->data.setassparms.hdr.length = 8 + len; + cmd->data.setassparms.hdr.command_code = cmd_code; + cmd->data.setassparms.hdr.return_code = 0; + cmd->data.setassparms.hdr.seq_no = 0; + + return iob; +} + +static int qeth_l3_send_setassparms(struct qeth_card *card, + struct qeth_cmd_buffer *iob, __u16 len, long data, + int (*reply_cb)(struct qeth_card *, struct qeth_reply *, + unsigned long), + void *reply_param) +{ + int rc; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 4, "sendassp"); + + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + if (len <= sizeof(__u32)) + cmd->data.setassparms.data.flags_32bit = (__u32) data; + else /* (len > sizeof(__u32)) */ + memcpy(&cmd->data.setassparms.data, (void *) data, len); + + rc = qeth_send_ipa_cmd(card, iob, reply_cb, reply_param); + return rc; +} + +#ifdef CONFIG_QETH_IPV6 +static int qeth_l3_send_simple_setassparms_ipv6(struct qeth_card *card, + enum qeth_ipa_funcs ipa_func, __u16 cmd_code) +{ + int rc; + struct qeth_cmd_buffer *iob; + + QETH_DBF_TEXT(trace, 4, "simassp6"); + iob = qeth_l3_get_setassparms_cmd(card, ipa_func, cmd_code, + 0, QETH_PROT_IPV6); + rc = qeth_l3_send_setassparms(card, iob, 0, 0, + qeth_l3_default_setassparms_cb, NULL); + return rc; +} +#endif + +static int qeth_l3_send_simple_setassparms(struct qeth_card *card, + enum qeth_ipa_funcs ipa_func, __u16 cmd_code, long data) +{ + int rc; + int length = 0; + struct qeth_cmd_buffer *iob; + + QETH_DBF_TEXT(trace, 4, "simassp4"); + if (data) + length = sizeof(__u32); + iob = qeth_l3_get_setassparms_cmd(card, ipa_func, cmd_code, + length, QETH_PROT_IPV4); + rc = qeth_l3_send_setassparms(card, iob, length, data, + qeth_l3_default_setassparms_cb, NULL); + return rc; +} + +static int qeth_l3_start_ipa_arp_processing(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "ipaarp"); + + if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return 0; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Could not start ARP processing " + "assist on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + } + return rc; +} + +static int qeth_l3_start_ipa_ip_fragmentation(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "ipaipfrg"); + + if (!qeth_is_supported(card, IPA_IP_FRAGMENTATION)) { + PRINT_INFO("Hardware IP fragmentation not supported on %s\n", + QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + + rc = qeth_l3_send_simple_setassparms(card, IPA_IP_FRAGMENTATION, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Could not start Hardware IP fragmentation " + "assist on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + } else + PRINT_INFO("Hardware IP fragmentation enabled \n"); + return rc; +} + +static int qeth_l3_start_ipa_source_mac(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "stsrcmac"); + + if (!card->options.fake_ll) + return -EOPNOTSUPP; + + if (!qeth_is_supported(card, IPA_SOURCE_MAC)) { + PRINT_INFO("Inbound source address not " + "supported on %s\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + + rc = qeth_l3_send_simple_setassparms(card, IPA_SOURCE_MAC, + IPA_CMD_ASS_START, 0); + if (rc) + PRINT_WARN("Could not start inbound source " + "assist on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + return rc; +} + +static int qeth_l3_start_ipa_vlan(struct qeth_card *card) +{ + int rc = 0; + + QETH_DBF_TEXT(trace, 3, "strtvlan"); + + if (!qeth_is_supported(card, IPA_FULL_VLAN)) { + PRINT_WARN("VLAN not supported on %s\n", + QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + + rc = qeth_l3_send_simple_setassparms(card, IPA_VLAN_PRIO, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Could not start vlan " + "assist on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + } else { + PRINT_INFO("VLAN enabled \n"); + } + return rc; +} + +static int qeth_l3_start_ipa_multicast(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "stmcast"); + + if (!qeth_is_supported(card, IPA_MULTICASTING)) { + PRINT_WARN("Multicast not supported on %s\n", + QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + + rc = qeth_l3_send_simple_setassparms(card, IPA_MULTICASTING, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Could not start multicast " + "assist on %s: rc=%i\n", + QETH_CARD_IFNAME(card), rc); + } else { + PRINT_INFO("Multicast enabled\n"); + card->dev->flags |= IFF_MULTICAST; + } + return rc; +} + +static int qeth_l3_query_ipassists_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) +{ + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(setup, 2, "qipasscb"); + + cmd = (struct qeth_ipa_cmd *) data; + if (cmd->hdr.prot_version == QETH_PROT_IPV4) { + card->options.ipa4.supported_funcs = cmd->hdr.ipa_supported; + card->options.ipa4.enabled_funcs = cmd->hdr.ipa_enabled; + } else { + card->options.ipa6.supported_funcs = cmd->hdr.ipa_supported; + card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled; + } + QETH_DBF_TEXT(setup, 2, "suppenbl"); + QETH_DBF_TEXT_(setup, 2, "%x", cmd->hdr.ipa_supported); + QETH_DBF_TEXT_(setup, 2, "%x", cmd->hdr.ipa_enabled); + return 0; +} + +static int qeth_l3_query_ipassists(struct qeth_card *card, + enum qeth_prot_versions prot) +{ + int rc; + struct qeth_cmd_buffer *iob; + + QETH_DBF_TEXT_(setup, 2, "qipassi%i", prot); + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_QIPASSIST, prot); + rc = qeth_send_ipa_cmd(card, iob, qeth_l3_query_ipassists_cb, NULL); + return rc; +} + +#ifdef CONFIG_QETH_IPV6 +static int qeth_l3_softsetup_ipv6(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "softipv6"); + + if (card->info.type == QETH_CARD_TYPE_IQD) + goto out; + + rc = qeth_l3_query_ipassists(card, QETH_PROT_IPV6); + if (rc) { + PRINT_ERR("IPv6 query ipassist failed on %s\n", + QETH_CARD_IFNAME(card)); + return rc; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_IPV6, + IPA_CMD_ASS_START, 3); + if (rc) { + PRINT_WARN("IPv6 start assist (version 4) failed " + "on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + return rc; + } + rc = qeth_l3_send_simple_setassparms_ipv6(card, IPA_IPV6, + IPA_CMD_ASS_START); + if (rc) { + PRINT_WARN("IPV6 start assist (version 6) failed " + "on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + return rc; + } + rc = qeth_l3_send_simple_setassparms_ipv6(card, IPA_PASSTHRU, + IPA_CMD_ASS_START); + if (rc) { + PRINT_WARN("Could not enable passthrough " + "on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + return rc; + } +out: + PRINT_INFO("IPV6 enabled \n"); + return 0; +} +#endif + +static int qeth_l3_start_ipa_ipv6(struct qeth_card *card) +{ + int rc = 0; + + QETH_DBF_TEXT(trace, 3, "strtipv6"); + + if (!qeth_is_supported(card, IPA_IPV6)) { + PRINT_WARN("IPv6 not supported on %s\n", + QETH_CARD_IFNAME(card)); + return 0; + } +#ifdef CONFIG_QETH_IPV6 + rc = qeth_l3_softsetup_ipv6(card); +#endif + return rc ; +} + +static int qeth_l3_start_ipa_broadcast(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "stbrdcst"); + card->info.broadcast_capable = 0; + if (!qeth_is_supported(card, IPA_FILTERING)) { + PRINT_WARN("Broadcast not supported on %s\n", + QETH_CARD_IFNAME(card)); + rc = -EOPNOTSUPP; + goto out; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_FILTERING, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Could not enable broadcasting filtering " + "on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + goto out; + } + + rc = qeth_l3_send_simple_setassparms(card, IPA_FILTERING, + IPA_CMD_ASS_CONFIGURE, 1); + if (rc) { + PRINT_WARN("Could not set up broadcast filtering on %s: 0x%x\n", + QETH_CARD_IFNAME(card), rc); + goto out; + } + card->info.broadcast_capable = QETH_BROADCAST_WITH_ECHO; + PRINT_INFO("Broadcast enabled \n"); + rc = qeth_l3_send_simple_setassparms(card, IPA_FILTERING, + IPA_CMD_ASS_ENABLE, 1); + if (rc) { + PRINT_WARN("Could not set up broadcast echo filtering on " + "%s: 0x%x\n", QETH_CARD_IFNAME(card), rc); + goto out; + } + card->info.broadcast_capable = QETH_BROADCAST_WITHOUT_ECHO; +out: + if (card->info.broadcast_capable) + card->dev->flags |= IFF_BROADCAST; + else + card->dev->flags &= ~IFF_BROADCAST; + return rc; +} + +static int qeth_l3_send_checksum_command(struct qeth_card *card) +{ + int rc; + + rc = qeth_l3_send_simple_setassparms(card, IPA_INBOUND_CHECKSUM, + IPA_CMD_ASS_START, 0); + if (rc) { + PRINT_WARN("Starting Inbound HW Checksumming failed on %s: " + "0x%x,\ncontinuing using Inbound SW Checksumming\n", + QETH_CARD_IFNAME(card), rc); + return rc; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_INBOUND_CHECKSUM, + IPA_CMD_ASS_ENABLE, + card->info.csum_mask); + if (rc) { + PRINT_WARN("Enabling Inbound HW Checksumming failed on %s: " + "0x%x,\ncontinuing using Inbound SW Checksumming\n", + QETH_CARD_IFNAME(card), rc); + return rc; + } + return 0; +} + +static int qeth_l3_start_ipa_checksum(struct qeth_card *card) +{ + int rc = 0; + + QETH_DBF_TEXT(trace, 3, "strtcsum"); + + if (card->options.checksum_type == NO_CHECKSUMMING) { + PRINT_WARN("Using no checksumming on %s.\n", + QETH_CARD_IFNAME(card)); + return 0; + } + if (card->options.checksum_type == SW_CHECKSUMMING) { + PRINT_WARN("Using SW checksumming on %s.\n", + QETH_CARD_IFNAME(card)); + return 0; + } + if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) { + PRINT_WARN("Inbound HW Checksumming not " + "supported on %s,\ncontinuing " + "using Inbound SW Checksumming\n", + QETH_CARD_IFNAME(card)); + card->options.checksum_type = SW_CHECKSUMMING; + return 0; + } + rc = qeth_l3_send_checksum_command(card); + if (!rc) + PRINT_INFO("HW Checksumming (inbound) enabled \n"); + + return rc; +} + +static int qeth_l3_start_ipa_tso(struct qeth_card *card) +{ + int rc; + + QETH_DBF_TEXT(trace, 3, "sttso"); + + if (!qeth_is_supported(card, IPA_OUTBOUND_TSO)) { + PRINT_WARN("Outbound TSO not supported on %s\n", + QETH_CARD_IFNAME(card)); + rc = -EOPNOTSUPP; + } else { + rc = qeth_l3_send_simple_setassparms(card, IPA_OUTBOUND_TSO, + IPA_CMD_ASS_START, 0); + if (rc) + PRINT_WARN("Could not start outbound TSO " + "assist on %s: rc=%i\n", + QETH_CARD_IFNAME(card), rc); + else + PRINT_INFO("Outbound TSO enabled\n"); + } + if (rc && (card->options.large_send == QETH_LARGE_SEND_TSO)) { + card->options.large_send = QETH_LARGE_SEND_NO; + card->dev->features &= ~(NETIF_F_TSO | NETIF_F_SG); + } + return rc; +} + +static int qeth_l3_start_ipassists(struct qeth_card *card) +{ + QETH_DBF_TEXT(trace, 3, "strtipas"); + qeth_l3_start_ipa_arp_processing(card); /* go on*/ + qeth_l3_start_ipa_ip_fragmentation(card); /* go on*/ + qeth_l3_start_ipa_source_mac(card); /* go on*/ + qeth_l3_start_ipa_vlan(card); /* go on*/ + qeth_l3_start_ipa_multicast(card); /* go on*/ + qeth_l3_start_ipa_ipv6(card); /* go on*/ + qeth_l3_start_ipa_broadcast(card); /* go on*/ + qeth_l3_start_ipa_checksum(card); /* go on*/ + qeth_l3_start_ipa_tso(card); /* go on*/ + return 0; +} + +static int qeth_l3_put_unique_id(struct qeth_card *card) +{ + + int rc = 0; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(trace, 2, "puniqeid"); + + if ((card->info.unique_id & UNIQUE_ID_NOT_BY_CARD) == + UNIQUE_ID_NOT_BY_CARD) + return -1; + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_DESTROY_ADDR, + QETH_PROT_IPV6); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + *((__u16 *) &cmd->data.create_destroy_addr.unique_id[6]) = + card->info.unique_id; + memcpy(&cmd->data.create_destroy_addr.unique_id[0], + card->dev->dev_addr, OSA_ADDR_LEN); + rc = qeth_send_ipa_cmd(card, iob, NULL, NULL); + return rc; +} + +static int qeth_l3_iqd_read_initial_mac_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) +{ + struct qeth_ipa_cmd *cmd; + + cmd = (struct qeth_ipa_cmd *) data; + if (cmd->hdr.return_code == 0) + memcpy(card->dev->dev_addr, + cmd->data.create_destroy_addr.unique_id, ETH_ALEN); + else + random_ether_addr(card->dev->dev_addr); + + return 0; +} + +static int qeth_l3_iqd_read_initial_mac(struct qeth_card *card) +{ + int rc = 0; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(setup, 2, "hsrmac"); + + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_CREATE_ADDR, + QETH_PROT_IPV6); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + *((__u16 *) &cmd->data.create_destroy_addr.unique_id[6]) = + card->info.unique_id; + + rc = qeth_send_ipa_cmd(card, iob, qeth_l3_iqd_read_initial_mac_cb, + NULL); + return rc; +} + +static int qeth_l3_get_unique_id_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) +{ + struct qeth_ipa_cmd *cmd; + + cmd = (struct qeth_ipa_cmd *) data; + if (cmd->hdr.return_code == 0) + card->info.unique_id = *((__u16 *) + &cmd->data.create_destroy_addr.unique_id[6]); + else { + card->info.unique_id = UNIQUE_ID_IF_CREATE_ADDR_FAILED | + UNIQUE_ID_NOT_BY_CARD; + PRINT_WARN("couldn't get a unique id from the card on device " + "%s (result=x%x), using default id. ipv6 " + "autoconfig on other lpars may lead to duplicate " + "ip addresses. please use manually " + "configured ones.\n", + CARD_BUS_ID(card), cmd->hdr.return_code); + } + return 0; +} + +static int qeth_l3_get_unique_id(struct qeth_card *card) +{ + int rc = 0; + struct qeth_cmd_buffer *iob; + struct qeth_ipa_cmd *cmd; + + QETH_DBF_TEXT(setup, 2, "guniqeid"); + + if (!qeth_is_supported(card, IPA_IPV6)) { + card->info.unique_id = UNIQUE_ID_IF_CREATE_ADDR_FAILED | + UNIQUE_ID_NOT_BY_CARD; + return 0; + } + + iob = qeth_get_ipacmd_buffer(card, IPA_CMD_CREATE_ADDR, + QETH_PROT_IPV6); + cmd = (struct qeth_ipa_cmd *)(iob->data+IPA_PDU_HEADER_SIZE); + *((__u16 *) &cmd->data.create_destroy_addr.unique_id[6]) = + card->info.unique_id; + + rc = qeth_send_ipa_cmd(card, iob, qeth_l3_get_unique_id_cb, NULL); + return rc; +} + +static void qeth_l3_get_mac_for_ipm(__u32 ipm, char *mac, + struct net_device *dev) +{ + if (dev->type == ARPHRD_IEEE802_TR) + ip_tr_mc_map(ipm, mac); + else + ip_eth_mc_map(ipm, mac); +} + +static void qeth_l3_add_mc(struct qeth_card *card, struct in_device *in4_dev) +{ + struct qeth_ipaddr *ipm; + struct ip_mc_list *im4; + char buf[MAX_ADDR_LEN]; + + QETH_DBF_TEXT(trace, 4, "addmc"); + for (im4 = in4_dev->mc_list; im4; im4 = im4->next) { + qeth_l3_get_mac_for_ipm(im4->multiaddr, buf, in4_dev->dev); + ipm = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (!ipm) + continue; + ipm->u.a4.addr = im4->multiaddr; + memcpy(ipm->mac, buf, OSA_ADDR_LEN); + ipm->is_multicast = 1; + if (!qeth_l3_add_ip(card, ipm)) + kfree(ipm); + } +} + +static void qeth_l3_add_vlan_mc(struct qeth_card *card) +{ + struct in_device *in_dev; + struct vlan_group *vg; + int i; + + QETH_DBF_TEXT(trace, 4, "addmcvl"); + if (!qeth_is_supported(card, IPA_FULL_VLAN) || (card->vlangrp == NULL)) + return; + + vg = card->vlangrp; + for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + struct net_device *netdev = vlan_group_get_device(vg, i); + if (netdev == NULL || + !(netdev->flags & IFF_UP)) + continue; + in_dev = in_dev_get(netdev); + if (!in_dev) + continue; + read_lock(&in_dev->mc_list_lock); + qeth_l3_add_mc(card, in_dev); + read_unlock(&in_dev->mc_list_lock); + in_dev_put(in_dev); + } +} + +static void qeth_l3_add_multicast_ipv4(struct qeth_card *card) +{ + struct in_device *in4_dev; + + QETH_DBF_TEXT(trace, 4, "chkmcv4"); + in4_dev = in_dev_get(card->dev); + if (in4_dev == NULL) + return; + read_lock(&in4_dev->mc_list_lock); + qeth_l3_add_mc(card, in4_dev); + qeth_l3_add_vlan_mc(card); + read_unlock(&in4_dev->mc_list_lock); + in_dev_put(in4_dev); +} + +#ifdef CONFIG_QETH_IPV6 +static void qeth_l3_add_mc6(struct qeth_card *card, struct inet6_dev *in6_dev) +{ + struct qeth_ipaddr *ipm; + struct ifmcaddr6 *im6; + char buf[MAX_ADDR_LEN]; + + QETH_DBF_TEXT(trace, 4, "addmc6"); + for (im6 = in6_dev->mc_list; im6 != NULL; im6 = im6->next) { + ndisc_mc_map(&im6->mca_addr, buf, in6_dev->dev, 0); + ipm = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); + if (!ipm) + continue; + ipm->is_multicast = 1; + memcpy(ipm->mac, buf, OSA_ADDR_LEN); + memcpy(&ipm->u.a6.addr, &im6->mca_addr.s6_addr, + sizeof(struct in6_addr)); + if (!qeth_l3_add_ip(card, ipm)) + kfree(ipm); + } +} + +static void qeth_l3_add_vlan_mc6(struct qeth_card *card) +{ + struct inet6_dev *in_dev; + struct vlan_group *vg; + int i; + + QETH_DBF_TEXT(trace, 4, "admc6vl"); + if (!qeth_is_supported(card, IPA_FULL_VLAN) || (card->vlangrp == NULL)) + return; + + vg = card->vlangrp; + for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + struct net_device *netdev = vlan_group_get_device(vg, i); + if (netdev == NULL || + !(netdev->flags & IFF_UP)) + continue; + in_dev = in6_dev_get(netdev); + if (!in_dev) + continue; + read_lock_bh(&in_dev->lock); + qeth_l3_add_mc6(card, in_dev); + read_unlock_bh(&in_dev->lock); + in6_dev_put(in_dev); + } +} + +static void qeth_l3_add_multicast_ipv6(struct qeth_card *card) +{ + struct inet6_dev *in6_dev; + + QETH_DBF_TEXT(trace, 4, "chkmcv6"); + if (!qeth_is_supported(card, IPA_IPV6)) + return ; + in6_dev = in6_dev_get(card->dev); + if (in6_dev == NULL) + return; + read_lock_bh(&in6_dev->lock); + qeth_l3_add_mc6(card, in6_dev); + qeth_l3_add_vlan_mc6(card); + read_unlock_bh(&in6_dev->lock); + in6_dev_put(in6_dev); +} +#endif /* CONFIG_QETH_IPV6 */ + +static void qeth_l3_free_vlan_addresses4(struct qeth_card *card, + unsigned short vid) +{ + struct in_device *in_dev; + struct in_ifaddr *ifa; + struct qeth_ipaddr *addr; + + QETH_DBF_TEXT(trace, 4, "frvaddr4"); + + in_dev = in_dev_get(vlan_group_get_device(card->vlangrp, vid)); + if (!in_dev) + return; + for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) { + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (addr) { + addr->u.a4.addr = ifa->ifa_address; + addr->u.a4.mask = ifa->ifa_mask; + addr->type = QETH_IP_TYPE_NORMAL; + if (!qeth_l3_delete_ip(card, addr)) + kfree(addr); + } + } + in_dev_put(in_dev); +} + +static void qeth_l3_free_vlan_addresses6(struct qeth_card *card, + unsigned short vid) +{ +#ifdef CONFIG_QETH_IPV6 + struct inet6_dev *in6_dev; + struct inet6_ifaddr *ifa; + struct qeth_ipaddr *addr; + + QETH_DBF_TEXT(trace, 4, "frvaddr6"); + + in6_dev = in6_dev_get(vlan_group_get_device(card->vlangrp, vid)); + if (!in6_dev) + return; + for (ifa = in6_dev->addr_list; ifa; ifa = ifa->lst_next) { + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); + if (addr) { + memcpy(&addr->u.a6.addr, &ifa->addr, + sizeof(struct in6_addr)); + addr->u.a6.pfxlen = ifa->prefix_len; + addr->type = QETH_IP_TYPE_NORMAL; + if (!qeth_l3_delete_ip(card, addr)) + kfree(addr); + } + } + in6_dev_put(in6_dev); +#endif /* CONFIG_QETH_IPV6 */ +} + +static void qeth_l3_free_vlan_addresses(struct qeth_card *card, + unsigned short vid) +{ + if (!card->vlangrp) + return; + qeth_l3_free_vlan_addresses4(card, vid); + qeth_l3_free_vlan_addresses6(card, vid); +} + +static void qeth_l3_vlan_rx_register(struct net_device *dev, + struct vlan_group *grp) +{ + struct qeth_card *card = netdev_priv(dev); + unsigned long flags; + + QETH_DBF_TEXT(trace, 4, "vlanreg"); + spin_lock_irqsave(&card->vlanlock, flags); + card->vlangrp = grp; + spin_unlock_irqrestore(&card->vlanlock, flags); +} + +static void qeth_l3_vlan_rx_add_vid(struct net_device *dev, unsigned short vid) +{ + struct net_device *vlandev; + struct qeth_card *card = (struct qeth_card *) dev->priv; + struct in_device *in_dev; + + if (card->info.type == QETH_CARD_TYPE_IQD) + return; + + vlandev = vlan_group_get_device(card->vlangrp, vid); + vlandev->neigh_setup = qeth_l3_neigh_setup; + + in_dev = in_dev_get(vlandev); +#ifdef CONFIG_SYSCTL + neigh_sysctl_unregister(in_dev->arp_parms); +#endif + neigh_parms_release(&arp_tbl, in_dev->arp_parms); + + in_dev->arp_parms = neigh_parms_alloc(vlandev, &arp_tbl); +#ifdef CONFIG_SYSCTL + neigh_sysctl_register(vlandev, in_dev->arp_parms, NET_IPV4, + NET_IPV4_NEIGH, "ipv4", NULL, NULL); +#endif + in_dev_put(in_dev); + return; +} + +static void qeth_l3_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid) +{ + struct qeth_card *card = netdev_priv(dev); + unsigned long flags; + + QETH_DBF_TEXT_(trace, 4, "kid:%d", vid); + spin_lock_irqsave(&card->vlanlock, flags); + /* unregister IP addresses of vlan device */ + qeth_l3_free_vlan_addresses(card, vid); + vlan_group_set_device(card->vlangrp, vid, NULL); + spin_unlock_irqrestore(&card->vlanlock, flags); + qeth_l3_set_multicast_list(card->dev); +} + +static inline __u16 qeth_l3_rebuild_skb(struct qeth_card *card, + struct sk_buff *skb, struct qeth_hdr *hdr) +{ + unsigned short vlan_id = 0; + __be16 prot; + struct iphdr *ip_hdr; + unsigned char tg_addr[MAX_ADDR_LEN]; + + if (!(hdr->hdr.l3.flags & QETH_HDR_PASSTHRU)) { + prot = htons((hdr->hdr.l3.flags & QETH_HDR_IPV6)? ETH_P_IPV6 : + ETH_P_IP); + switch (hdr->hdr.l3.flags & QETH_HDR_CAST_MASK) { + case QETH_CAST_MULTICAST: + switch (prot) { +#ifdef CONFIG_QETH_IPV6 + case __constant_htons(ETH_P_IPV6): + ndisc_mc_map((struct in6_addr *) + skb->data + 24, + tg_addr, card->dev, 0); + break; +#endif + case __constant_htons(ETH_P_IP): + ip_hdr = (struct iphdr *)skb->data; + (card->dev->type == ARPHRD_IEEE802_TR) ? + ip_tr_mc_map(ip_hdr->daddr, tg_addr): + ip_eth_mc_map(ip_hdr->daddr, tg_addr); + break; + default: + memcpy(tg_addr, card->dev->broadcast, + card->dev->addr_len); + } + card->stats.multicast++; + skb->pkt_type = PACKET_MULTICAST; + break; + case QETH_CAST_BROADCAST: + memcpy(tg_addr, card->dev->broadcast, + card->dev->addr_len); + card->stats.multicast++; + skb->pkt_type = PACKET_BROADCAST; + break; + case QETH_CAST_UNICAST: + case QETH_CAST_ANYCAST: + case QETH_CAST_NOCAST: + default: + skb->pkt_type = PACKET_HOST; + memcpy(tg_addr, card->dev->dev_addr, + card->dev->addr_len); + } + card->dev->header_ops->create(skb, card->dev, prot, tg_addr, + "FAKELL", card->dev->addr_len); + } + +#ifdef CONFIG_TR + if (card->dev->type == ARPHRD_IEEE802_TR) + skb->protocol = tr_type_trans(skb, card->dev); + else +#endif + skb->protocol = eth_type_trans(skb, card->dev); + + if (hdr->hdr.l3.ext_flags & + (QETH_HDR_EXT_VLAN_FRAME | QETH_HDR_EXT_INCLUDE_VLAN_TAG)) { + vlan_id = (hdr->hdr.l3.ext_flags & QETH_HDR_EXT_VLAN_FRAME)? + hdr->hdr.l3.vlan_id : *((u16 *)&hdr->hdr.l3.dest_addr[12]); + } + + skb->ip_summed = card->options.checksum_type; + if (card->options.checksum_type == HW_CHECKSUMMING) { + if ((hdr->hdr.l3.ext_flags & + (QETH_HDR_EXT_CSUM_HDR_REQ | + QETH_HDR_EXT_CSUM_TRANSP_REQ)) == + (QETH_HDR_EXT_CSUM_HDR_REQ | + QETH_HDR_EXT_CSUM_TRANSP_REQ)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + else + skb->ip_summed = SW_CHECKSUMMING; + } + + return vlan_id; +} + +static void qeth_l3_process_inbound_buffer(struct qeth_card *card, + struct qeth_qdio_buffer *buf, int index) +{ + struct qdio_buffer_element *element; + struct sk_buff *skb; + struct qeth_hdr *hdr; + int offset; + __u16 vlan_tag = 0; + unsigned int len; + + /* get first element of current buffer */ + element = (struct qdio_buffer_element *)&buf->buffer->element[0]; + offset = 0; + if (card->options.performance_stats) + card->perf_stats.bufs_rec++; + while ((skb = qeth_core_get_next_skb(card, buf->buffer, &element, + &offset, &hdr))) { + skb->dev = card->dev; + /* is device UP ? */ + if (!(card->dev->flags & IFF_UP)) { + dev_kfree_skb_any(skb); + continue; + } + + switch (hdr->hdr.l3.id) { + case QETH_HEADER_TYPE_LAYER3: + vlan_tag = qeth_l3_rebuild_skb(card, skb, hdr); + len = skb->len; + if (vlan_tag) + if (card->vlangrp) + vlan_hwaccel_rx(skb, card->vlangrp, + vlan_tag); + else { + dev_kfree_skb_any(skb); + continue; + } + else + netif_rx(skb); + break; + default: + dev_kfree_skb_any(skb); + QETH_DBF_TEXT(trace, 3, "inbunkno"); + QETH_DBF_HEX(control, 3, hdr, QETH_DBF_CONTROL_LEN); + continue; + } + + card->dev->last_rx = jiffies; + card->stats.rx_packets++; + card->stats.rx_bytes += len; + } +} + +static int qeth_l3_verify_vlan_dev(struct net_device *dev, + struct qeth_card *card) +{ + int rc = 0; + struct vlan_group *vg; + int i; + + vg = card->vlangrp; + if (!vg) + return rc; + + for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + if (vlan_group_get_device(vg, i) == dev) { + rc = QETH_VLAN_CARD; + break; + } + } + + if (rc && !(netdev_priv(vlan_dev_info(dev)->real_dev) == (void *)card)) + return 0; + + return rc; +} + +static int qeth_l3_verify_dev(struct net_device *dev) +{ + struct qeth_card *card; + unsigned long flags; + int rc = 0; + + read_lock_irqsave(&qeth_core_card_list.rwlock, flags); + list_for_each_entry(card, &qeth_core_card_list.list, list) { + if (card->dev == dev) { + rc = QETH_REAL_CARD; + break; + } + rc = qeth_l3_verify_vlan_dev(dev, card); + if (rc) + break; + } + read_unlock_irqrestore(&qeth_core_card_list.rwlock, flags); + + return rc; +} + +static struct qeth_card *qeth_l3_get_card_from_dev(struct net_device *dev) +{ + struct qeth_card *card = NULL; + int rc; + + rc = qeth_l3_verify_dev(dev); + if (rc == QETH_REAL_CARD) + card = netdev_priv(dev); + else if (rc == QETH_VLAN_CARD) + card = netdev_priv(vlan_dev_info(dev)->real_dev); + if (card->options.layer2) + card = NULL; + QETH_DBF_TEXT_(trace, 4, "%d", rc); + return card ; +} + +static int qeth_l3_stop_card(struct qeth_card *card, int recovery_mode) +{ + int rc = 0; + + QETH_DBF_TEXT(setup, 2, "stopcard"); + QETH_DBF_HEX(setup, 2, &card, sizeof(void *)); + + qeth_set_allowed_threads(card, 0, 1); + if (qeth_wait_for_threads(card, ~QETH_RECOVER_THREAD)) + return -ERESTARTSYS; + if (card->read.state == CH_STATE_UP && + card->write.state == CH_STATE_UP && + (card->state == CARD_STATE_UP)) { + if (recovery_mode) + qeth_l3_stop(card->dev); + if (!card->use_hard_stop) { + rc = qeth_send_stoplan(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + } + card->state = CARD_STATE_SOFTSETUP; + } + if (card->state == CARD_STATE_SOFTSETUP) { + qeth_l3_clear_ip_list(card, !card->use_hard_stop, 1); + qeth_clear_ipacmd_list(card); + card->state = CARD_STATE_HARDSETUP; + } + if (card->state == CARD_STATE_HARDSETUP) { + if (!card->use_hard_stop && + (card->info.type != QETH_CARD_TYPE_IQD)) { + rc = qeth_l3_put_unique_id(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "2err%d", rc); + } + qeth_qdio_clear_card(card, 0); + qeth_clear_qdio_buffers(card); + qeth_clear_working_pool_list(card); + card->state = CARD_STATE_DOWN; + } + if (card->state == CARD_STATE_DOWN) { + qeth_clear_cmd_buffers(&card->read); + qeth_clear_cmd_buffers(&card->write); + } + card->use_hard_stop = 0; + return rc; +} + +static void qeth_l3_set_multicast_list(struct net_device *dev) +{ + struct qeth_card *card = netdev_priv(dev); + + QETH_DBF_TEXT(trace, 3, "setmulti"); + qeth_l3_delete_mc_addresses(card); + qeth_l3_add_multicast_ipv4(card); +#ifdef CONFIG_QETH_IPV6 + qeth_l3_add_multicast_ipv6(card); +#endif + qeth_l3_set_ip_addr_list(card); + if (!qeth_adp_supported(card, IPA_SETADP_SET_PROMISC_MODE)) + return; + qeth_setadp_promisc_mode(card); +} + +static const char *qeth_l3_arp_get_error_cause(int *rc) +{ + switch (*rc) { + case QETH_IPA_ARP_RC_FAILED: + *rc = -EIO; + return "operation failed"; + case QETH_IPA_ARP_RC_NOTSUPP: + *rc = -EOPNOTSUPP; + return "operation not supported"; + case QETH_IPA_ARP_RC_OUT_OF_RANGE: + *rc = -EINVAL; + return "argument out of range"; + case QETH_IPA_ARP_RC_Q_NOTSUPP: + *rc = -EOPNOTSUPP; + return "query operation not supported"; + case QETH_IPA_ARP_RC_Q_NO_DATA: + *rc = -ENOENT; + return "no query data available"; + default: + return "unknown error"; + } +} + +static int qeth_l3_arp_set_no_entries(struct qeth_card *card, int no_entries) +{ + int tmp; + int rc; + + QETH_DBF_TEXT(trace, 3, "arpstnoe"); + + /* + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_SET_NO_ENTRIES; + * thus we say EOPNOTSUPP for this ARP function + */ + if (card->info.guestlan) + return -EOPNOTSUPP; + if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_ARP_SET_NO_ENTRIES, + no_entries); + if (rc) { + tmp = rc; + PRINT_WARN("Could not set number of ARP entries on %s: " + "%s (0x%x/%d)\n", QETH_CARD_IFNAME(card), + qeth_l3_arp_get_error_cause(&rc), tmp, tmp); + } + return rc; +} + +static void qeth_l3_copy_arp_entries_stripped(struct qeth_arp_query_info *qinfo, + struct qeth_arp_query_data *qdata, int entry_size, + int uentry_size) +{ + char *entry_ptr; + char *uentry_ptr; + int i; + + entry_ptr = (char *)&qdata->data; + uentry_ptr = (char *)(qinfo->udata + qinfo->udata_offset); + for (i = 0; i < qdata->no_entries; ++i) { + /* strip off 32 bytes "media specific information" */ + memcpy(uentry_ptr, (entry_ptr + 32), entry_size - 32); + entry_ptr += entry_size; + uentry_ptr += uentry_size; + } +} + +static int qeth_l3_arp_query_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) +{ + struct qeth_ipa_cmd *cmd; + struct qeth_arp_query_data *qdata; + struct qeth_arp_query_info *qinfo; + int entry_size; + int uentry_size; + int i; + + QETH_DBF_TEXT(trace, 4, "arpquecb"); + + qinfo = (struct qeth_arp_query_info *) reply->param; + cmd = (struct qeth_ipa_cmd *) data; + if (cmd->hdr.return_code) { + QETH_DBF_TEXT_(trace, 4, "qaer1%i", cmd->hdr.return_code); + return 0; + } + if (cmd->data.setassparms.hdr.return_code) { + cmd->hdr.return_code = cmd->data.setassparms.hdr.return_code; + QETH_DBF_TEXT_(trace, 4, "qaer2%i", cmd->hdr.return_code); + return 0; + } + qdata = &cmd->data.setassparms.data.query_arp; + switch (qdata->reply_bits) { + case 5: + uentry_size = entry_size = sizeof(struct qeth_arp_qi_entry5); + if (qinfo->mask_bits & QETH_QARP_STRIP_ENTRIES) + uentry_size = sizeof(struct qeth_arp_qi_entry5_short); + break; + case 7: + /* fall through to default */ + default: + /* tr is the same as eth -> entry7 */ + uentry_size = entry_size = sizeof(struct qeth_arp_qi_entry7); + if (qinfo->mask_bits & QETH_QARP_STRIP_ENTRIES) + uentry_size = sizeof(struct qeth_arp_qi_entry7_short); + break; + } + /* check if there is enough room in userspace */ + if ((qinfo->udata_len - qinfo->udata_offset) < + qdata->no_entries * uentry_size){ + QETH_DBF_TEXT_(trace, 4, "qaer3%i", -ENOMEM); + cmd->hdr.return_code = -ENOMEM; + PRINT_WARN("query ARP user space buffer is too small for " + "the returned number of ARP entries. " + "Aborting query!\n"); + goto out_error; + } + QETH_DBF_TEXT_(trace, 4, "anore%i", + cmd->data.setassparms.hdr.number_of_replies); + QETH_DBF_TEXT_(trace, 4, "aseqn%i", cmd->data.setassparms.hdr.seq_no); + QETH_DBF_TEXT_(trace, 4, "anoen%i", qdata->no_entries); + + if (qinfo->mask_bits & QETH_QARP_STRIP_ENTRIES) { + /* strip off "media specific information" */ + qeth_l3_copy_arp_entries_stripped(qinfo, qdata, entry_size, + uentry_size); + } else + /*copy entries to user buffer*/ + memcpy(qinfo->udata + qinfo->udata_offset, + (char *)&qdata->data, qdata->no_entries*uentry_size); + + qinfo->no_entries += qdata->no_entries; + qinfo->udata_offset += (qdata->no_entries*uentry_size); + /* check if all replies received ... */ + if (cmd->data.setassparms.hdr.seq_no < + cmd->data.setassparms.hdr.number_of_replies) + return 1; + memcpy(qinfo->udata, &qinfo->no_entries, 4); + /* keep STRIP_ENTRIES flag so the user program can distinguish + * stripped entries from normal ones */ + if (qinfo->mask_bits & QETH_QARP_STRIP_ENTRIES) + qdata->reply_bits |= QETH_QARP_STRIP_ENTRIES; + memcpy(qinfo->udata + QETH_QARP_MASK_OFFSET, &qdata->reply_bits, 2); + return 0; +out_error: + i = 0; + memcpy(qinfo->udata, &i, 4); + return 0; +} + +static int qeth_l3_send_ipa_arp_cmd(struct qeth_card *card, + struct qeth_cmd_buffer *iob, int len, + int (*reply_cb)(struct qeth_card *, struct qeth_reply *, + unsigned long), + void *reply_param) +{ + QETH_DBF_TEXT(trace, 4, "sendarp"); + + memcpy(iob->data, IPA_PDU_HEADER, IPA_PDU_HEADER_SIZE); + memcpy(QETH_IPA_CMD_DEST_ADDR(iob->data), + &card->token.ulp_connection_r, QETH_MPC_TOKEN_LENGTH); + return qeth_send_control_data(card, IPA_PDU_HEADER_SIZE + len, iob, + reply_cb, reply_param); +} + +static int qeth_l3_arp_query(struct qeth_card *card, char __user *udata) +{ + struct qeth_cmd_buffer *iob; + struct qeth_arp_query_info qinfo = {0, }; + int tmp; + int rc; + + QETH_DBF_TEXT(trace, 3, "arpquery"); + + if (!qeth_is_supported(card,/*IPA_QUERY_ARP_ADDR_INFO*/ + IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + /* get size of userspace buffer and mask_bits -> 6 bytes */ + if (copy_from_user(&qinfo, udata, 6)) + return -EFAULT; + qinfo.udata = kzalloc(qinfo.udata_len, GFP_KERNEL); + if (!qinfo.udata) + return -ENOMEM; + qinfo.udata_offset = QETH_QARP_ENTRIES_OFFSET; + iob = qeth_l3_get_setassparms_cmd(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_ARP_QUERY_INFO, + sizeof(int), QETH_PROT_IPV4); + + rc = qeth_l3_send_ipa_arp_cmd(card, iob, + QETH_SETASS_BASE_LEN+QETH_ARP_CMD_LEN, + qeth_l3_arp_query_cb, (void *)&qinfo); + if (rc) { + tmp = rc; + PRINT_WARN("Error while querying ARP cache on %s: %s " + "(0x%x/%d)\n", QETH_CARD_IFNAME(card), + qeth_l3_arp_get_error_cause(&rc), tmp, tmp); + if (copy_to_user(udata, qinfo.udata, 4)) + rc = -EFAULT; + } else { + if (copy_to_user(udata, qinfo.udata, qinfo.udata_len)) + rc = -EFAULT; + } + kfree(qinfo.udata); + return rc; +} + +static int qeth_l3_arp_add_entry(struct qeth_card *card, + struct qeth_arp_cache_entry *entry) +{ + struct qeth_cmd_buffer *iob; + char buf[16]; + int tmp; + int rc; + + QETH_DBF_TEXT(trace, 3, "arpadent"); + + /* + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_ADD_ENTRY; + * thus we say EOPNOTSUPP for this ARP function + */ + if (card->info.guestlan) + return -EOPNOTSUPP; + if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + + iob = qeth_l3_get_setassparms_cmd(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_ARP_ADD_ENTRY, + sizeof(struct qeth_arp_cache_entry), + QETH_PROT_IPV4); + rc = qeth_l3_send_setassparms(card, iob, + sizeof(struct qeth_arp_cache_entry), + (unsigned long) entry, + qeth_l3_default_setassparms_cb, NULL); + if (rc) { + tmp = rc; + qeth_l3_ipaddr4_to_string((u8 *)entry->ipaddr, buf); + PRINT_WARN("Could not add ARP entry for address %s on %s: " + "%s (0x%x/%d)\n", + buf, QETH_CARD_IFNAME(card), + qeth_l3_arp_get_error_cause(&rc), tmp, tmp); + } + return rc; +} + +static int qeth_l3_arp_remove_entry(struct qeth_card *card, + struct qeth_arp_cache_entry *entry) +{ + struct qeth_cmd_buffer *iob; + char buf[16] = {0, }; + int tmp; + int rc; + + QETH_DBF_TEXT(trace, 3, "arprment"); + + /* + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_REMOVE_ENTRY; + * thus we say EOPNOTSUPP for this ARP function + */ + if (card->info.guestlan) + return -EOPNOTSUPP; + if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + memcpy(buf, entry, 12); + iob = qeth_l3_get_setassparms_cmd(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_ARP_REMOVE_ENTRY, + 12, + QETH_PROT_IPV4); + rc = qeth_l3_send_setassparms(card, iob, + 12, (unsigned long)buf, + qeth_l3_default_setassparms_cb, NULL); + if (rc) { + tmp = rc; + memset(buf, 0, 16); + qeth_l3_ipaddr4_to_string((u8 *)entry->ipaddr, buf); + PRINT_WARN("Could not delete ARP entry for address %s on %s: " + "%s (0x%x/%d)\n", + buf, QETH_CARD_IFNAME(card), + qeth_l3_arp_get_error_cause(&rc), tmp, tmp); + } + return rc; +} + +static int qeth_l3_arp_flush_cache(struct qeth_card *card) +{ + int rc; + int tmp; + + QETH_DBF_TEXT(trace, 3, "arpflush"); + + /* + * currently GuestLAN only supports the ARP assist function + * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_FLUSH_CACHE; + * thus we say EOPNOTSUPP for this ARP function + */ + if (card->info.guestlan || (card->info.type == QETH_CARD_TYPE_IQD)) + return -EOPNOTSUPP; + if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) { + PRINT_WARN("ARP processing not supported " + "on %s!\n", QETH_CARD_IFNAME(card)); + return -EOPNOTSUPP; + } + rc = qeth_l3_send_simple_setassparms(card, IPA_ARP_PROCESSING, + IPA_CMD_ASS_ARP_FLUSH_CACHE, 0); + if (rc) { + tmp = rc; + PRINT_WARN("Could not flush ARP cache on %s: %s (0x%x/%d)\n", + QETH_CARD_IFNAME(card), + qeth_l3_arp_get_error_cause(&rc), tmp, tmp); + } + return rc; +} + +static int qeth_l3_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) +{ + struct qeth_card *card = netdev_priv(dev); + struct qeth_arp_cache_entry arp_entry; + struct mii_ioctl_data *mii_data; + int rc = 0; + + if (!card) + return -ENODEV; + + if ((card->state != CARD_STATE_UP) && + (card->state != CARD_STATE_SOFTSETUP)) + return -ENODEV; + + switch (cmd) { + case SIOC_QETH_ARP_SET_NO_ENTRIES: + if (!capable(CAP_NET_ADMIN)) { + rc = -EPERM; + break; + } + rc = qeth_l3_arp_set_no_entries(card, rq->ifr_ifru.ifru_ivalue); + break; + case SIOC_QETH_ARP_QUERY_INFO: + if (!capable(CAP_NET_ADMIN)) { + rc = -EPERM; + break; + } + rc = qeth_l3_arp_query(card, rq->ifr_ifru.ifru_data); + break; + case SIOC_QETH_ARP_ADD_ENTRY: + if (!capable(CAP_NET_ADMIN)) { + rc = -EPERM; + break; + } + if (copy_from_user(&arp_entry, rq->ifr_ifru.ifru_data, + sizeof(struct qeth_arp_cache_entry))) + rc = -EFAULT; + else + rc = qeth_l3_arp_add_entry(card, &arp_entry); + break; + case SIOC_QETH_ARP_REMOVE_ENTRY: + if (!capable(CAP_NET_ADMIN)) { + rc = -EPERM; + break; + } + if (copy_from_user(&arp_entry, rq->ifr_ifru.ifru_data, + sizeof(struct qeth_arp_cache_entry))) + rc = -EFAULT; + else + rc = qeth_l3_arp_remove_entry(card, &arp_entry); + break; + case SIOC_QETH_ARP_FLUSH_CACHE: + if (!capable(CAP_NET_ADMIN)) { + rc = -EPERM; + break; + } + rc = qeth_l3_arp_flush_cache(card); + break; + case SIOC_QETH_ADP_SET_SNMP_CONTROL: + rc = qeth_snmp_command(card, rq->ifr_ifru.ifru_data); + break; + case SIOC_QETH_GET_CARD_TYPE: + if ((card->info.type == QETH_CARD_TYPE_OSAE) && + !card->info.guestlan) + return 1; + return 0; + break; + case SIOCGMIIPHY: + mii_data = if_mii(rq); + mii_data->phy_id = 0; + break; + case SIOCGMIIREG: + mii_data = if_mii(rq); + if (mii_data->phy_id != 0) + rc = -EINVAL; + else + mii_data->val_out = qeth_mdio_read(dev, + mii_data->phy_id, + mii_data->reg_num); + break; + default: + rc = -EOPNOTSUPP; + } + if (rc) + QETH_DBF_TEXT_(trace, 2, "ioce%d", rc); + return rc; +} + +static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr, + struct sk_buff *skb, int ipv, int cast_type) +{ + QETH_DBF_TEXT(trace, 6, "fillhdr"); + + memset(hdr, 0, sizeof(struct qeth_hdr)); + hdr->hdr.l3.id = QETH_HEADER_TYPE_LAYER3; + hdr->hdr.l3.ext_flags = 0; + + /* + * before we're going to overwrite this location with next hop ip. + * v6 uses passthrough, v4 sets the tag in the QDIO header. + */ + if (card->vlangrp && vlan_tx_tag_present(skb)) { + hdr->hdr.l3.ext_flags = (ipv == 4) ? + QETH_HDR_EXT_VLAN_FRAME : + QETH_HDR_EXT_INCLUDE_VLAN_TAG; + hdr->hdr.l3.vlan_id = vlan_tx_tag_get(skb); + } + + hdr->hdr.l3.length = skb->len - sizeof(struct qeth_hdr); + if (ipv == 4) { + /* IPv4 */ + hdr->hdr.l3.flags = qeth_l3_get_qeth_hdr_flags4(cast_type); + memset(hdr->hdr.l3.dest_addr, 0, 12); + if ((skb->dst) && (skb->dst->neighbour)) { + *((u32 *) (&hdr->hdr.l3.dest_addr[12])) = + *((u32 *) skb->dst->neighbour->primary_key); + } else { + /* fill in destination address used in ip header */ + *((u32 *) (&hdr->hdr.l3.dest_addr[12])) = + ip_hdr(skb)->daddr; + } + } else if (ipv == 6) { + /* IPv6 */ + hdr->hdr.l3.flags = qeth_l3_get_qeth_hdr_flags6(cast_type); + if (card->info.type == QETH_CARD_TYPE_IQD) + hdr->hdr.l3.flags &= ~QETH_HDR_PASSTHRU; + if ((skb->dst) && (skb->dst->neighbour)) { + memcpy(hdr->hdr.l3.dest_addr, + skb->dst->neighbour->primary_key, 16); + } else { + /* fill in destination address used in ip header */ + memcpy(hdr->hdr.l3.dest_addr, + &ipv6_hdr(skb)->daddr, 16); + } + } else { + /* passthrough */ + if ((skb->dev->type == ARPHRD_IEEE802_TR) && + !memcmp(skb->data + sizeof(struct qeth_hdr) + + sizeof(__u16), skb->dev->broadcast, 6)) { + hdr->hdr.l3.flags = QETH_CAST_BROADCAST | + QETH_HDR_PASSTHRU; + } else if (!memcmp(skb->data + sizeof(struct qeth_hdr), + skb->dev->broadcast, 6)) { + /* broadcast? */ + hdr->hdr.l3.flags = QETH_CAST_BROADCAST | + QETH_HDR_PASSTHRU; + } else { + hdr->hdr.l3.flags = (cast_type == RTN_MULTICAST) ? + QETH_CAST_MULTICAST | QETH_HDR_PASSTHRU : + QETH_CAST_UNICAST | QETH_HDR_PASSTHRU; + } + } +} + +static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + int rc; + u16 *tag; + struct qeth_hdr *hdr = NULL; + int elements_needed = 0; + struct qeth_card *card = netdev_priv(dev); + struct sk_buff *new_skb = NULL; + int ipv = qeth_get_ip_version(skb); + int cast_type = qeth_get_cast_type(card, skb); + struct qeth_qdio_out_q *queue = card->qdio.out_qs + [qeth_get_priority_queue(card, skb, ipv, cast_type)]; + int tx_bytes = skb->len; + enum qeth_large_send_types large_send = QETH_LARGE_SEND_NO; + struct qeth_eddp_context *ctx = NULL; + + QETH_DBF_TEXT(trace, 6, "l3xmit"); + + if ((card->info.type == QETH_CARD_TYPE_IQD) && + (skb->protocol != htons(ETH_P_IPV6)) && + (skb->protocol != htons(ETH_P_IP))) + goto tx_drop; + + if ((card->state != CARD_STATE_UP) || !card->lan_online) { + card->stats.tx_carrier_errors++; + goto tx_drop; + } + + if ((cast_type == RTN_BROADCAST) && + (card->info.broadcast_capable == 0)) + goto tx_drop; + + if (card->options.performance_stats) { + card->perf_stats.outbound_cnt++; + card->perf_stats.outbound_start_time = qeth_get_micros(); + } + + /* create a clone with writeable headroom */ + new_skb = skb_realloc_headroom(skb, sizeof(struct qeth_hdr_tso) + + VLAN_HLEN); + if (!new_skb) + goto tx_drop; + + if (card->info.type == QETH_CARD_TYPE_IQD) { + skb_pull(new_skb, ETH_HLEN); + } else { + if (new_skb->protocol == htons(ETH_P_IP)) { + if (card->dev->type == ARPHRD_IEEE802_TR) + skb_pull(new_skb, TR_HLEN); + else + skb_pull(new_skb, ETH_HLEN); + } + + if (new_skb->protocol == ETH_P_IPV6 && card->vlangrp && + vlan_tx_tag_present(new_skb)) { + skb_push(new_skb, VLAN_HLEN); + skb_copy_to_linear_data(new_skb, new_skb->data + 4, 4); + skb_copy_to_linear_data_offset(new_skb, 4, + new_skb->data + 8, 4); + skb_copy_to_linear_data_offset(new_skb, 8, + new_skb->data + 12, 4); + tag = (u16 *)(new_skb->data + 12); + *tag = __constant_htons(ETH_P_8021Q); + *(tag + 1) = htons(vlan_tx_tag_get(new_skb)); + VLAN_TX_SKB_CB(new_skb)->magic = 0; + } + } + + netif_stop_queue(dev); + + if (skb_is_gso(new_skb)) + large_send = card->options.large_send; + + /* fix hardware limitation: as long as we do not have sbal + * chaining we can not send long frag lists so we temporary + * switch to EDDP + */ + if ((large_send == QETH_LARGE_SEND_TSO) && + ((skb_shinfo(new_skb)->nr_frags + 2) > 16)) + large_send = QETH_LARGE_SEND_EDDP; + + if ((large_send == QETH_LARGE_SEND_TSO) && + (cast_type == RTN_UNSPEC)) { + hdr = (struct qeth_hdr *)skb_push(new_skb, + sizeof(struct qeth_hdr_tso)); + memset(hdr, 0, sizeof(struct qeth_hdr_tso)); + qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type); + qeth_tso_fill_header(card, hdr, new_skb); + elements_needed++; + } else { + hdr = (struct qeth_hdr *)skb_push(new_skb, + sizeof(struct qeth_hdr)); + qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type); + } + + if (large_send == QETH_LARGE_SEND_EDDP) { + /* new_skb is not owned by a socket so we use skb to get + * the protocol + */ + ctx = qeth_eddp_create_context(card, new_skb, hdr, + skb->sk->sk_protocol); + if (ctx == NULL) { + PRINT_WARN("could not create eddp context\n"); + goto tx_drop; + } + } else { + int elems = qeth_get_elements_no(card, (void *)hdr, new_skb, + elements_needed); + if (!elems) + goto tx_drop; + elements_needed += elems; + } + + if ((large_send == QETH_LARGE_SEND_NO) && + (new_skb->ip_summed == CHECKSUM_PARTIAL)) + qeth_tx_csum(new_skb); + + if (card->info.type != QETH_CARD_TYPE_IQD) + rc = qeth_do_send_packet(card, queue, new_skb, hdr, + elements_needed, ctx); + else + rc = qeth_do_send_packet_fast(card, queue, new_skb, hdr, + elements_needed, ctx); + + if (!rc) { + card->stats.tx_packets++; + card->stats.tx_bytes += tx_bytes; + if (new_skb != skb) + dev_kfree_skb_any(skb); + if (card->options.performance_stats) { + if (large_send != QETH_LARGE_SEND_NO) { + card->perf_stats.large_send_bytes += tx_bytes; + card->perf_stats.large_send_cnt++; + } + if (skb_shinfo(new_skb)->nr_frags > 0) { + card->perf_stats.sg_skbs_sent++; + /* nr_frags + skb->data */ + card->perf_stats.sg_frags_sent += + skb_shinfo(new_skb)->nr_frags + 1; + } + } + + if (ctx != NULL) { + qeth_eddp_put_context(ctx); + dev_kfree_skb_any(new_skb); + } + } else { + if (ctx != NULL) + qeth_eddp_put_context(ctx); + + if (rc == -EBUSY) { + if (new_skb != skb) + dev_kfree_skb_any(new_skb); + return NETDEV_TX_BUSY; + } else + goto tx_drop; + } + + netif_wake_queue(dev); + if (card->options.performance_stats) + card->perf_stats.outbound_time += qeth_get_micros() - + card->perf_stats.outbound_start_time; + return rc; + +tx_drop: + card->stats.tx_dropped++; + card->stats.tx_errors++; + if ((new_skb != skb) && new_skb) + dev_kfree_skb_any(new_skb); + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +} + +static int qeth_l3_open(struct net_device *dev) +{ + struct qeth_card *card = netdev_priv(dev); + + QETH_DBF_TEXT(trace, 4, "qethopen"); + if (card->state != CARD_STATE_SOFTSETUP) + return -ENODEV; + card->data.state = CH_STATE_UP; + card->state = CARD_STATE_UP; + card->dev->flags |= IFF_UP; + netif_start_queue(dev); + + if (!card->lan_online && netif_carrier_ok(dev)) + netif_carrier_off(dev); + return 0; +} + +static int qeth_l3_stop(struct net_device *dev) +{ + struct qeth_card *card = netdev_priv(dev); + + QETH_DBF_TEXT(trace, 4, "qethstop"); + netif_tx_disable(dev); + card->dev->flags &= ~IFF_UP; + if (card->state == CARD_STATE_UP) + card->state = CARD_STATE_SOFTSETUP; + return 0; +} + +static u32 qeth_l3_ethtool_get_rx_csum(struct net_device *dev) +{ + struct qeth_card *card = netdev_priv(dev); + + return (card->options.checksum_type == HW_CHECKSUMMING); +} + +static int qeth_l3_ethtool_set_rx_csum(struct net_device *dev, u32 data) +{ + struct qeth_card *card = netdev_priv(dev); + enum qeth_card_states old_state; + enum qeth_checksum_types csum_type; + + if ((card->state != CARD_STATE_UP) && + (card->state != CARD_STATE_DOWN)) + return -EPERM; + + if (data) + csum_type = HW_CHECKSUMMING; + else + csum_type = SW_CHECKSUMMING; + + if (card->options.checksum_type != csum_type) { + old_state = card->state; + if (card->state == CARD_STATE_UP) + __qeth_l3_set_offline(card->gdev, 1); + card->options.checksum_type = csum_type; + if (old_state == CARD_STATE_UP) + __qeth_l3_set_online(card->gdev, 1); + } + return 0; +} + +static int qeth_l3_ethtool_set_tso(struct net_device *dev, u32 data) +{ + struct qeth_card *card = netdev_priv(dev); + + if (data) { + if (card->options.large_send == QETH_LARGE_SEND_NO) { + if (card->info.type == QETH_CARD_TYPE_IQD) + card->options.large_send = QETH_LARGE_SEND_EDDP; + else + card->options.large_send = QETH_LARGE_SEND_TSO; + dev->features |= NETIF_F_TSO; + } + } else { + dev->features &= ~NETIF_F_TSO; + card->options.large_send = QETH_LARGE_SEND_NO; + } + return 0; +} + +static struct ethtool_ops qeth_l3_ethtool_ops = { + .get_link = ethtool_op_get_link, + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = ethtool_op_set_tx_hw_csum, + .get_rx_csum = qeth_l3_ethtool_get_rx_csum, + .set_rx_csum = qeth_l3_ethtool_set_rx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = ethtool_op_set_sg, + .get_tso = ethtool_op_get_tso, + .set_tso = qeth_l3_ethtool_set_tso, + .get_strings = qeth_core_get_strings, + .get_ethtool_stats = qeth_core_get_ethtool_stats, + .get_stats_count = qeth_core_get_stats_count, + .get_drvinfo = qeth_core_get_drvinfo, +}; + +/* + * we need NOARP for IPv4 but we want neighbor solicitation for IPv6. Setting + * NOARP on the netdevice is no option because it also turns off neighbor + * solicitation. For IPv4 we install a neighbor_setup function. We don't want + * arp resolution but we want the hard header (packet socket will work + * e.g. tcpdump) + */ +static int qeth_l3_neigh_setup_noarp(struct neighbour *n) +{ + n->nud_state = NUD_NOARP; + memcpy(n->ha, "FAKELL", 6); + n->output = n->ops->connected_output; + return 0; +} + +static int +qeth_l3_neigh_setup(struct net_device *dev, struct neigh_parms *np) +{ + if (np->tbl->family == AF_INET) + np->neigh_setup = qeth_l3_neigh_setup_noarp; + + return 0; +} + +static int qeth_l3_setup_netdev(struct qeth_card *card) +{ + if (card->info.type == QETH_CARD_TYPE_OSAE) { + if ((card->info.link_type == QETH_LINK_TYPE_LANE_TR) || + (card->info.link_type == QETH_LINK_TYPE_HSTR)) { +#ifdef CONFIG_TR + card->dev = alloc_trdev(0); +#endif + if (!card->dev) + return -ENODEV; + } else { + card->dev = alloc_etherdev(0); + if (!card->dev) + return -ENODEV; + card->dev->neigh_setup = qeth_l3_neigh_setup; + + /*IPv6 address autoconfiguration stuff*/ + qeth_l3_get_unique_id(card); + if (!(card->info.unique_id & UNIQUE_ID_NOT_BY_CARD)) + card->dev->dev_id = card->info.unique_id & + 0xffff; + } + } else if (card->info.type == QETH_CARD_TYPE_IQD) { + card->dev = alloc_netdev(0, "hsi%d", ether_setup); + if (!card->dev) + return -ENODEV; + card->dev->flags |= IFF_NOARP; + qeth_l3_iqd_read_initial_mac(card); + } else + return -ENODEV; + + card->dev->hard_start_xmit = qeth_l3_hard_start_xmit; + card->dev->priv = card; + card->dev->tx_timeout = &qeth_tx_timeout; + card->dev->watchdog_timeo = QETH_TX_TIMEOUT; + card->dev->open = qeth_l3_open; + card->dev->stop = qeth_l3_stop; + card->dev->do_ioctl = qeth_l3_do_ioctl; + card->dev->get_stats = qeth_get_stats; + card->dev->change_mtu = qeth_change_mtu; + card->dev->set_multicast_list = qeth_l3_set_multicast_list; + card->dev->vlan_rx_register = qeth_l3_vlan_rx_register; + card->dev->vlan_rx_add_vid = qeth_l3_vlan_rx_add_vid; + card->dev->vlan_rx_kill_vid = qeth_l3_vlan_rx_kill_vid; + card->dev->mtu = card->info.initial_mtu; + SET_ETHTOOL_OPS(card->dev, &qeth_l3_ethtool_ops); + card->dev->features |= NETIF_F_HW_VLAN_TX | + NETIF_F_HW_VLAN_RX | + NETIF_F_HW_VLAN_FILTER; + + SET_NETDEV_DEV(card->dev, &card->gdev->dev); + return register_netdev(card->dev); +} + +static void qeth_l3_qdio_input_handler(struct ccw_device *ccwdev, + unsigned int status, unsigned int qdio_err, + unsigned int siga_err, unsigned int queue, int first_element, + int count, unsigned long card_ptr) +{ + struct net_device *net_dev; + struct qeth_card *card; + struct qeth_qdio_buffer *buffer; + int index; + int i; + + QETH_DBF_TEXT(trace, 6, "qdinput"); + card = (struct qeth_card *) card_ptr; + net_dev = card->dev; + if (card->options.performance_stats) { + card->perf_stats.inbound_cnt++; + card->perf_stats.inbound_start_time = qeth_get_micros(); + } + if (status & QDIO_STATUS_LOOK_FOR_ERROR) { + if (status & QDIO_STATUS_ACTIVATE_CHECK_CONDITION) { + QETH_DBF_TEXT(trace, 1, "qdinchk"); + QETH_DBF_TEXT_(trace, 1, "%s", CARD_BUS_ID(card)); + QETH_DBF_TEXT_(trace, 1, "%04X%04X", + first_element, count); + QETH_DBF_TEXT_(trace, 1, "%04X%04X", queue, status); + qeth_schedule_recovery(card); + return; + } + } + for (i = first_element; i < (first_element + count); ++i) { + index = i % QDIO_MAX_BUFFERS_PER_Q; + buffer = &card->qdio.in_q->bufs[index]; + if (!((status & QDIO_STATUS_LOOK_FOR_ERROR) && + qeth_check_qdio_errors(buffer->buffer, + qdio_err, siga_err, "qinerr"))) + qeth_l3_process_inbound_buffer(card, buffer, index); + /* clear buffer and give back to hardware */ + qeth_put_buffer_pool_entry(card, buffer->pool_entry); + qeth_queue_input_buffer(card, index); + } + if (card->options.performance_stats) + card->perf_stats.inbound_time += qeth_get_micros() - + card->perf_stats.inbound_start_time; +} + +static int qeth_l3_probe_device(struct ccwgroup_device *gdev) +{ + struct qeth_card *card = dev_get_drvdata(&gdev->dev); + + qeth_l3_create_device_attributes(&gdev->dev); + card->options.layer2 = 0; + card->discipline.input_handler = (qdio_handler_t *) + qeth_l3_qdio_input_handler; + card->discipline.output_handler = (qdio_handler_t *) + qeth_qdio_output_handler; + card->discipline.recover = qeth_l3_recover; + return 0; +} + +static void qeth_l3_remove_device(struct ccwgroup_device *cgdev) +{ + struct qeth_card *card = dev_get_drvdata(&cgdev->dev); + + wait_event(card->wait_q, qeth_threads_running(card, 0xffffffff) == 0); + + if (cgdev->state == CCWGROUP_ONLINE) { + card->use_hard_stop = 1; + qeth_l3_set_offline(cgdev); + } + + if (card->dev) { + unregister_netdev(card->dev); + card->dev = NULL; + } + + qeth_l3_remove_device_attributes(&cgdev->dev); + qeth_l3_clear_ip_list(card, 0, 0); + qeth_l3_clear_ipato_list(card); + return; +} + +static int __qeth_l3_set_online(struct ccwgroup_device *gdev, int recovery_mode) +{ + struct qeth_card *card = dev_get_drvdata(&gdev->dev); + int rc = 0; + enum qeth_card_states recover_flag; + + BUG_ON(!card); + QETH_DBF_TEXT(setup, 2, "setonlin"); + QETH_DBF_HEX(setup, 2, &card, sizeof(void *)); + + qeth_set_allowed_threads(card, QETH_RECOVER_THREAD, 1); + if (qeth_wait_for_threads(card, ~QETH_RECOVER_THREAD)) { + PRINT_WARN("set_online of card %s interrupted by user!\n", + CARD_BUS_ID(card)); + return -ERESTARTSYS; + } + + recover_flag = card->state; + rc = ccw_device_set_online(CARD_RDEV(card)); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + return -EIO; + } + rc = ccw_device_set_online(CARD_WDEV(card)); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + return -EIO; + } + rc = ccw_device_set_online(CARD_DDEV(card)); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + return -EIO; + } + + rc = qeth_core_hardsetup_card(card); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "2err%d", rc); + goto out_remove; + } + + qeth_l3_query_ipassists(card, QETH_PROT_IPV4); + + if (!card->dev && qeth_l3_setup_netdev(card)) + goto out_remove; + + card->state = CARD_STATE_HARDSETUP; + qeth_print_status_message(card); + + /* softsetup */ + QETH_DBF_TEXT(setup, 2, "softsetp"); + + rc = qeth_send_startlan(card); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + if (rc == 0xe080) { + PRINT_WARN("LAN on card %s if offline! " + "Waiting for STARTLAN from card.\n", + CARD_BUS_ID(card)); + card->lan_online = 0; + } + return rc; + } else + card->lan_online = 1; + qeth_set_large_send(card, card->options.large_send); + + rc = qeth_l3_setadapter_parms(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "2err%d", rc); + rc = qeth_l3_start_ipassists(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "3err%d", rc); + rc = qeth_l3_setrouting_v4(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "4err%d", rc); + rc = qeth_l3_setrouting_v6(card); + if (rc) + QETH_DBF_TEXT_(setup, 2, "5err%d", rc); + netif_tx_disable(card->dev); + + rc = qeth_init_qdio_queues(card); + if (rc) { + QETH_DBF_TEXT_(setup, 2, "6err%d", rc); + goto out_remove; + } + card->state = CARD_STATE_SOFTSETUP; + netif_carrier_on(card->dev); + + qeth_set_allowed_threads(card, 0xffffffff, 0); + if ((recover_flag == CARD_STATE_RECOVER) && recovery_mode) { + qeth_l3_open(card->dev); + qeth_l3_set_multicast_list(card->dev); + } + /* let user_space know that device is online */ + kobject_uevent(&gdev->dev.kobj, KOBJ_CHANGE); + return 0; +out_remove: + card->use_hard_stop = 1; + qeth_l3_stop_card(card, 0); + ccw_device_set_offline(CARD_DDEV(card)); + ccw_device_set_offline(CARD_WDEV(card)); + ccw_device_set_offline(CARD_RDEV(card)); + if (recover_flag == CARD_STATE_RECOVER) + card->state = CARD_STATE_RECOVER; + else + card->state = CARD_STATE_DOWN; + return -ENODEV; +} + +static int qeth_l3_set_online(struct ccwgroup_device *gdev) +{ + return __qeth_l3_set_online(gdev, 0); +} + +static int __qeth_l3_set_offline(struct ccwgroup_device *cgdev, + int recovery_mode) +{ + struct qeth_card *card = dev_get_drvdata(&cgdev->dev); + int rc = 0, rc2 = 0, rc3 = 0; + enum qeth_card_states recover_flag; + + QETH_DBF_TEXT(setup, 3, "setoffl"); + QETH_DBF_HEX(setup, 3, &card, sizeof(void *)); + + if (card->dev && netif_carrier_ok(card->dev)) + netif_carrier_off(card->dev); + recover_flag = card->state; + if (qeth_l3_stop_card(card, recovery_mode) == -ERESTARTSYS) { + PRINT_WARN("Stopping card %s interrupted by user!\n", + CARD_BUS_ID(card)); + return -ERESTARTSYS; + } + rc = ccw_device_set_offline(CARD_DDEV(card)); + rc2 = ccw_device_set_offline(CARD_WDEV(card)); + rc3 = ccw_device_set_offline(CARD_RDEV(card)); + if (!rc) + rc = (rc2) ? rc2 : rc3; + if (rc) + QETH_DBF_TEXT_(setup, 2, "1err%d", rc); + if (recover_flag == CARD_STATE_UP) + card->state = CARD_STATE_RECOVER; + /* let user_space know that device is offline */ + kobject_uevent(&cgdev->dev.kobj, KOBJ_CHANGE); + return 0; +} + +static int qeth_l3_set_offline(struct ccwgroup_device *cgdev) +{ + return __qeth_l3_set_offline(cgdev, 0); +} + +static int qeth_l3_recover(void *ptr) +{ + struct qeth_card *card; + int rc = 0; + + card = (struct qeth_card *) ptr; + QETH_DBF_TEXT(trace, 2, "recover1"); + QETH_DBF_HEX(trace, 2, &card, sizeof(void *)); + if (!qeth_do_run_thread(card, QETH_RECOVER_THREAD)) + return 0; + QETH_DBF_TEXT(trace, 2, "recover2"); + PRINT_WARN("Recovery of device %s started ...\n", + CARD_BUS_ID(card)); + card->use_hard_stop = 1; + __qeth_l3_set_offline(card->gdev, 1); + rc = __qeth_l3_set_online(card->gdev, 1); + /* don't run another scheduled recovery */ + qeth_clear_thread_start_bit(card, QETH_RECOVER_THREAD); + qeth_clear_thread_running_bit(card, QETH_RECOVER_THREAD); + if (!rc) + PRINT_INFO("Device %s successfully recovered!\n", + CARD_BUS_ID(card)); + else + PRINT_INFO("Device %s could not be recovered!\n", + CARD_BUS_ID(card)); + return 0; +} + +static void qeth_l3_shutdown(struct ccwgroup_device *gdev) +{ + struct qeth_card *card = dev_get_drvdata(&gdev->dev); + qeth_l3_clear_ip_list(card, 0, 0); + qeth_qdio_clear_card(card, 0); + qeth_clear_qdio_buffers(card); +} + +struct ccwgroup_driver qeth_l3_ccwgroup_driver = { + .probe = qeth_l3_probe_device, + .remove = qeth_l3_remove_device, + .set_online = qeth_l3_set_online, + .set_offline = qeth_l3_set_offline, + .shutdown = qeth_l3_shutdown, +}; +EXPORT_SYMBOL_GPL(qeth_l3_ccwgroup_driver); + +static int qeth_l3_ip_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; + struct net_device *dev = (struct net_device *)ifa->ifa_dev->dev; + struct qeth_ipaddr *addr; + struct qeth_card *card; + + if (dev->nd_net != &init_net) + return NOTIFY_DONE; + + QETH_DBF_TEXT(trace, 3, "ipevent"); + card = qeth_l3_get_card_from_dev(dev); + if (!card) + return NOTIFY_DONE; + + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (addr != NULL) { + addr->u.a4.addr = ifa->ifa_address; + addr->u.a4.mask = ifa->ifa_mask; + addr->type = QETH_IP_TYPE_NORMAL; + } else + goto out; + + switch (event) { + case NETDEV_UP: + if (!qeth_l3_add_ip(card, addr)) + kfree(addr); + break; + case NETDEV_DOWN: + if (!qeth_l3_delete_ip(card, addr)) + kfree(addr); + break; + default: + break; + } + qeth_l3_set_ip_addr_list(card); +out: + return NOTIFY_DONE; +} + +static struct notifier_block qeth_l3_ip_notifier = { + qeth_l3_ip_event, + NULL, +}; + +#ifdef CONFIG_QETH_IPV6 +/** + * IPv6 event handler + */ +static int qeth_l3_ip6_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct inet6_ifaddr *ifa = (struct inet6_ifaddr *)ptr; + struct net_device *dev = (struct net_device *)ifa->idev->dev; + struct qeth_ipaddr *addr; + struct qeth_card *card; + + QETH_DBF_TEXT(trace, 3, "ip6event"); + + card = qeth_l3_get_card_from_dev(dev); + if (!card) + return NOTIFY_DONE; + if (!qeth_is_supported(card, IPA_IPV6)) + return NOTIFY_DONE; + + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); + if (addr != NULL) { + memcpy(&addr->u.a6.addr, &ifa->addr, sizeof(struct in6_addr)); + addr->u.a6.pfxlen = ifa->prefix_len; + addr->type = QETH_IP_TYPE_NORMAL; + } else + goto out; + + switch (event) { + case NETDEV_UP: + if (!qeth_l3_add_ip(card, addr)) + kfree(addr); + break; + case NETDEV_DOWN: + if (!qeth_l3_delete_ip(card, addr)) + kfree(addr); + break; + default: + break; + } + qeth_l3_set_ip_addr_list(card); +out: + return NOTIFY_DONE; +} + +static struct notifier_block qeth_l3_ip6_notifier = { + qeth_l3_ip6_event, + NULL, +}; +#endif + +static int qeth_l3_register_notifiers(void) +{ + int rc; + + QETH_DBF_TEXT(trace, 5, "regnotif"); + rc = register_inetaddr_notifier(&qeth_l3_ip_notifier); + if (rc) + return rc; +#ifdef CONFIG_QETH_IPV6 + rc = register_inet6addr_notifier(&qeth_l3_ip6_notifier); + if (rc) { + unregister_inetaddr_notifier(&qeth_l3_ip_notifier); + return rc; + } +#else + PRINT_WARN("layer 3 discipline no IPv6 support\n"); +#endif + return 0; +} + +static void qeth_l3_unregister_notifiers(void) +{ + + QETH_DBF_TEXT(trace, 5, "unregnot"); + BUG_ON(unregister_inetaddr_notifier(&qeth_l3_ip_notifier)); +#ifdef CONFIG_QETH_IPV6 + BUG_ON(unregister_inet6addr_notifier(&qeth_l3_ip6_notifier)); +#endif /* QETH_IPV6 */ +} + +static int __init qeth_l3_init(void) +{ + int rc = 0; + + PRINT_INFO("register layer 3 discipline\n"); + rc = qeth_l3_register_notifiers(); + return rc; +} + +static void __exit qeth_l3_exit(void) +{ + qeth_l3_unregister_notifiers(); + PRINT_INFO("unregister layer 3 discipline\n"); +} + +module_init(qeth_l3_init); +module_exit(qeth_l3_exit); +MODULE_AUTHOR("Frank Blaschka <frank.blaschka@de.ibm.com>"); +MODULE_DESCRIPTION("qeth layer 3 discipline"); +MODULE_LICENSE("GPL"); |