linux_dsm_epyc7002/drivers/s390/net/qeth_l3.h
Julian Wiedmann b1d5e36b41 s390/qeth: shrink qeth_ipaddr struct
Using up 8 bytes in every ipaddr object to store SETIP/DELIP flags is
rather wasteful. Except for takeover eligibility, the flag values all
just depend on the address type, so determine them on demand.

While at it reorder the struct to fill an alignment hole.

Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-09 13:10:05 -05:00

134 lines
3.4 KiB
C

/* SPDX-License-Identifier: GPL-2.0 */
/*
* 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>
*/
#ifndef __QETH_L3_H__
#define __QETH_L3_H__
#include "qeth_core.h"
#include <linux/hashtable.h>
#define QETH_SNIFF_AVAIL 0x0008
enum qeth_ip_types {
QETH_IP_TYPE_NORMAL,
QETH_IP_TYPE_VIPA,
QETH_IP_TYPE_RXIP,
};
struct qeth_ipaddr {
struct hlist_node hnode;
enum qeth_ip_types type;
unsigned char mac[ETH_ALEN];
u8 is_multicast:1;
u8 in_progress:1;
u8 disp_flag:2;
u8 ipato:1; /* ucast only */
/* is changed only for normal ip addresses
* for non-normal addresses it always is 1
*/
int ref_counter;
enum qeth_prot_versions proto;
union {
struct {
unsigned int addr;
unsigned int mask;
} a4;
struct {
struct in6_addr addr;
unsigned int pfxlen;
} a6;
} u;
};
static inline void qeth_l3_init_ipaddr(struct qeth_ipaddr *addr,
enum qeth_ip_types type,
enum qeth_prot_versions proto)
{
memset(addr, 0, sizeof(*addr));
addr->type = type;
addr->proto = proto;
addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING;
}
static inline bool qeth_l3_addr_match_ip(struct qeth_ipaddr *a1,
struct qeth_ipaddr *a2)
{
if (a1->proto != a2->proto)
return false;
if (a1->proto == QETH_PROT_IPV6)
return ipv6_addr_equal(&a1->u.a6.addr, &a2->u.a6.addr);
return a1->u.a4.addr == a2->u.a4.addr;
}
static inline bool qeth_l3_addr_match_all(struct qeth_ipaddr *a1,
struct qeth_ipaddr *a2)
{
/* Assumes that the pair was obtained via qeth_l3_addr_find_by_ip(),
* so 'proto' and 'addr' match for sure.
*
* For ucast:
* - 'mac' is always 0.
* - 'mask'/'pfxlen' for RXIP/VIPA is always 0. For NORMAL, matching
* values are required to avoid mixups in takeover eligibility.
*
* For mcast,
* - 'mac' is mapped from the IP, and thus always matches.
* - 'mask'/'pfxlen' is always 0.
*/
if (a1->type != a2->type)
return false;
if (a1->proto == QETH_PROT_IPV6)
return a1->u.a6.pfxlen == a2->u.a6.pfxlen;
return a1->u.a4.mask == a2->u.a4.mask;
}
static inline u64 qeth_l3_ipaddr_hash(struct qeth_ipaddr *addr)
{
u64 ret = 0;
u8 *point;
if (addr->proto == QETH_PROT_IPV6) {
point = (u8 *) &addr->u.a6.addr;
ret = get_unaligned((u64 *)point) ^
get_unaligned((u64 *) (point + 8));
}
if (addr->proto == QETH_PROT_IPV4) {
point = (u8 *) &addr->u.a4.addr;
ret = get_unaligned((u32 *) point);
}
return ret;
}
struct qeth_ipato_entry {
struct list_head entry;
enum qeth_prot_versions proto;
char addr[16];
int mask_bits;
};
extern const struct attribute_group *qeth_l3_attr_groups[];
void qeth_l3_ipaddr_to_string(enum qeth_prot_versions, const __u8 *, char *);
int qeth_l3_create_device_attributes(struct device *);
void qeth_l3_remove_device_attributes(struct device *);
int qeth_l3_setrouting_v4(struct qeth_card *);
int qeth_l3_setrouting_v6(struct qeth_card *);
int qeth_l3_add_ipato_entry(struct qeth_card *, struct qeth_ipato_entry *);
int qeth_l3_del_ipato_entry(struct qeth_card *card,
enum qeth_prot_versions proto, u8 *addr,
int mask_bits);
void qeth_l3_update_ipato(struct qeth_card *card);
int qeth_l3_modify_hsuid(struct qeth_card *card, bool add);
int qeth_l3_modify_rxip_vipa(struct qeth_card *card, bool add, const u8 *ip,
enum qeth_ip_types type,
enum qeth_prot_versions proto);
#endif /* __QETH_L3_H__ */