Initial commit: Linux kernel 4.19 for Unplugged OS

This commit is contained in:
ziv committed 2025-10-06 13:52:42 +00:00
commit d30235b04c
80519 files changed
+35800681

No files matched your search

+106
View File
@@ -0,0 +1,106 @@
/*
* Copyright (c) 2010 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(_RDMA_IB_H)
#define _RDMA_IB_H
#include <linux/types.h>
#include <linux/sched.h>
#include <linux/cred.h>
struct ib_addr {
union {
__u8 uib_addr8[16];
__be16 uib_addr16[8];
__be32 uib_addr32[4];
__be64 uib_addr64[2];
} ib_u;
#define sib_addr8 ib_u.uib_addr8
#define sib_addr16 ib_u.uib_addr16
#define sib_addr32 ib_u.uib_addr32
#define sib_addr64 ib_u.uib_addr64
#define sib_raw ib_u.uib_addr8
#define sib_subnet_prefix ib_u.uib_addr64[0]
#define sib_interface_id ib_u.uib_addr64[1]
};
static inline bool ib_addr_any(const struct ib_addr *a)
{
return ((a->sib_addr64[0] | a->sib_addr64[1]) == 0);
}
static inline bool ib_addr_loopback(const struct ib_addr *a)
{
return ((a->sib_addr32[0] | a->sib_addr32[1] |
a->sib_addr32[2] | (a->sib_addr32[3] ^ htonl(1))) == 0);
}
static inline void ib_addr_set(struct ib_addr *addr,
__be32 w1, __be32 w2, __be32 w3, __be32 w4)
{
addr->sib_addr32[0] = w1;
addr->sib_addr32[1] = w2;
addr->sib_addr32[2] = w3;
addr->sib_addr32[3] = w4;
}
static inline int ib_addr_cmp(const struct ib_addr *a1, const struct ib_addr *a2)
{
return memcmp(a1, a2, sizeof(struct ib_addr));
}
struct sockaddr_ib {
unsigned short int sib_family; /* AF_IB */
__be16 sib_pkey;
__be32 sib_flowinfo;
struct ib_addr sib_addr;
__be64 sib_sid;
__be64 sib_sid_mask;
__u64 sib_scope_id;
};
/*
* The IB interfaces that use write() as bi-directional ioctl() are
* fundamentally unsafe, since there are lots of ways to trigger "write()"
* calls from various contexts with elevated privileges. That includes the
* traditional suid executable error message writes, but also various kernel
* interfaces that can write to file descriptors.
*
* This function provides protection for the legacy API by restricting the
* calling context.
*/
static inline bool ib_safe_file_access(struct file *filp)
{
return filp->f_cred == current_cred() && !uaccess_kernel();
}
#endif /* _RDMA_IB_H */
+299
View File
@@ -0,0 +1,299 @@
/*
* Copyright (c) 2005 Voltaire Inc. All rights reserved.
* Copyright (c) 2005 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_ADDR_H)
#define IB_ADDR_H
#include <linux/in.h>
#include <linux/in6.h>
#include <linux/if_arp.h>
#include <linux/netdevice.h>
#include <linux/inetdevice.h>
#include <linux/socket.h>
#include <linux/if_vlan.h>
#include <net/ipv6.h>
#include <net/if_inet6.h>
#include <net/ip.h>
#include <rdma/ib_verbs.h>
#include <rdma/ib_pack.h>
#include <net/ipv6.h>
#include <net/net_namespace.h>
/**
* struct rdma_dev_addr - Contains resolved RDMA hardware addresses
* @src_dev_addr: Source MAC address.
* @dst_dev_addr: Destination MAC address.
* @broadcast: Broadcast address of the device.
* @dev_type: The interface hardware type of the device.
* @bound_dev_if: An optional device interface index.
* @transport: The transport type used.
* @net: Network namespace containing the bound_dev_if net_dev.
* @sgid_attr: GID attribute to use for identified SGID
*/
struct rdma_dev_addr {
unsigned char src_dev_addr[MAX_ADDR_LEN];
unsigned char dst_dev_addr[MAX_ADDR_LEN];
unsigned char broadcast[MAX_ADDR_LEN];
unsigned short dev_type;
int bound_dev_if;
enum rdma_transport_type transport;
struct net *net;
const struct ib_gid_attr *sgid_attr;
enum rdma_network_type network;
int hoplimit;
};
/**
* rdma_translate_ip - Translate a local IP address to an RDMA hardware
* address.
*
* The dev_addr->net field must be initialized.
*/
int rdma_translate_ip(const struct sockaddr *addr,
struct rdma_dev_addr *dev_addr);
/**
* rdma_resolve_ip - Resolve source and destination IP addresses to
* RDMA hardware addresses.
* @src_addr: An optional source address to use in the resolution. If a
* source address is not provided, a usable address will be returned via
* the callback.
* @dst_addr: The destination address to resolve.
* @addr: A reference to a data location that will receive the resolved
* addresses. The data location must remain valid until the callback has
* been invoked. The net field of the addr struct must be valid.
* @timeout_ms: Amount of time to wait for the address resolution to complete.
* @callback: Call invoked once address resolution has completed, timed out,
* or been canceled. A status of 0 indicates success.
* @context: User-specified context associated with the call.
*/
int rdma_resolve_ip(struct sockaddr *src_addr, const struct sockaddr *dst_addr,
struct rdma_dev_addr *addr, int timeout_ms,
void (*callback)(int status, struct sockaddr *src_addr,
struct rdma_dev_addr *addr, void *context),
void *context);
void rdma_addr_cancel(struct rdma_dev_addr *addr);
void rdma_copy_addr(struct rdma_dev_addr *dev_addr,
const struct net_device *dev,
const unsigned char *dst_dev_addr);
int rdma_addr_size(const struct sockaddr *addr);
int rdma_addr_size_in6(struct sockaddr_in6 *addr);
int rdma_addr_size_kss(struct __kernel_sockaddr_storage *addr);
static inline u16 ib_addr_get_pkey(struct rdma_dev_addr *dev_addr)
{
return ((u16)dev_addr->broadcast[8] << 8) | (u16)dev_addr->broadcast[9];
}
static inline void ib_addr_set_pkey(struct rdma_dev_addr *dev_addr, u16 pkey)
{
dev_addr->broadcast[8] = pkey >> 8;
dev_addr->broadcast[9] = (unsigned char) pkey;
}
static inline void ib_addr_get_mgid(struct rdma_dev_addr *dev_addr,
union ib_gid *gid)
{
memcpy(gid, dev_addr->broadcast + 4, sizeof *gid);
}
static inline int rdma_addr_gid_offset(struct rdma_dev_addr *dev_addr)
{
return dev_addr->dev_type == ARPHRD_INFINIBAND ? 4 : 0;
}
static inline u16 rdma_vlan_dev_vlan_id(const struct net_device *dev)
{
return is_vlan_dev(dev) ? vlan_dev_vlan_id(dev) : 0xffff;
}
static inline int rdma_ip2gid(struct sockaddr *addr, union ib_gid *gid)
{
switch (addr->sa_family) {
case AF_INET:
ipv6_addr_set_v4mapped(((struct sockaddr_in *)
addr)->sin_addr.s_addr,
(struct in6_addr *)gid);
break;
case AF_INET6:
*(struct in6_addr *)&gid->raw =
((struct sockaddr_in6 *)addr)->sin6_addr;
break;
default:
return -EINVAL;
}
return 0;
}
/* Important - sockaddr should be a union of sockaddr_in and sockaddr_in6 */
static inline void rdma_gid2ip(struct sockaddr *out, const union ib_gid *gid)
{
if (ipv6_addr_v4mapped((struct in6_addr *)gid)) {
struct sockaddr_in *out_in = (struct sockaddr_in *)out;
memset(out_in, 0, sizeof(*out_in));
out_in->sin_family = AF_INET;
memcpy(&out_in->sin_addr.s_addr, gid->raw + 12, 4);
} else {
struct sockaddr_in6 *out_in = (struct sockaddr_in6 *)out;
memset(out_in, 0, sizeof(*out_in));
out_in->sin6_family = AF_INET6;
memcpy(&out_in->sin6_addr.s6_addr, gid->raw, 16);
}
}
/*
* rdma_get/set_sgid/dgid() APIs are applicable to IB, and iWarp.
* They are not applicable to RoCE.
* RoCE GIDs are derived from the IP addresses.
*/
static inline void rdma_addr_get_sgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
{
memcpy(gid, dev_addr->src_dev_addr + rdma_addr_gid_offset(dev_addr),
sizeof(*gid));
}
static inline void rdma_addr_set_sgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
{
memcpy(dev_addr->src_dev_addr + rdma_addr_gid_offset(dev_addr), gid, sizeof *gid);
}
static inline void rdma_addr_get_dgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
{
memcpy(gid, dev_addr->dst_dev_addr + rdma_addr_gid_offset(dev_addr), sizeof *gid);
}
static inline void rdma_addr_set_dgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
{
memcpy(dev_addr->dst_dev_addr + rdma_addr_gid_offset(dev_addr), gid, sizeof *gid);
}
static inline enum ib_mtu iboe_get_mtu(int mtu)
{
/*
* Reduce IB headers from effective IBoE MTU.
*/
mtu = mtu - (IB_GRH_BYTES + IB_UDP_BYTES + IB_BTH_BYTES +
IB_EXT_XRC_BYTES + IB_EXT_ATOMICETH_BYTES +
IB_ICRC_BYTES);
if (mtu >= ib_mtu_enum_to_int(IB_MTU_4096))
return IB_MTU_4096;
else if (mtu >= ib_mtu_enum_to_int(IB_MTU_2048))
return IB_MTU_2048;
else if (mtu >= ib_mtu_enum_to_int(IB_MTU_1024))
return IB_MTU_1024;
else if (mtu >= ib_mtu_enum_to_int(IB_MTU_512))
return IB_MTU_512;
else if (mtu >= ib_mtu_enum_to_int(IB_MTU_256))
return IB_MTU_256;
else
return 0;
}
static inline int iboe_get_rate(struct net_device *dev)
{
struct ethtool_link_ksettings cmd;
int err;
rtnl_lock();
err = __ethtool_get_link_ksettings(dev, &cmd);
rtnl_unlock();
if (err)
return IB_RATE_PORT_CURRENT;
if (cmd.base.speed >= 40000)
return IB_RATE_40_GBPS;
else if (cmd.base.speed >= 30000)
return IB_RATE_30_GBPS;
else if (cmd.base.speed >= 20000)
return IB_RATE_20_GBPS;
else if (cmd.base.speed >= 10000)
return IB_RATE_10_GBPS;
else
return IB_RATE_PORT_CURRENT;
}
static inline int rdma_link_local_addr(struct in6_addr *addr)
{
if (addr->s6_addr32[0] == htonl(0xfe800000) &&
addr->s6_addr32[1] == 0)
return 1;
return 0;
}
static inline void rdma_get_ll_mac(struct in6_addr *addr, u8 *mac)
{
memcpy(mac, &addr->s6_addr[8], 3);
memcpy(mac + 3, &addr->s6_addr[13], 3);
mac[0] ^= 2;
}
static inline int rdma_is_multicast_addr(struct in6_addr *addr)
{
__be32 ipv4_addr;
if (addr->s6_addr[0] == 0xff)
return 1;
ipv4_addr = addr->s6_addr32[3];
return (ipv6_addr_v4mapped(addr) && ipv4_is_multicast(ipv4_addr));
}
static inline void rdma_get_mcast_mac(struct in6_addr *addr, u8 *mac)
{
int i;
mac[0] = 0x33;
mac[1] = 0x33;
for (i = 2; i < 6; ++i)
mac[i] = addr->s6_addr[i + 10];
}
static inline u16 rdma_get_vlan_id(union ib_gid *dgid)
{
u16 vid;
vid = dgid->raw[11] << 8 | dgid->raw[12];
return vid < 0x1000 ? vid : 0xffff;
}
static inline struct net_device *rdma_vlan_dev_real_dev(const struct net_device *dev)
{
return is_vlan_dev(dev) ? vlan_dev_real_dev(dev) : NULL;
}
#endif /* IB_ADDR_H */
+136
View File
@@ -0,0 +1,136 @@
/*
* Copyright (c) 2004 Topspin Communications. All rights reserved.
* Copyright (c) 2005 Intel Corporation. All rights reserved.
* Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _IB_CACHE_H
#define _IB_CACHE_H
#include <rdma/ib_verbs.h>
int rdma_query_gid(struct ib_device *device, u8 port_num, int index,
union ib_gid *gid);
const struct ib_gid_attr *rdma_find_gid(struct ib_device *device,
const union ib_gid *gid,
enum ib_gid_type gid_type,
struct net_device *ndev);
const struct ib_gid_attr *rdma_find_gid_by_port(struct ib_device *ib_dev,
const union ib_gid *gid,
enum ib_gid_type gid_type,
u8 port,
struct net_device *ndev);
const struct ib_gid_attr *rdma_find_gid_by_filter(
struct ib_device *device, const union ib_gid *gid, u8 port_num,
bool (*filter)(const union ib_gid *gid, const struct ib_gid_attr *,
void *),
void *context);
/**
* ib_get_cached_pkey - Returns a cached PKey table entry
* @device: The device to query.
* @port_num: The port number of the device to query.
* @index: The index into the cached PKey table to query.
* @pkey: The PKey value found at the specified index.
*
* ib_get_cached_pkey() fetches the specified PKey table entry stored in
* the local software cache.
*/
int ib_get_cached_pkey(struct ib_device *device_handle,
u8 port_num,
int index,
u16 *pkey);
/**
* ib_find_cached_pkey - Returns the PKey table index where a specified
* PKey value occurs.
* @device: The device to query.
* @port_num: The port number of the device to search for the PKey.
* @pkey: The PKey value to search for.
* @index: The index into the cached PKey table where the PKey was found.
*
* ib_find_cached_pkey() searches the specified PKey table in
* the local software cache.
*/
int ib_find_cached_pkey(struct ib_device *device,
u8 port_num,
u16 pkey,
u16 *index);
/**
* ib_find_exact_cached_pkey - Returns the PKey table index where a specified
* PKey value occurs. Comparison uses the FULL 16 bits (incl membership bit)
* @device: The device to query.
* @port_num: The port number of the device to search for the PKey.
* @pkey: The PKey value to search for.
* @index: The index into the cached PKey table where the PKey was found.
*
* ib_find_exact_cached_pkey() searches the specified PKey table in
* the local software cache.
*/
int ib_find_exact_cached_pkey(struct ib_device *device,
u8 port_num,
u16 pkey,
u16 *index);
/**
* ib_get_cached_lmc - Returns a cached lmc table entry
* @device: The device to query.
* @port_num: The port number of the device to query.
* @lmc: The lmc value for the specified port for that device.
*
* ib_get_cached_lmc() fetches the specified lmc table entry stored in
* the local software cache.
*/
int ib_get_cached_lmc(struct ib_device *device,
u8 port_num,
u8 *lmc);
/**
* ib_get_cached_port_state - Returns a cached port state table entry
* @device: The device to query.
* @port_num: The port number of the device to query.
* @port_state: port_state for the specified port for that device.
*
* ib_get_cached_port_state() fetches the specified port_state table entry stored in
* the local software cache.
*/
int ib_get_cached_port_state(struct ib_device *device,
u8 port_num,
enum ib_port_state *port_active);
bool rdma_is_zero_gid(const union ib_gid *gid);
const struct ib_gid_attr *rdma_get_gid_attr(struct ib_device *device,
u8 port_num, int index);
void rdma_put_gid_attr(const struct ib_gid_attr *attr);
void rdma_hold_gid_attr(const struct ib_gid_attr *attr);
#endif /* _IB_CACHE_H */
+628
View File
@@ -0,0 +1,628 @@
/*
* Copyright (c) 2004, 2005 Intel Corporation. All rights reserved.
* Copyright (c) 2004 Topspin Corporation. All rights reserved.
* Copyright (c) 2004 Voltaire Corporation. All rights reserved.
* Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_CM_H)
#define IB_CM_H
#include <rdma/ib_mad.h>
#include <rdma/ib_sa.h>
/* ib_cm and ib_user_cm modules share /sys/class/infiniband_cm */
extern struct class cm_class;
enum ib_cm_state {
IB_CM_IDLE,
IB_CM_LISTEN,
IB_CM_REQ_SENT,
IB_CM_REQ_RCVD,
IB_CM_MRA_REQ_SENT,
IB_CM_MRA_REQ_RCVD,
IB_CM_REP_SENT,
IB_CM_REP_RCVD,
IB_CM_MRA_REP_SENT,
IB_CM_MRA_REP_RCVD,
IB_CM_ESTABLISHED,
IB_CM_DREQ_SENT,
IB_CM_DREQ_RCVD,
IB_CM_TIMEWAIT,
IB_CM_SIDR_REQ_SENT,
IB_CM_SIDR_REQ_RCVD
};
enum ib_cm_lap_state {
IB_CM_LAP_UNINIT,
IB_CM_LAP_IDLE,
IB_CM_LAP_SENT,
IB_CM_LAP_RCVD,
IB_CM_MRA_LAP_SENT,
IB_CM_MRA_LAP_RCVD,
};
enum ib_cm_event_type {
IB_CM_REQ_ERROR,
IB_CM_REQ_RECEIVED,
IB_CM_REP_ERROR,
IB_CM_REP_RECEIVED,
IB_CM_RTU_RECEIVED,
IB_CM_USER_ESTABLISHED,
IB_CM_DREQ_ERROR,
IB_CM_DREQ_RECEIVED,
IB_CM_DREP_RECEIVED,
IB_CM_TIMEWAIT_EXIT,
IB_CM_MRA_RECEIVED,
IB_CM_REJ_RECEIVED,
IB_CM_LAP_ERROR,
IB_CM_LAP_RECEIVED,
IB_CM_APR_RECEIVED,
IB_CM_SIDR_REQ_ERROR,
IB_CM_SIDR_REQ_RECEIVED,
IB_CM_SIDR_REP_RECEIVED
};
enum ib_cm_data_size {
IB_CM_REQ_PRIVATE_DATA_SIZE = 92,
IB_CM_MRA_PRIVATE_DATA_SIZE = 222,
IB_CM_REJ_PRIVATE_DATA_SIZE = 148,
IB_CM_REP_PRIVATE_DATA_SIZE = 196,
IB_CM_RTU_PRIVATE_DATA_SIZE = 224,
IB_CM_DREQ_PRIVATE_DATA_SIZE = 220,
IB_CM_DREP_PRIVATE_DATA_SIZE = 224,
IB_CM_REJ_ARI_LENGTH = 72,
IB_CM_LAP_PRIVATE_DATA_SIZE = 168,
IB_CM_APR_PRIVATE_DATA_SIZE = 148,
IB_CM_APR_INFO_LENGTH = 72,
IB_CM_SIDR_REQ_PRIVATE_DATA_SIZE = 216,
IB_CM_SIDR_REP_PRIVATE_DATA_SIZE = 136,
IB_CM_SIDR_REP_INFO_LENGTH = 72,
};
struct ib_cm_id;
struct ib_cm_req_event_param {
struct ib_cm_id *listen_id;
/* P_Key that was used by the GMP's BTH header */
u16 bth_pkey;
u8 port;
struct sa_path_rec *primary_path;
struct sa_path_rec *alternate_path;
/*
* SGID attribute of the primary path. Currently only
* useful for RoCE. Alternate path GID attributes
* are not yet supported.
*/
const struct ib_gid_attr *ppath_sgid_attr;
__be64 remote_ca_guid;
u32 remote_qkey;
u32 remote_qpn;
enum ib_qp_type qp_type;
u32 starting_psn;
u8 responder_resources;
u8 initiator_depth;
unsigned int local_cm_response_timeout:5;
unsigned int flow_control:1;
unsigned int remote_cm_response_timeout:5;
unsigned int retry_count:3;
unsigned int rnr_retry_count:3;
unsigned int srq:1;
};
struct ib_cm_rep_event_param {
__be64 remote_ca_guid;
u32 remote_qkey;
u32 remote_qpn;
u32 starting_psn;
u8 responder_resources;
u8 initiator_depth;
unsigned int target_ack_delay:5;
unsigned int failover_accepted:2;
unsigned int flow_control:1;
unsigned int rnr_retry_count:3;
unsigned int srq:1;
};
enum ib_cm_rej_reason {
IB_CM_REJ_NO_QP = 1,
IB_CM_REJ_NO_EEC = 2,
IB_CM_REJ_NO_RESOURCES = 3,
IB_CM_REJ_TIMEOUT = 4,
IB_CM_REJ_UNSUPPORTED = 5,
IB_CM_REJ_INVALID_COMM_ID = 6,
IB_CM_REJ_INVALID_COMM_INSTANCE = 7,
IB_CM_REJ_INVALID_SERVICE_ID = 8,
IB_CM_REJ_INVALID_TRANSPORT_TYPE = 9,
IB_CM_REJ_STALE_CONN = 10,
IB_CM_REJ_RDC_NOT_EXIST = 11,
IB_CM_REJ_INVALID_GID = 12,
IB_CM_REJ_INVALID_LID = 13,
IB_CM_REJ_INVALID_SL = 14,
IB_CM_REJ_INVALID_TRAFFIC_CLASS = 15,
IB_CM_REJ_INVALID_HOP_LIMIT = 16,
IB_CM_REJ_INVALID_PACKET_RATE = 17,
IB_CM_REJ_INVALID_ALT_GID = 18,
IB_CM_REJ_INVALID_ALT_LID = 19,
IB_CM_REJ_INVALID_ALT_SL = 20,
IB_CM_REJ_INVALID_ALT_TRAFFIC_CLASS = 21,
IB_CM_REJ_INVALID_ALT_HOP_LIMIT = 22,
IB_CM_REJ_INVALID_ALT_PACKET_RATE = 23,
IB_CM_REJ_PORT_CM_REDIRECT = 24,
IB_CM_REJ_PORT_REDIRECT = 25,
IB_CM_REJ_INVALID_MTU = 26,
IB_CM_REJ_INSUFFICIENT_RESP_RESOURCES = 27,
IB_CM_REJ_CONSUMER_DEFINED = 28,
IB_CM_REJ_INVALID_RNR_RETRY = 29,
IB_CM_REJ_DUPLICATE_LOCAL_COMM_ID = 30,
IB_CM_REJ_INVALID_CLASS_VERSION = 31,
IB_CM_REJ_INVALID_FLOW_LABEL = 32,
IB_CM_REJ_INVALID_ALT_FLOW_LABEL = 33
};
struct ib_cm_rej_event_param {
enum ib_cm_rej_reason reason;
void *ari;
u8 ari_length;
};
struct ib_cm_mra_event_param {
u8 service_timeout;
};
struct ib_cm_lap_event_param {
struct sa_path_rec *alternate_path;
};
enum ib_cm_apr_status {
IB_CM_APR_SUCCESS,
IB_CM_APR_INVALID_COMM_ID,
IB_CM_APR_UNSUPPORTED,
IB_CM_APR_REJECT,
IB_CM_APR_REDIRECT,
IB_CM_APR_IS_CURRENT,
IB_CM_APR_INVALID_QPN_EECN,
IB_CM_APR_INVALID_LID,
IB_CM_APR_INVALID_GID,
IB_CM_APR_INVALID_FLOW_LABEL,
IB_CM_APR_INVALID_TCLASS,
IB_CM_APR_INVALID_HOP_LIMIT,
IB_CM_APR_INVALID_PACKET_RATE,
IB_CM_APR_INVALID_SL
};
struct ib_cm_apr_event_param {
enum ib_cm_apr_status ap_status;
void *apr_info;
u8 info_len;
};
struct ib_cm_sidr_req_event_param {
struct ib_cm_id *listen_id;
__be64 service_id;
/*
* SGID attribute of the request. Currently only
* useful for RoCE.
*/
const struct ib_gid_attr *sgid_attr;
/* P_Key that was used by the GMP's BTH header */
u16 bth_pkey;
u8 port;
u16 pkey;
};
enum ib_cm_sidr_status {
IB_SIDR_SUCCESS,
IB_SIDR_UNSUPPORTED,
IB_SIDR_REJECT,
IB_SIDR_NO_QP,
IB_SIDR_REDIRECT,
IB_SIDR_UNSUPPORTED_VERSION
};
struct ib_cm_sidr_rep_event_param {
enum ib_cm_sidr_status status;
u32 qkey;
u32 qpn;
void *info;
const struct ib_gid_attr *sgid_attr;
u8 info_len;
};
struct ib_cm_event {
enum ib_cm_event_type event;
union {
struct ib_cm_req_event_param req_rcvd;
struct ib_cm_rep_event_param rep_rcvd;
/* No data for RTU received events. */
struct ib_cm_rej_event_param rej_rcvd;
struct ib_cm_mra_event_param mra_rcvd;
struct ib_cm_lap_event_param lap_rcvd;
struct ib_cm_apr_event_param apr_rcvd;
/* No data for DREQ/DREP received events. */
struct ib_cm_sidr_req_event_param sidr_req_rcvd;
struct ib_cm_sidr_rep_event_param sidr_rep_rcvd;
enum ib_wc_status send_status;
} param;
void *private_data;
};
#define CM_REQ_ATTR_ID cpu_to_be16(0x0010)
#define CM_MRA_ATTR_ID cpu_to_be16(0x0011)
#define CM_REJ_ATTR_ID cpu_to_be16(0x0012)
#define CM_REP_ATTR_ID cpu_to_be16(0x0013)
#define CM_RTU_ATTR_ID cpu_to_be16(0x0014)
#define CM_DREQ_ATTR_ID cpu_to_be16(0x0015)
#define CM_DREP_ATTR_ID cpu_to_be16(0x0016)
#define CM_SIDR_REQ_ATTR_ID cpu_to_be16(0x0017)
#define CM_SIDR_REP_ATTR_ID cpu_to_be16(0x0018)
#define CM_LAP_ATTR_ID cpu_to_be16(0x0019)
#define CM_APR_ATTR_ID cpu_to_be16(0x001A)
/**
* ib_cm_handler - User-defined callback to process communication events.
* @cm_id: Communication identifier associated with the reported event.
* @event: Information about the communication event.
*
* IB_CM_REQ_RECEIVED and IB_CM_SIDR_REQ_RECEIVED communication events
* generated as a result of listen requests result in the allocation of a
* new @cm_id. The new @cm_id is returned to the user through this callback.
* Clients are responsible for destroying the new @cm_id. For peer-to-peer
* IB_CM_REQ_RECEIVED and all other events, the returned @cm_id corresponds
* to a user's existing communication identifier.
*
* Users may not call ib_destroy_cm_id while in the context of this callback;
* however, returning a non-zero value instructs the communication manager to
* destroy the @cm_id after the callback completes.
*/
typedef int (*ib_cm_handler)(struct ib_cm_id *cm_id,
const struct ib_cm_event *event);
struct ib_cm_id {
ib_cm_handler cm_handler;
void *context;
struct ib_device *device;
__be64 service_id;
__be64 service_mask;
enum ib_cm_state state; /* internal CM/debug use */
enum ib_cm_lap_state lap_state; /* internal CM/debug use */
__be32 local_id;
__be32 remote_id;
u32 remote_cm_qpn; /* 1 unless redirected */
};
/**
* ib_create_cm_id - Allocate a communication identifier.
* @device: Device associated with the cm_id. All related communication will
* be associated with the specified device.
* @cm_handler: Callback invoked to notify the user of CM events.
* @context: User specified context associated with the communication
* identifier.
*
* Communication identifiers are used to track connection states, service
* ID resolution requests, and listen requests.
*/
struct ib_cm_id *ib_create_cm_id(struct ib_device *device,
ib_cm_handler cm_handler,
void *context);
/**
* ib_destroy_cm_id - Destroy a connection identifier.
* @cm_id: Connection identifier to destroy.
*
* This call blocks until the connection identifier is destroyed.
*/
void ib_destroy_cm_id(struct ib_cm_id *cm_id);
#define IB_SERVICE_ID_AGN_MASK cpu_to_be64(0xFF00000000000000ULL)
#define IB_CM_ASSIGN_SERVICE_ID cpu_to_be64(0x0200000000000000ULL)
#define IB_CMA_SERVICE_ID cpu_to_be64(0x0000000001000000ULL)
#define IB_CMA_SERVICE_ID_MASK cpu_to_be64(0xFFFFFFFFFF000000ULL)
#define IB_SDP_SERVICE_ID cpu_to_be64(0x0000000000010000ULL)
#define IB_SDP_SERVICE_ID_MASK cpu_to_be64(0xFFFFFFFFFFFF0000ULL)
/**
* ib_cm_listen - Initiates listening on the specified service ID for
* connection and service ID resolution requests.
* @cm_id: Connection identifier associated with the listen request.
* @service_id: Service identifier matched against incoming connection
* and service ID resolution requests. The service ID should be specified
* network-byte order. If set to IB_CM_ASSIGN_SERVICE_ID, the CM will
* assign a service ID to the caller.
* @service_mask: Mask applied to service ID used to listen across a
* range of service IDs. If set to 0, the service ID is matched
* exactly. This parameter is ignored if %service_id is set to
* IB_CM_ASSIGN_SERVICE_ID.
*/
int ib_cm_listen(struct ib_cm_id *cm_id, __be64 service_id,
__be64 service_mask);
struct ib_cm_id *ib_cm_insert_listen(struct ib_device *device,
ib_cm_handler cm_handler,
__be64 service_id);
struct ib_cm_req_param {
struct sa_path_rec *primary_path;
struct sa_path_rec *alternate_path;
const struct ib_gid_attr *ppath_sgid_attr;
__be64 service_id;
u32 qp_num;
enum ib_qp_type qp_type;
u32 starting_psn;
const void *private_data;
u8 private_data_len;
u8 peer_to_peer;
u8 responder_resources;
u8 initiator_depth;
u8 remote_cm_response_timeout;
u8 flow_control;
u8 local_cm_response_timeout;
u8 retry_count;
u8 rnr_retry_count;
u8 max_cm_retries;
u8 srq;
};
/**
* ib_send_cm_req - Sends a connection request to the remote node.
* @cm_id: Connection identifier that will be associated with the
* connection request.
* @param: Connection request information needed to establish the
* connection.
*/
int ib_send_cm_req(struct ib_cm_id *cm_id,
struct ib_cm_req_param *param);
struct ib_cm_rep_param {
u32 qp_num;
u32 starting_psn;
const void *private_data;
u8 private_data_len;
u8 responder_resources;
u8 initiator_depth;
u8 failover_accepted;
u8 flow_control;
u8 rnr_retry_count;
u8 srq;
};
/**
* ib_send_cm_rep - Sends a connection reply in response to a connection
* request.
* @cm_id: Connection identifier that will be associated with the
* connection request.
* @param: Connection reply information needed to establish the
* connection.
*/
int ib_send_cm_rep(struct ib_cm_id *cm_id,
struct ib_cm_rep_param *param);
/**
* ib_send_cm_rtu - Sends a connection ready to use message in response
* to a connection reply message.
* @cm_id: Connection identifier associated with the connection request.
* @private_data: Optional user-defined private data sent with the
* ready to use message.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_rtu(struct ib_cm_id *cm_id,
const void *private_data,
u8 private_data_len);
/**
* ib_send_cm_dreq - Sends a disconnection request for an existing
* connection.
* @cm_id: Connection identifier associated with the connection being
* released.
* @private_data: Optional user-defined private data sent with the
* disconnection request message.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_dreq(struct ib_cm_id *cm_id,
const void *private_data,
u8 private_data_len);
/**
* ib_send_cm_drep - Sends a disconnection reply to a disconnection request.
* @cm_id: Connection identifier associated with the connection being
* released.
* @private_data: Optional user-defined private data sent with the
* disconnection reply message.
* @private_data_len: Size of the private data buffer, in bytes.
*
* If the cm_id is in the correct state, the CM will transition the connection
* to the timewait state, even if an error occurs sending the DREP message.
*/
int ib_send_cm_drep(struct ib_cm_id *cm_id,
const void *private_data,
u8 private_data_len);
/**
* ib_cm_notify - Notifies the CM of an event reported to the consumer.
* @cm_id: Connection identifier to transition to established.
* @event: Type of event.
*
* This routine should be invoked by users to notify the CM of relevant
* communication events. Events that should be reported to the CM and
* when to report them are:
*
* IB_EVENT_COMM_EST - Used when a message is received on a connected
* QP before an RTU has been received.
* IB_EVENT_PATH_MIG - Notifies the CM that the connection has failed over
* to the alternate path.
*/
int ib_cm_notify(struct ib_cm_id *cm_id, enum ib_event_type event);
/**
* ib_send_cm_rej - Sends a connection rejection message to the
* remote node.
* @cm_id: Connection identifier associated with the connection being
* rejected.
* @reason: Reason for the connection request rejection.
* @ari: Optional additional rejection information.
* @ari_length: Size of the additional rejection information, in bytes.
* @private_data: Optional user-defined private data sent with the
* rejection message.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_rej(struct ib_cm_id *cm_id,
enum ib_cm_rej_reason reason,
void *ari,
u8 ari_length,
const void *private_data,
u8 private_data_len);
#define IB_CM_MRA_FLAG_DELAY 0x80 /* Send MRA only after a duplicate msg */
/**
* ib_send_cm_mra - Sends a message receipt acknowledgement to a connection
* message.
* @cm_id: Connection identifier associated with the connection message.
* @service_timeout: The lower 5-bits specify the maximum time required for
* the sender to reply to the connection message. The upper 3-bits
* specify additional control flags.
* @private_data: Optional user-defined private data sent with the
* message receipt acknowledgement.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_mra(struct ib_cm_id *cm_id,
u8 service_timeout,
const void *private_data,
u8 private_data_len);
/**
* ib_send_cm_lap - Sends a load alternate path request.
* @cm_id: Connection identifier associated with the load alternate path
* message.
* @alternate_path: A path record that identifies the alternate path to
* load.
* @private_data: Optional user-defined private data sent with the
* load alternate path message.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_lap(struct ib_cm_id *cm_id,
struct sa_path_rec *alternate_path,
const void *private_data,
u8 private_data_len);
/**
* ib_cm_init_qp_attr - Initializes the QP attributes for use in transitioning
* to a specified QP state.
* @cm_id: Communication identifier associated with the QP attributes to
* initialize.
* @qp_attr: On input, specifies the desired QP state. On output, the
* mandatory and desired optional attributes will be set in order to
* modify the QP to the specified state.
* @qp_attr_mask: The QP attribute mask that may be used to transition the
* QP to the specified state.
*
* Users must set the @qp_attr->qp_state to the desired QP state. This call
* will set all required attributes for the given transition, along with
* known optional attributes. Users may override the attributes returned from
* this call before calling ib_modify_qp.
*/
int ib_cm_init_qp_attr(struct ib_cm_id *cm_id,
struct ib_qp_attr *qp_attr,
int *qp_attr_mask);
/**
* ib_send_cm_apr - Sends an alternate path response message in response to
* a load alternate path request.
* @cm_id: Connection identifier associated with the alternate path response.
* @status: Reply status sent with the alternate path response.
* @info: Optional additional information sent with the alternate path
* response.
* @info_length: Size of the additional information, in bytes.
* @private_data: Optional user-defined private data sent with the
* alternate path response message.
* @private_data_len: Size of the private data buffer, in bytes.
*/
int ib_send_cm_apr(struct ib_cm_id *cm_id,
enum ib_cm_apr_status status,
void *info,
u8 info_length,
const void *private_data,
u8 private_data_len);
struct ib_cm_sidr_req_param {
struct sa_path_rec *path;
const struct ib_gid_attr *sgid_attr;
__be64 service_id;
int timeout_ms;
const void *private_data;
u8 private_data_len;
u8 max_cm_retries;
};
/**
* ib_send_cm_sidr_req - Sends a service ID resolution request to the
* remote node.
* @cm_id: Communication identifier that will be associated with the
* service ID resolution request.
* @param: Service ID resolution request information.
*/
int ib_send_cm_sidr_req(struct ib_cm_id *cm_id,
struct ib_cm_sidr_req_param *param);
struct ib_cm_sidr_rep_param {
u32 qp_num;
u32 qkey;
enum ib_cm_sidr_status status;
const void *info;
u8 info_length;
const void *private_data;
u8 private_data_len;
};
/**
* ib_send_cm_sidr_rep - Sends a service ID resolution reply to the
* remote node.
* @cm_id: Communication identifier associated with the received service ID
* resolution request.
* @param: Service ID resolution reply information.
*/
int ib_send_cm_sidr_rep(struct ib_cm_id *cm_id,
struct ib_cm_sidr_rep_param *param);
/**
* ibcm_reject_msg - return a pointer to a reject message string.
* @reason: Value returned in the REJECT event status field.
*/
const char *__attribute_const__ ibcm_reject_msg(int reason);
#endif /* IB_CM_H */
+93
View File
@@ -0,0 +1,93 @@
/*
* Copyright (c) 2004 Topspin Corporation. All rights reserved.
* Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_FMR_POOL_H)
#define IB_FMR_POOL_H
#include <rdma/ib_verbs.h>
struct ib_fmr_pool;
/**
* struct ib_fmr_pool_param - Parameters for creating FMR pool
* @max_pages_per_fmr:Maximum number of pages per map request.
* @page_shift: Log2 of sizeof "pages" mapped by this fmr
* @access:Access flags for FMRs in pool.
* @pool_size:Number of FMRs to allocate for pool.
* @dirty_watermark:Flush is triggered when @dirty_watermark dirty
* FMRs are present.
* @flush_function:Callback called when unmapped FMRs are flushed and
* more FMRs are possibly available for mapping
* @flush_arg:Context passed to user's flush function.
* @cache:If set, FMRs may be reused after unmapping for identical map
* requests.
*/
struct ib_fmr_pool_param {
int max_pages_per_fmr;
int page_shift;
enum ib_access_flags access;
int pool_size;
int dirty_watermark;
void (*flush_function)(struct ib_fmr_pool *pool,
void *arg);
void *flush_arg;
unsigned cache:1;
};
struct ib_pool_fmr {
struct ib_fmr *fmr;
struct ib_fmr_pool *pool;
struct list_head list;
struct hlist_node cache_node;
int ref_count;
int remap_count;
u64 io_virtual_address;
int page_list_len;
u64 page_list[0];
};
struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd *pd,
struct ib_fmr_pool_param *params);
void ib_destroy_fmr_pool(struct ib_fmr_pool *pool);
int ib_flush_fmr_pool(struct ib_fmr_pool *pool);
struct ib_pool_fmr *ib_fmr_pool_map_phys(struct ib_fmr_pool *pool_handle,
u64 *page_list,
int list_len,
u64 io_virtual_address);
int ib_fmr_pool_unmap(struct ib_pool_fmr *fmr);
#endif /* IB_FMR_POOL_H */
+341
View File
@@ -0,0 +1,341 @@
/*
* Copyright(c) 2016 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#ifndef IB_HDRS_H
#define IB_HDRS_H
#include <linux/types.h>
#include <asm/unaligned.h>
#include <rdma/ib_verbs.h>
#define IB_SEQ_NAK (3 << 29)
/* AETH NAK opcode values */
#define IB_RNR_NAK 0x20
#define IB_NAK_PSN_ERROR 0x60
#define IB_NAK_INVALID_REQUEST 0x61
#define IB_NAK_REMOTE_ACCESS_ERROR 0x62
#define IB_NAK_REMOTE_OPERATIONAL_ERROR 0x63
#define IB_NAK_INVALID_RD_REQUEST 0x64
#define IB_BTH_REQ_ACK BIT(31)
#define IB_BTH_SOLICITED BIT(23)
#define IB_BTH_MIG_REQ BIT(22)
#define IB_GRH_VERSION 6
#define IB_GRH_VERSION_MASK 0xF
#define IB_GRH_VERSION_SHIFT 28
#define IB_GRH_TCLASS_MASK 0xFF
#define IB_GRH_TCLASS_SHIFT 20
#define IB_GRH_FLOW_MASK 0xFFFFF
#define IB_GRH_FLOW_SHIFT 0
#define IB_GRH_NEXT_HDR 0x1B
#define IB_FECN_SHIFT 31
#define IB_FECN_MASK 1
#define IB_FECN_SMASK BIT(IB_FECN_SHIFT)
#define IB_BECN_SHIFT 30
#define IB_BECN_MASK 1
#define IB_BECN_SMASK BIT(IB_BECN_SHIFT)
#define IB_AETH_CREDIT_SHIFT 24
#define IB_AETH_CREDIT_MASK 0x1F
#define IB_AETH_CREDIT_INVAL 0x1F
#define IB_AETH_NAK_SHIFT 29
#define IB_MSN_MASK 0xFFFFFF
struct ib_reth {
__be64 vaddr; /* potentially unaligned */
__be32 rkey;
__be32 length;
} __packed;
struct ib_atomic_eth {
__be64 vaddr; /* potentially unaligned */
__be32 rkey;
__be64 swap_data; /* potentially unaligned */
__be64 compare_data; /* potentially unaligned */
} __packed;
union ib_ehdrs {
struct {
__be32 deth[2];
__be32 imm_data;
} ud;
struct {
struct ib_reth reth;
__be32 imm_data;
} rc;
struct {
__be32 aeth;
__be64 atomic_ack_eth; /* potentially unaligned */
} __packed at;
__be32 imm_data;
__be32 aeth;
__be32 ieth;
struct ib_atomic_eth atomic_eth;
} __packed;
struct ib_other_headers {
__be32 bth[3];
union ib_ehdrs u;
} __packed;
struct ib_header {
__be16 lrh[4];
union {
struct {
struct ib_grh grh;
struct ib_other_headers oth;
} l;
struct ib_other_headers oth;
} u;
} __packed;
/* accessors for unaligned __be64 items */
static inline u64 ib_u64_get(__be64 *p)
{
return get_unaligned_be64(p);
}
static inline void ib_u64_put(u64 val, __be64 *p)
{
put_unaligned_be64(val, p);
}
static inline u64 get_ib_reth_vaddr(struct ib_reth *reth)
{
return ib_u64_get(&reth->vaddr);
}
static inline void put_ib_reth_vaddr(u64 val, struct ib_reth *reth)
{
ib_u64_put(val, &reth->vaddr);
}
static inline u64 get_ib_ateth_vaddr(struct ib_atomic_eth *ateth)
{
return ib_u64_get(&ateth->vaddr);
}
static inline void put_ib_ateth_vaddr(u64 val, struct ib_atomic_eth *ateth)
{
ib_u64_put(val, &ateth->vaddr);
}
static inline u64 get_ib_ateth_swap(struct ib_atomic_eth *ateth)
{
return ib_u64_get(&ateth->swap_data);
}
static inline void put_ib_ateth_swap(u64 val, struct ib_atomic_eth *ateth)
{
ib_u64_put(val, &ateth->swap_data);
}
static inline u64 get_ib_ateth_compare(struct ib_atomic_eth *ateth)
{
return ib_u64_get(&ateth->compare_data);
}
static inline void put_ib_ateth_compare(u64 val, struct ib_atomic_eth *ateth)
{
ib_u64_put(val, &ateth->compare_data);
}
/*
* 9B/IB Packet Format
*/
#define IB_LNH_MASK 3
#define IB_SC_MASK 0xf
#define IB_SC_SHIFT 12
#define IB_SC5_MASK 0x10
#define IB_SL_MASK 0xf
#define IB_SL_SHIFT 4
#define IB_SL_SHIFT 4
#define IB_LVER_MASK 0xf
#define IB_LVER_SHIFT 8
static inline u8 ib_get_lnh(struct ib_header *hdr)
{
return (be16_to_cpu(hdr->lrh[0]) & IB_LNH_MASK);
}
static inline u8 ib_get_sc(struct ib_header *hdr)
{
return ((be16_to_cpu(hdr->lrh[0]) >> IB_SC_SHIFT) & IB_SC_MASK);
}
static inline bool ib_is_sc5(u16 sc5)
{
return !!(sc5 & IB_SC5_MASK);
}
static inline u8 ib_get_sl(struct ib_header *hdr)
{
return ((be16_to_cpu(hdr->lrh[0]) >> IB_SL_SHIFT) & IB_SL_MASK);
}
static inline u16 ib_get_dlid(struct ib_header *hdr)
{
return (be16_to_cpu(hdr->lrh[1]));
}
static inline u16 ib_get_slid(struct ib_header *hdr)
{
return (be16_to_cpu(hdr->lrh[3]));
}
static inline u8 ib_get_lver(struct ib_header *hdr)
{
return (u8)((be16_to_cpu(hdr->lrh[0]) >> IB_LVER_SHIFT) &
IB_LVER_MASK);
}
static inline u16 ib_get_len(struct ib_header *hdr)
{
return (u16)(be16_to_cpu(hdr->lrh[2]));
}
static inline u32 ib_get_qkey(struct ib_other_headers *ohdr)
{
return be32_to_cpu(ohdr->u.ud.deth[0]);
}
static inline u32 ib_get_sqpn(struct ib_other_headers *ohdr)
{
return ((be32_to_cpu(ohdr->u.ud.deth[1])) & IB_QPN_MASK);
}
/*
* BTH
*/
#define IB_BTH_OPCODE_MASK 0xff
#define IB_BTH_OPCODE_SHIFT 24
#define IB_BTH_PAD_MASK 3
#define IB_BTH_PKEY_MASK 0xffff
#define IB_BTH_PAD_SHIFT 20
#define IB_BTH_A_MASK 1
#define IB_BTH_A_SHIFT 31
#define IB_BTH_M_MASK 1
#define IB_BTH_M_SHIFT 22
#define IB_BTH_SE_MASK 1
#define IB_BTH_SE_SHIFT 23
#define IB_BTH_TVER_MASK 0xf
#define IB_BTH_TVER_SHIFT 16
static inline u8 ib_bth_get_pad(struct ib_other_headers *ohdr)
{
return ((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_PAD_SHIFT) &
IB_BTH_PAD_MASK);
}
static inline u16 ib_bth_get_pkey(struct ib_other_headers *ohdr)
{
return (be32_to_cpu(ohdr->bth[0]) & IB_BTH_PKEY_MASK);
}
static inline u8 ib_bth_get_opcode(struct ib_other_headers *ohdr)
{
return ((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_OPCODE_SHIFT) &
IB_BTH_OPCODE_MASK);
}
static inline u8 ib_bth_get_ackreq(struct ib_other_headers *ohdr)
{
return (u8)((be32_to_cpu(ohdr->bth[2]) >> IB_BTH_A_SHIFT) &
IB_BTH_A_MASK);
}
static inline u8 ib_bth_get_migreq(struct ib_other_headers *ohdr)
{
return (u8)((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_M_SHIFT) &
IB_BTH_M_MASK);
}
static inline u8 ib_bth_get_se(struct ib_other_headers *ohdr)
{
return (u8)((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_SE_SHIFT) &
IB_BTH_SE_MASK);
}
static inline u32 ib_bth_get_psn(struct ib_other_headers *ohdr)
{
return (u32)(be32_to_cpu(ohdr->bth[2]));
}
static inline u32 ib_bth_get_qpn(struct ib_other_headers *ohdr)
{
return (u32)((be32_to_cpu(ohdr->bth[1])) & IB_QPN_MASK);
}
static inline bool ib_bth_get_becn(struct ib_other_headers *ohdr)
{
return (ohdr->bth[1]) & cpu_to_be32(IB_BECN_SMASK);
}
static inline bool ib_bth_get_fecn(struct ib_other_headers *ohdr)
{
return (ohdr->bth[1]) & cpu_to_be32(IB_FECN_SMASK);
}
static inline u8 ib_bth_get_tver(struct ib_other_headers *ohdr)
{
return (u8)((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_TVER_SHIFT) &
IB_BTH_TVER_MASK);
}
static inline bool ib_bth_is_solicited(struct ib_other_headers *ohdr)
{
return ohdr->bth[0] & cpu_to_be32(IB_BTH_SOLICITED);
}
static inline bool ib_bth_is_migration(struct ib_other_headers *ohdr)
{
return ohdr->bth[0] & cpu_to_be32(IB_BTH_MIG_REQ);
}
#endif /* IB_HDRS_H */
+924
View File
@@ -0,0 +1,924 @@
/*
* Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved.
* Copyright (c) 2004 Infinicon Corporation. All rights reserved.
* Copyright (c) 2004 Intel Corporation. All rights reserved.
* Copyright (c) 2004 Topspin Corporation. All rights reserved.
* Copyright (c) 2004-2006 Voltaire Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_MAD_H)
#define IB_MAD_H
#include <linux/list.h>
#include <rdma/ib_verbs.h>
#include <uapi/rdma/ib_user_mad.h>
/* Management base versions */
#define IB_MGMT_BASE_VERSION 1
#define OPA_MGMT_BASE_VERSION 0x80
#define OPA_SM_CLASS_VERSION 0x80
/* Management classes */
#define IB_MGMT_CLASS_SUBN_LID_ROUTED 0x01
#define IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE 0x81
#define IB_MGMT_CLASS_SUBN_ADM 0x03
#define IB_MGMT_CLASS_PERF_MGMT 0x04
#define IB_MGMT_CLASS_BM 0x05
#define IB_MGMT_CLASS_DEVICE_MGMT 0x06
#define IB_MGMT_CLASS_CM 0x07
#define IB_MGMT_CLASS_SNMP 0x08
#define IB_MGMT_CLASS_DEVICE_ADM 0x10
#define IB_MGMT_CLASS_BOOT_MGMT 0x11
#define IB_MGMT_CLASS_BIS 0x12
#define IB_MGMT_CLASS_CONG_MGMT 0x21
#define IB_MGMT_CLASS_VENDOR_RANGE2_START 0x30
#define IB_MGMT_CLASS_VENDOR_RANGE2_END 0x4F
#define IB_OPENIB_OUI (0x001405)
/* Management methods */
#define IB_MGMT_METHOD_GET 0x01
#define IB_MGMT_METHOD_SET 0x02
#define IB_MGMT_METHOD_GET_RESP 0x81
#define IB_MGMT_METHOD_SEND 0x03
#define IB_MGMT_METHOD_TRAP 0x05
#define IB_MGMT_METHOD_REPORT 0x06
#define IB_MGMT_METHOD_REPORT_RESP 0x86
#define IB_MGMT_METHOD_TRAP_REPRESS 0x07
#define IB_MGMT_METHOD_RESP 0x80
#define IB_BM_ATTR_MOD_RESP cpu_to_be32(1)
#define IB_MGMT_MAX_METHODS 128
/* MAD Status field bit masks */
#define IB_MGMT_MAD_STATUS_SUCCESS 0x0000
#define IB_MGMT_MAD_STATUS_BUSY 0x0001
#define IB_MGMT_MAD_STATUS_REDIRECT_REQD 0x0002
#define IB_MGMT_MAD_STATUS_BAD_VERSION 0x0004
#define IB_MGMT_MAD_STATUS_UNSUPPORTED_METHOD 0x0008
#define IB_MGMT_MAD_STATUS_UNSUPPORTED_METHOD_ATTRIB 0x000c
#define IB_MGMT_MAD_STATUS_INVALID_ATTRIB_VALUE 0x001c
/* RMPP information */
#define IB_MGMT_RMPP_VERSION 1
#define IB_MGMT_RMPP_TYPE_DATA 1
#define IB_MGMT_RMPP_TYPE_ACK 2
#define IB_MGMT_RMPP_TYPE_STOP 3
#define IB_MGMT_RMPP_TYPE_ABORT 4
#define IB_MGMT_RMPP_FLAG_ACTIVE 1
#define IB_MGMT_RMPP_FLAG_FIRST (1<<1)
#define IB_MGMT_RMPP_FLAG_LAST (1<<2)
#define IB_MGMT_RMPP_NO_RESPTIME 0x1F
#define IB_MGMT_RMPP_STATUS_SUCCESS 0
#define IB_MGMT_RMPP_STATUS_RESX 1
#define IB_MGMT_RMPP_STATUS_ABORT_MIN 118
#define IB_MGMT_RMPP_STATUS_T2L 118
#define IB_MGMT_RMPP_STATUS_BAD_LEN 119
#define IB_MGMT_RMPP_STATUS_BAD_SEG 120
#define IB_MGMT_RMPP_STATUS_BADT 121
#define IB_MGMT_RMPP_STATUS_W2S 122
#define IB_MGMT_RMPP_STATUS_S2B 123
#define IB_MGMT_RMPP_STATUS_BAD_STATUS 124
#define IB_MGMT_RMPP_STATUS_UNV 125
#define IB_MGMT_RMPP_STATUS_TMR 126
#define IB_MGMT_RMPP_STATUS_UNSPEC 127
#define IB_MGMT_RMPP_STATUS_ABORT_MAX 127
#define IB_QP0 0
#define IB_QP1 cpu_to_be32(1)
#define IB_QP1_QKEY 0x80010000
#define IB_QP_SET_QKEY 0x80000000
#define IB_DEFAULT_PKEY_PARTIAL 0x7FFF
#define IB_DEFAULT_PKEY_FULL 0xFFFF
/*
* Generic trap/notice types
*/
#define IB_NOTICE_TYPE_FATAL 0x80
#define IB_NOTICE_TYPE_URGENT 0x81
#define IB_NOTICE_TYPE_SECURITY 0x82
#define IB_NOTICE_TYPE_SM 0x83
#define IB_NOTICE_TYPE_INFO 0x84
/*
* Generic trap/notice producers
*/
#define IB_NOTICE_PROD_CA cpu_to_be16(1)
#define IB_NOTICE_PROD_SWITCH cpu_to_be16(2)
#define IB_NOTICE_PROD_ROUTER cpu_to_be16(3)
#define IB_NOTICE_PROD_CLASS_MGR cpu_to_be16(4)
enum {
IB_MGMT_MAD_HDR = 24,
IB_MGMT_MAD_DATA = 232,
IB_MGMT_RMPP_HDR = 36,
IB_MGMT_RMPP_DATA = 220,
IB_MGMT_VENDOR_HDR = 40,
IB_MGMT_VENDOR_DATA = 216,
IB_MGMT_SA_HDR = 56,
IB_MGMT_SA_DATA = 200,
IB_MGMT_DEVICE_HDR = 64,
IB_MGMT_DEVICE_DATA = 192,
IB_MGMT_MAD_SIZE = IB_MGMT_MAD_HDR + IB_MGMT_MAD_DATA,
OPA_MGMT_MAD_DATA = 2024,
OPA_MGMT_RMPP_DATA = 2012,
OPA_MGMT_MAD_SIZE = IB_MGMT_MAD_HDR + OPA_MGMT_MAD_DATA,
};
struct ib_mad_hdr {
u8 base_version;
u8 mgmt_class;
u8 class_version;
u8 method;
__be16 status;
__be16 class_specific;
__be64 tid;
__be16 attr_id;
__be16 resv;
__be32 attr_mod;
};
struct ib_rmpp_hdr {
u8 rmpp_version;
u8 rmpp_type;
u8 rmpp_rtime_flags;
u8 rmpp_status;
__be32 seg_num;
__be32 paylen_newwin;
};
typedef u64 __bitwise ib_sa_comp_mask;
#define IB_SA_COMP_MASK(n) ((__force ib_sa_comp_mask) cpu_to_be64(1ull << (n)))
/*
* ib_sa_hdr and ib_sa_mad structures must be packed because they have
* 64-bit fields that are only 32-bit aligned. 64-bit architectures will
* lay them out wrong otherwise. (And unfortunately they are sent on
* the wire so we can't change the layout)
*/
struct ib_sa_hdr {
__be64 sm_key;
__be16 attr_offset;
__be16 reserved;
ib_sa_comp_mask comp_mask;
} __attribute__ ((packed));
struct ib_mad {
struct ib_mad_hdr mad_hdr;
u8 data[IB_MGMT_MAD_DATA];
};
struct opa_mad {
struct ib_mad_hdr mad_hdr;
u8 data[OPA_MGMT_MAD_DATA];
};
struct ib_rmpp_mad {
struct ib_mad_hdr mad_hdr;
struct ib_rmpp_hdr rmpp_hdr;
u8 data[IB_MGMT_RMPP_DATA];
};
struct opa_rmpp_mad {
struct ib_mad_hdr mad_hdr;
struct ib_rmpp_hdr rmpp_hdr;
u8 data[OPA_MGMT_RMPP_DATA];
};
struct ib_sa_mad {
struct ib_mad_hdr mad_hdr;
struct ib_rmpp_hdr rmpp_hdr;
struct ib_sa_hdr sa_hdr;
u8 data[IB_MGMT_SA_DATA];
} __attribute__ ((packed));
struct ib_vendor_mad {
struct ib_mad_hdr mad_hdr;
struct ib_rmpp_hdr rmpp_hdr;
u8 reserved;
u8 oui[3];
u8 data[IB_MGMT_VENDOR_DATA];
};
#define IB_MGMT_CLASSPORTINFO_ATTR_ID cpu_to_be16(0x0001)
#define IB_CLASS_PORT_INFO_RESP_TIME_MASK 0x1F
#define IB_CLASS_PORT_INFO_RESP_TIME_FIELD_SIZE 5
struct ib_class_port_info {
u8 base_version;
u8 class_version;
__be16 capability_mask;
/* 27 bits for cap_mask2, 5 bits for resp_time */
__be32 cap_mask2_resp_time;
u8 redirect_gid[16];
__be32 redirect_tcslfl;
__be16 redirect_lid;
__be16 redirect_pkey;
__be32 redirect_qp;
__be32 redirect_qkey;
u8 trap_gid[16];
__be32 trap_tcslfl;
__be16 trap_lid;
__be16 trap_pkey;
__be32 trap_hlqp;
__be32 trap_qkey;
};
/* PortInfo CapabilityMask */
enum ib_port_capability_mask_bits {
IB_PORT_SM = 1 << 1,
IB_PORT_NOTICE_SUP = 1 << 2,
IB_PORT_TRAP_SUP = 1 << 3,
IB_PORT_OPT_IPD_SUP = 1 << 4,
IB_PORT_AUTO_MIGR_SUP = 1 << 5,
IB_PORT_SL_MAP_SUP = 1 << 6,
IB_PORT_MKEY_NVRAM = 1 << 7,
IB_PORT_PKEY_NVRAM = 1 << 8,
IB_PORT_LED_INFO_SUP = 1 << 9,
IB_PORT_SM_DISABLED = 1 << 10,
IB_PORT_SYS_IMAGE_GUID_SUP = 1 << 11,
IB_PORT_PKEY_SW_EXT_PORT_TRAP_SUP = 1 << 12,
IB_PORT_EXTENDED_SPEEDS_SUP = 1 << 14,
IB_PORT_CM_SUP = 1 << 16,
IB_PORT_SNMP_TUNNEL_SUP = 1 << 17,
IB_PORT_REINIT_SUP = 1 << 18,
IB_PORT_DEVICE_MGMT_SUP = 1 << 19,
IB_PORT_VENDOR_CLASS_SUP = 1 << 20,
IB_PORT_DR_NOTICE_SUP = 1 << 21,
IB_PORT_CAP_MASK_NOTICE_SUP = 1 << 22,
IB_PORT_BOOT_MGMT_SUP = 1 << 23,
IB_PORT_LINK_LATENCY_SUP = 1 << 24,
IB_PORT_CLIENT_REG_SUP = 1 << 25,
IB_PORT_OTHER_LOCAL_CHANGES_SUP = 1 << 26,
IB_PORT_LINK_SPEED_WIDTH_TABLE_SUP = 1 << 27,
IB_PORT_VENDOR_SPECIFIC_MADS_TABLE_SUP = 1 << 28,
IB_PORT_MCAST_PKEY_TRAP_SUPPRESSION_SUP = 1 << 29,
IB_PORT_MCAST_FDB_TOP_SUP = 1 << 30,
IB_PORT_HIERARCHY_INFO_SUP = 1ULL << 31,
};
#define OPA_CLASS_PORT_INFO_PR_SUPPORT BIT(26)
struct opa_class_port_info {
u8 base_version;
u8 class_version;
__be16 cap_mask;
__be32 cap_mask2_resp_time;
u8 redirect_gid[16];
__be32 redirect_tc_fl;
__be32 redirect_lid;
__be32 redirect_sl_qp;
__be32 redirect_qkey;
u8 trap_gid[16];
__be32 trap_tc_fl;
__be32 trap_lid;
__be32 trap_hl_qp;
__be32 trap_qkey;
__be16 trap_pkey;
__be16 redirect_pkey;
u8 trap_sl_rsvd;
u8 reserved[3];
} __packed;
/**
* ib_get_cpi_resp_time - Returns the resp_time value from
* cap_mask2_resp_time in ib_class_port_info.
* @cpi: A struct ib_class_port_info mad.
*/
static inline u8 ib_get_cpi_resp_time(struct ib_class_port_info *cpi)
{
return (u8)(be32_to_cpu(cpi->cap_mask2_resp_time) &
IB_CLASS_PORT_INFO_RESP_TIME_MASK);
}
/**
* ib_set_cpi_resptime - Sets the response time in an
* ib_class_port_info mad.
* @cpi: A struct ib_class_port_info.
* @rtime: The response time to set.
*/
static inline void ib_set_cpi_resp_time(struct ib_class_port_info *cpi,
u8 rtime)
{
cpi->cap_mask2_resp_time =
(cpi->cap_mask2_resp_time &
cpu_to_be32(~IB_CLASS_PORT_INFO_RESP_TIME_MASK)) |
cpu_to_be32(rtime & IB_CLASS_PORT_INFO_RESP_TIME_MASK);
}
/**
* ib_get_cpi_capmask2 - Returns the capmask2 value from
* cap_mask2_resp_time in ib_class_port_info.
* @cpi: A struct ib_class_port_info mad.
*/
static inline u32 ib_get_cpi_capmask2(struct ib_class_port_info *cpi)
{
return (be32_to_cpu(cpi->cap_mask2_resp_time) >>
IB_CLASS_PORT_INFO_RESP_TIME_FIELD_SIZE);
}
/**
* ib_set_cpi_capmask2 - Sets the capmask2 in an
* ib_class_port_info mad.
* @cpi: A struct ib_class_port_info.
* @capmask2: The capmask2 to set.
*/
static inline void ib_set_cpi_capmask2(struct ib_class_port_info *cpi,
u32 capmask2)
{
cpi->cap_mask2_resp_time =
(cpi->cap_mask2_resp_time &
cpu_to_be32(IB_CLASS_PORT_INFO_RESP_TIME_MASK)) |
cpu_to_be32(capmask2 <<
IB_CLASS_PORT_INFO_RESP_TIME_FIELD_SIZE);
}
/**
* opa_get_cpi_capmask2 - Returns the capmask2 value from
* cap_mask2_resp_time in ib_class_port_info.
* @cpi: A struct opa_class_port_info mad.
*/
static inline u32 opa_get_cpi_capmask2(struct opa_class_port_info *cpi)
{
return (be32_to_cpu(cpi->cap_mask2_resp_time) >>
IB_CLASS_PORT_INFO_RESP_TIME_FIELD_SIZE);
}
struct ib_mad_notice_attr {
u8 generic_type;
u8 prod_type_msb;
__be16 prod_type_lsb;
__be16 trap_num;
__be16 issuer_lid;
__be16 toggle_count;
union {
struct {
u8 details[54];
} raw_data;
struct {
__be16 reserved;
__be16 lid; /* where violation happened */
u8 port_num; /* where violation happened */
} __packed ntc_129_131;
struct {
__be16 reserved;
__be16 lid; /* LID where change occurred */
u8 reserved2;
u8 local_changes; /* low bit - local changes */
__be32 new_cap_mask; /* new capability mask */
u8 reserved3;
u8 change_flags; /* low 3 bits only */
} __packed ntc_144;
struct {
__be16 reserved;
__be16 lid; /* lid where sys guid changed */
__be16 reserved2;
__be64 new_sys_guid;
} __packed ntc_145;
struct {
__be16 reserved;
__be16 lid;
__be16 dr_slid;
u8 method;
u8 reserved2;
__be16 attr_id;
__be32 attr_mod;
__be64 mkey;
u8 reserved3;
u8 dr_trunc_hop;
u8 dr_rtn_path[30];
} __packed ntc_256;
struct {
__be16 reserved;
__be16 lid1;
__be16 lid2;
__be32 key;
__be32 sl_qp1; /* SL: high 4 bits */
__be32 qp2; /* high 8 bits reserved */
union ib_gid gid1;
union ib_gid gid2;
} __packed ntc_257_258;
} details;
};
/**
* ib_mad_send_buf - MAD data buffer and work request for sends.
* @next: A pointer used to chain together MADs for posting.
* @mad: References an allocated MAD data buffer for MADs that do not have
* RMPP active. For MADs using RMPP, references the common and management
* class specific headers.
* @mad_agent: MAD agent that allocated the buffer.
* @ah: The address handle to use when sending the MAD.
* @context: User-controlled context fields.
* @hdr_len: Indicates the size of the data header of the MAD. This length
* includes the common MAD, RMPP, and class specific headers.
* @data_len: Indicates the total size of user-transferred data.
* @seg_count: The number of RMPP segments allocated for this send.
* @seg_size: Size of the data in each RMPP segment. This does not include
* class specific headers.
* @seg_rmpp_size: Size of each RMPP segment including the class specific
* headers.
* @timeout_ms: Time to wait for a response.
* @retries: Number of times to retry a request for a response. For MADs
* using RMPP, this applies per window. On completion, returns the number
* of retries needed to complete the transfer.
*
* Users are responsible for initializing the MAD buffer itself, with the
* exception of any RMPP header. Additional segment buffer space allocated
* beyond data_len is padding.
*/
struct ib_mad_send_buf {
struct ib_mad_send_buf *next;
void *mad;
struct ib_mad_agent *mad_agent;
struct ib_ah *ah;
void *context[2];
int hdr_len;
int data_len;
int seg_count;
int seg_size;
int seg_rmpp_size;
int timeout_ms;
int retries;
};
/**
* ib_response_mad - Returns if the specified MAD has been generated in
* response to a sent request or trap.
*/
int ib_response_mad(const struct ib_mad_hdr *hdr);
/**
* ib_get_rmpp_resptime - Returns the RMPP response time.
* @rmpp_hdr: An RMPP header.
*/
static inline u8 ib_get_rmpp_resptime(struct ib_rmpp_hdr *rmpp_hdr)
{
return rmpp_hdr->rmpp_rtime_flags >> 3;
}
/**
* ib_get_rmpp_flags - Returns the RMPP flags.
* @rmpp_hdr: An RMPP header.
*/
static inline u8 ib_get_rmpp_flags(struct ib_rmpp_hdr *rmpp_hdr)
{
return rmpp_hdr->rmpp_rtime_flags & 0x7;
}
/**
* ib_set_rmpp_resptime - Sets the response time in an RMPP header.
* @rmpp_hdr: An RMPP header.
* @rtime: The response time to set.
*/
static inline void ib_set_rmpp_resptime(struct ib_rmpp_hdr *rmpp_hdr, u8 rtime)
{
rmpp_hdr->rmpp_rtime_flags = ib_get_rmpp_flags(rmpp_hdr) | (rtime << 3);
}
/**
* ib_set_rmpp_flags - Sets the flags in an RMPP header.
* @rmpp_hdr: An RMPP header.
* @flags: The flags to set.
*/
static inline void ib_set_rmpp_flags(struct ib_rmpp_hdr *rmpp_hdr, u8 flags)
{
rmpp_hdr->rmpp_rtime_flags = (rmpp_hdr->rmpp_rtime_flags & 0xF8) |
(flags & 0x7);
}
struct ib_mad_agent;
struct ib_mad_send_wc;
struct ib_mad_recv_wc;
/**
* ib_mad_send_handler - callback handler for a sent MAD.
* @mad_agent: MAD agent that sent the MAD.
* @mad_send_wc: Send work completion information on the sent MAD.
*/
typedef void (*ib_mad_send_handler)(struct ib_mad_agent *mad_agent,
struct ib_mad_send_wc *mad_send_wc);
/**
* ib_mad_snoop_handler - Callback handler for snooping sent MADs.
* @mad_agent: MAD agent that snooped the MAD.
* @send_buf: send MAD data buffer.
* @mad_send_wc: Work completion information on the sent MAD. Valid
* only for snooping that occurs on a send completion.
*
* Clients snooping MADs should not modify data referenced by the @send_buf
* or @mad_send_wc.
*/
typedef void (*ib_mad_snoop_handler)(struct ib_mad_agent *mad_agent,
struct ib_mad_send_buf *send_buf,
struct ib_mad_send_wc *mad_send_wc);
/**
* ib_mad_recv_handler - callback handler for a received MAD.
* @mad_agent: MAD agent requesting the received MAD.
* @send_buf: Send buffer if found, else NULL
* @mad_recv_wc: Received work completion information on the received MAD.
*
* MADs received in response to a send request operation will be handed to
* the user before the send operation completes. All data buffers given
* to registered agents through this routine are owned by the receiving
* client, except for snooping agents. Clients snooping MADs should not
* modify the data referenced by @mad_recv_wc.
*/
typedef void (*ib_mad_recv_handler)(struct ib_mad_agent *mad_agent,
struct ib_mad_send_buf *send_buf,
struct ib_mad_recv_wc *mad_recv_wc);
/**
* ib_mad_agent - Used to track MAD registration with the access layer.
* @device: Reference to device registration is on.
* @qp: Reference to QP used for sending and receiving MADs.
* @mr: Memory region for system memory usable for DMA.
* @recv_handler: Callback handler for a received MAD.
* @send_handler: Callback handler for a sent MAD.
* @snoop_handler: Callback handler for snooped sent MADs.
* @context: User-specified context associated with this registration.
* @hi_tid: Access layer assigned transaction ID for this client.
* Unsolicited MADs sent by this client will have the upper 32-bits
* of their TID set to this value.
* @flags: registration flags
* @port_num: Port number on which QP is registered
* @rmpp_version: If set, indicates the RMPP version used by this agent.
*/
enum {
IB_MAD_USER_RMPP = IB_USER_MAD_USER_RMPP,
};
struct ib_mad_agent {
struct ib_device *device;
struct ib_qp *qp;
ib_mad_recv_handler recv_handler;
ib_mad_send_handler send_handler;
ib_mad_snoop_handler snoop_handler;
void *context;
u32 hi_tid;
u32 flags;
u8 port_num;
u8 rmpp_version;
void *security;
bool smp_allowed;
bool lsm_nb_reg;
struct notifier_block lsm_nb;
};
/**
* ib_mad_send_wc - MAD send completion information.
* @send_buf: Send MAD data buffer associated with the send MAD request.
* @status: Completion status.
* @vendor_err: Optional vendor error information returned with a failed
* request.
*/
struct ib_mad_send_wc {
struct ib_mad_send_buf *send_buf;
enum ib_wc_status status;
u32 vendor_err;
};
/**
* ib_mad_recv_buf - received MAD buffer information.
* @list: Reference to next data buffer for a received RMPP MAD.
* @grh: References a data buffer containing the global route header.
* The data refereced by this buffer is only valid if the GRH is
* valid.
* @mad: References the start of the received MAD.
*/
struct ib_mad_recv_buf {
struct list_head list;
struct ib_grh *grh;
union {
struct ib_mad *mad;
struct opa_mad *opa_mad;
};
};
/**
* ib_mad_recv_wc - received MAD information.
* @wc: Completion information for the received data.
* @recv_buf: Specifies the location of the received data buffer(s).
* @rmpp_list: Specifies a list of RMPP reassembled received MAD buffers.
* @mad_len: The length of the received MAD, without duplicated headers.
* @mad_seg_size: The size of individual MAD segments
*
* For received response, the wr_id contains a pointer to the ib_mad_send_buf
* for the corresponding send request.
*/
struct ib_mad_recv_wc {
struct ib_wc *wc;
struct ib_mad_recv_buf recv_buf;
struct list_head rmpp_list;
int mad_len;
size_t mad_seg_size;
};
/**
* ib_mad_reg_req - MAD registration request
* @mgmt_class: Indicates which management class of MADs should be receive
* by the caller. This field is only required if the user wishes to
* receive unsolicited MADs, otherwise it should be 0.
* @mgmt_class_version: Indicates which version of MADs for the given
* management class to receive.
* @oui: Indicates IEEE OUI when mgmt_class is a vendor class
* in the range from 0x30 to 0x4f. Otherwise not used.
* @method_mask: The caller will receive unsolicited MADs for any method
* where @method_mask = 1.
*
*/
struct ib_mad_reg_req {
u8 mgmt_class;
u8 mgmt_class_version;
u8 oui[3];
DECLARE_BITMAP(method_mask, IB_MGMT_MAX_METHODS);
};
/**
* ib_register_mad_agent - Register to send/receive MADs.
* @device: The device to register with.
* @port_num: The port on the specified device to use.
* @qp_type: Specifies which QP to access. Must be either
* IB_QPT_SMI or IB_QPT_GSI.
* @mad_reg_req: Specifies which unsolicited MADs should be received
* by the caller. This parameter may be NULL if the caller only
* wishes to receive solicited responses.
* @rmpp_version: If set, indicates that the client will send
* and receive MADs that contain the RMPP header for the given version.
* If set to 0, indicates that RMPP is not used by this client.
* @send_handler: The completion callback routine invoked after a send
* request has completed.
* @recv_handler: The completion callback routine invoked for a received
* MAD.
* @context: User specified context associated with the registration.
* @registration_flags: Registration flags to set for this agent
*/
struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
u8 port_num,
enum ib_qp_type qp_type,
struct ib_mad_reg_req *mad_reg_req,
u8 rmpp_version,
ib_mad_send_handler send_handler,
ib_mad_recv_handler recv_handler,
void *context,
u32 registration_flags);
enum ib_mad_snoop_flags {
/*IB_MAD_SNOOP_POSTED_SENDS = 1,*/
/*IB_MAD_SNOOP_RMPP_SENDS = (1<<1),*/
IB_MAD_SNOOP_SEND_COMPLETIONS = (1<<2),
/*IB_MAD_SNOOP_RMPP_SEND_COMPLETIONS = (1<<3),*/
IB_MAD_SNOOP_RECVS = (1<<4)
/*IB_MAD_SNOOP_RMPP_RECVS = (1<<5),*/
/*IB_MAD_SNOOP_REDIRECTED_QPS = (1<<6)*/
};
/**
* ib_register_mad_snoop - Register to snoop sent and received MADs.
* @device: The device to register with.
* @port_num: The port on the specified device to use.
* @qp_type: Specifies which QP traffic to snoop. Must be either
* IB_QPT_SMI or IB_QPT_GSI.
* @mad_snoop_flags: Specifies information where snooping occurs.
* @send_handler: The callback routine invoked for a snooped send.
* @recv_handler: The callback routine invoked for a snooped receive.
* @context: User specified context associated with the registration.
*/
struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device,
u8 port_num,
enum ib_qp_type qp_type,
int mad_snoop_flags,
ib_mad_snoop_handler snoop_handler,
ib_mad_recv_handler recv_handler,
void *context);
/**
* ib_unregister_mad_agent - Unregisters a client from using MAD services.
* @mad_agent: Corresponding MAD registration request to deregister.
*
* After invoking this routine, MAD services are no longer usable by the
* client on the associated QP.
*/
void ib_unregister_mad_agent(struct ib_mad_agent *mad_agent);
/**
* ib_post_send_mad - Posts MAD(s) to the send queue of the QP associated
* with the registered client.
* @send_buf: Specifies the information needed to send the MAD(s).
* @bad_send_buf: Specifies the MAD on which an error was encountered. This
* parameter is optional if only a single MAD is posted.
*
* Sent MADs are not guaranteed to complete in the order that they were posted.
*
* If the MAD requires RMPP, the data buffer should contain a single copy
* of the common MAD, RMPP, and class specific headers, followed by the class
* defined data. If the class defined data would not divide evenly into
* RMPP segments, then space must be allocated at the end of the referenced
* buffer for any required padding. To indicate the amount of class defined
* data being transferred, the paylen_newwin field in the RMPP header should
* be set to the size of the class specific header plus the amount of class
* defined data being transferred. The paylen_newwin field should be
* specified in network-byte order.
*/
int ib_post_send_mad(struct ib_mad_send_buf *send_buf,
struct ib_mad_send_buf **bad_send_buf);
/**
* ib_free_recv_mad - Returns data buffers used to receive a MAD.
* @mad_recv_wc: Work completion information for a received MAD.
*
* Clients receiving MADs through their ib_mad_recv_handler must call this
* routine to return the work completion buffers to the access layer.
*/
void ib_free_recv_mad(struct ib_mad_recv_wc *mad_recv_wc);
/**
* ib_cancel_mad - Cancels an outstanding send MAD operation.
* @mad_agent: Specifies the registration associated with sent MAD.
* @send_buf: Indicates the MAD to cancel.
*
* MADs will be returned to the user through the corresponding
* ib_mad_send_handler.
*/
void ib_cancel_mad(struct ib_mad_agent *mad_agent,
struct ib_mad_send_buf *send_buf);
/**
* ib_modify_mad - Modifies an outstanding send MAD operation.
* @mad_agent: Specifies the registration associated with sent MAD.
* @send_buf: Indicates the MAD to modify.
* @timeout_ms: New timeout value for sent MAD.
*
* This call will reset the timeout value for a sent MAD to the specified
* value.
*/
int ib_modify_mad(struct ib_mad_agent *mad_agent,
struct ib_mad_send_buf *send_buf, u32 timeout_ms);
/**
* ib_redirect_mad_qp - Registers a QP for MAD services.
* @qp: Reference to a QP that requires MAD services.
* @rmpp_version: If set, indicates that the client will send
* and receive MADs that contain the RMPP header for the given version.
* If set to 0, indicates that RMPP is not used by this client.
* @send_handler: The completion callback routine invoked after a send
* request has completed.
* @recv_handler: The completion callback routine invoked for a received
* MAD.
* @context: User specified context associated with the registration.
*
* Use of this call allows clients to use MAD services, such as RMPP,
* on user-owned QPs. After calling this routine, users may send
* MADs on the specified QP by calling ib_mad_post_send.
*/
struct ib_mad_agent *ib_redirect_mad_qp(struct ib_qp *qp,
u8 rmpp_version,
ib_mad_send_handler send_handler,
ib_mad_recv_handler recv_handler,
void *context);
/**
* ib_process_mad_wc - Processes a work completion associated with a
* MAD sent or received on a redirected QP.
* @mad_agent: Specifies the registered MAD service using the redirected QP.
* @wc: References a work completion associated with a sent or received
* MAD segment.
*
* This routine is used to complete or continue processing on a MAD request.
* If the work completion is associated with a send operation, calling
* this routine is required to continue an RMPP transfer or to wait for a
* corresponding response, if it is a request. If the work completion is
* associated with a receive operation, calling this routine is required to
* process an inbound or outbound RMPP transfer, or to match a response MAD
* with its corresponding request.
*/
int ib_process_mad_wc(struct ib_mad_agent *mad_agent,
struct ib_wc *wc);
/**
* ib_create_send_mad - Allocate and initialize a data buffer and work request
* for sending a MAD.
* @mad_agent: Specifies the registered MAD service to associate with the MAD.
* @remote_qpn: Specifies the QPN of the receiving node.
* @pkey_index: Specifies which PKey the MAD will be sent using. This field
* is valid only if the remote_qpn is QP 1.
* @rmpp_active: Indicates if the send will enable RMPP.
* @hdr_len: Indicates the size of the data header of the MAD. This length
* should include the common MAD header, RMPP header, plus any class
* specific header.
* @data_len: Indicates the size of any user-transferred data. The call will
* automatically adjust the allocated buffer size to account for any
* additional padding that may be necessary.
* @gfp_mask: GFP mask used for the memory allocation.
* @base_version: Base Version of this MAD
*
* This routine allocates a MAD for sending. The returned MAD send buffer
* will reference a data buffer usable for sending a MAD, along
* with an initialized work request structure. Users may modify the returned
* MAD data buffer before posting the send.
*
* The returned MAD header, class specific headers, and any padding will be
* cleared. Users are responsible for initializing the common MAD header,
* any class specific header, and MAD data area.
* If @rmpp_active is set, the RMPP header will be initialized for sending.
*/
struct ib_mad_send_buf *ib_create_send_mad(struct ib_mad_agent *mad_agent,
u32 remote_qpn, u16 pkey_index,
int rmpp_active,
int hdr_len, int data_len,
gfp_t gfp_mask,
u8 base_version);
/**
* ib_is_mad_class_rmpp - returns whether given management class
* supports RMPP.
* @mgmt_class: management class
*
* This routine returns whether the management class supports RMPP.
*/
int ib_is_mad_class_rmpp(u8 mgmt_class);
/**
* ib_get_mad_data_offset - returns the data offset for a given
* management class.
* @mgmt_class: management class
*
* This routine returns the data offset in the MAD for the management
* class requested.
*/
int ib_get_mad_data_offset(u8 mgmt_class);
/**
* ib_get_rmpp_segment - returns the data buffer for a given RMPP segment.
* @send_buf: Previously allocated send data buffer.
* @seg_num: number of segment to return
*
* This routine returns a pointer to the data buffer of an RMPP MAD.
* Users must provide synchronization to @send_buf around this call.
*/
void *ib_get_rmpp_segment(struct ib_mad_send_buf *send_buf, int seg_num);
/**
* ib_free_send_mad - Returns data buffers used to send a MAD.
* @send_buf: Previously allocated send data buffer.
*/
void ib_free_send_mad(struct ib_mad_send_buf *send_buf);
/**
* ib_mad_kernel_rmpp_agent - Returns if the agent is performing RMPP.
* @agent: the agent in question
* @return: true if agent is performing rmpp, false otherwise.
*/
int ib_mad_kernel_rmpp_agent(const struct ib_mad_agent *agent);
#endif /* IB_MAD_H */
+55
View File
@@ -0,0 +1,55 @@
/*
* Copyright (c) 2005-2006 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_USER_MARSHALL_H)
#define IB_USER_MARSHALL_H
#include <rdma/ib_verbs.h>
#include <rdma/ib_sa.h>
#include <rdma/ib_user_verbs.h>
#include <rdma/ib_user_sa.h>
void ib_copy_qp_attr_to_user(struct ib_device *device,
struct ib_uverbs_qp_attr *dst,
struct ib_qp_attr *src);
void ib_copy_ah_attr_to_user(struct ib_device *device,
struct ib_uverbs_ah_attr *dst,
struct rdma_ah_attr *src);
void ib_copy_path_rec_to_user(struct ib_user_path_rec *dst,
struct sa_path_rec *src);
void ib_copy_path_rec_from_user(struct sa_path_rec *dst,
struct ib_user_path_rec *src);
#endif /* IB_USER_MARSHALL_H */
+311
View File
@@ -0,0 +1,311 @@
/*
* Copyright (c) 2004 Topspin Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef IB_PACK_H
#define IB_PACK_H
#include <rdma/ib_verbs.h>
#include <uapi/linux/if_ether.h>
enum {
IB_LRH_BYTES = 8,
IB_ETH_BYTES = 14,
IB_VLAN_BYTES = 4,
IB_GRH_BYTES = 40,
IB_IP4_BYTES = 20,
IB_UDP_BYTES = 8,
IB_BTH_BYTES = 12,
IB_DETH_BYTES = 8,
IB_EXT_ATOMICETH_BYTES = 28,
IB_EXT_XRC_BYTES = 4,
IB_ICRC_BYTES = 4
};
struct ib_field {
size_t struct_offset_bytes;
size_t struct_size_bytes;
int offset_words;
int offset_bits;
int size_bits;
char *field_name;
};
#define RESERVED \
.field_name = "reserved"
/*
* This macro cleans up the definitions of constants for BTH opcodes.
* It is used to define constants such as IB_OPCODE_UD_SEND_ONLY,
* which becomes IB_OPCODE_UD + IB_OPCODE_SEND_ONLY, and this gives
* the correct value.
*
* In short, user code should use the constants defined using the
* macro rather than worrying about adding together other constants.
*/
#define IB_OPCODE(transport, op) \
IB_OPCODE_ ## transport ## _ ## op = \
IB_OPCODE_ ## transport + IB_OPCODE_ ## op
enum {
/* transport types -- just used to define real constants */
IB_OPCODE_RC = 0x00,
IB_OPCODE_UC = 0x20,
IB_OPCODE_RD = 0x40,
IB_OPCODE_UD = 0x60,
/* per IBTA 1.3 vol 1 Table 38, A10.3.2 */
IB_OPCODE_CNP = 0x80,
/* Manufacturer specific */
IB_OPCODE_MSP = 0xe0,
/* operations -- just used to define real constants */
IB_OPCODE_SEND_FIRST = 0x00,
IB_OPCODE_SEND_MIDDLE = 0x01,
IB_OPCODE_SEND_LAST = 0x02,
IB_OPCODE_SEND_LAST_WITH_IMMEDIATE = 0x03,
IB_OPCODE_SEND_ONLY = 0x04,
IB_OPCODE_SEND_ONLY_WITH_IMMEDIATE = 0x05,
IB_OPCODE_RDMA_WRITE_FIRST = 0x06,
IB_OPCODE_RDMA_WRITE_MIDDLE = 0x07,
IB_OPCODE_RDMA_WRITE_LAST = 0x08,
IB_OPCODE_RDMA_WRITE_LAST_WITH_IMMEDIATE = 0x09,
IB_OPCODE_RDMA_WRITE_ONLY = 0x0a,
IB_OPCODE_RDMA_WRITE_ONLY_WITH_IMMEDIATE = 0x0b,
IB_OPCODE_RDMA_READ_REQUEST = 0x0c,
IB_OPCODE_RDMA_READ_RESPONSE_FIRST = 0x0d,
IB_OPCODE_RDMA_READ_RESPONSE_MIDDLE = 0x0e,
IB_OPCODE_RDMA_READ_RESPONSE_LAST = 0x0f,
IB_OPCODE_RDMA_READ_RESPONSE_ONLY = 0x10,
IB_OPCODE_ACKNOWLEDGE = 0x11,
IB_OPCODE_ATOMIC_ACKNOWLEDGE = 0x12,
IB_OPCODE_COMPARE_SWAP = 0x13,
IB_OPCODE_FETCH_ADD = 0x14,
/* opcode 0x15 is reserved */
IB_OPCODE_SEND_LAST_WITH_INVALIDATE = 0x16,
IB_OPCODE_SEND_ONLY_WITH_INVALIDATE = 0x17,
/* real constants follow -- see comment about above IB_OPCODE()
macro for more details */
/* RC */
IB_OPCODE(RC, SEND_FIRST),
IB_OPCODE(RC, SEND_MIDDLE),
IB_OPCODE(RC, SEND_LAST),
IB_OPCODE(RC, SEND_LAST_WITH_IMMEDIATE),
IB_OPCODE(RC, SEND_ONLY),
IB_OPCODE(RC, SEND_ONLY_WITH_IMMEDIATE),
IB_OPCODE(RC, RDMA_WRITE_FIRST),
IB_OPCODE(RC, RDMA_WRITE_MIDDLE),
IB_OPCODE(RC, RDMA_WRITE_LAST),
IB_OPCODE(RC, RDMA_WRITE_LAST_WITH_IMMEDIATE),
IB_OPCODE(RC, RDMA_WRITE_ONLY),
IB_OPCODE(RC, RDMA_WRITE_ONLY_WITH_IMMEDIATE),
IB_OPCODE(RC, RDMA_READ_REQUEST),
IB_OPCODE(RC, RDMA_READ_RESPONSE_FIRST),
IB_OPCODE(RC, RDMA_READ_RESPONSE_MIDDLE),
IB_OPCODE(RC, RDMA_READ_RESPONSE_LAST),
IB_OPCODE(RC, RDMA_READ_RESPONSE_ONLY),
IB_OPCODE(RC, ACKNOWLEDGE),
IB_OPCODE(RC, ATOMIC_ACKNOWLEDGE),
IB_OPCODE(RC, COMPARE_SWAP),
IB_OPCODE(RC, FETCH_ADD),
IB_OPCODE(RC, SEND_LAST_WITH_INVALIDATE),
IB_OPCODE(RC, SEND_ONLY_WITH_INVALIDATE),
/* UC */
IB_OPCODE(UC, SEND_FIRST),
IB_OPCODE(UC, SEND_MIDDLE),
IB_OPCODE(UC, SEND_LAST),
IB_OPCODE(UC, SEND_LAST_WITH_IMMEDIATE),
IB_OPCODE(UC, SEND_ONLY),
IB_OPCODE(UC, SEND_ONLY_WITH_IMMEDIATE),
IB_OPCODE(UC, RDMA_WRITE_FIRST),
IB_OPCODE(UC, RDMA_WRITE_MIDDLE),
IB_OPCODE(UC, RDMA_WRITE_LAST),
IB_OPCODE(UC, RDMA_WRITE_LAST_WITH_IMMEDIATE),
IB_OPCODE(UC, RDMA_WRITE_ONLY),
IB_OPCODE(UC, RDMA_WRITE_ONLY_WITH_IMMEDIATE),
/* RD */
IB_OPCODE(RD, SEND_FIRST),
IB_OPCODE(RD, SEND_MIDDLE),
IB_OPCODE(RD, SEND_LAST),
IB_OPCODE(RD, SEND_LAST_WITH_IMMEDIATE),
IB_OPCODE(RD, SEND_ONLY),
IB_OPCODE(RD, SEND_ONLY_WITH_IMMEDIATE),
IB_OPCODE(RD, RDMA_WRITE_FIRST),
IB_OPCODE(RD, RDMA_WRITE_MIDDLE),
IB_OPCODE(RD, RDMA_WRITE_LAST),
IB_OPCODE(RD, RDMA_WRITE_LAST_WITH_IMMEDIATE),
IB_OPCODE(RD, RDMA_WRITE_ONLY),
IB_OPCODE(RD, RDMA_WRITE_ONLY_WITH_IMMEDIATE),
IB_OPCODE(RD, RDMA_READ_REQUEST),
IB_OPCODE(RD, RDMA_READ_RESPONSE_FIRST),
IB_OPCODE(RD, RDMA_READ_RESPONSE_MIDDLE),
IB_OPCODE(RD, RDMA_READ_RESPONSE_LAST),
IB_OPCODE(RD, RDMA_READ_RESPONSE_ONLY),
IB_OPCODE(RD, ACKNOWLEDGE),
IB_OPCODE(RD, ATOMIC_ACKNOWLEDGE),
IB_OPCODE(RD, COMPARE_SWAP),
IB_OPCODE(RD, FETCH_ADD),
/* UD */
IB_OPCODE(UD, SEND_ONLY),
IB_OPCODE(UD, SEND_ONLY_WITH_IMMEDIATE)
};
enum {
IB_LNH_RAW = 0,
IB_LNH_IP = 1,
IB_LNH_IBA_LOCAL = 2,
IB_LNH_IBA_GLOBAL = 3
};
struct ib_unpacked_lrh {
u8 virtual_lane;
u8 link_version;
u8 service_level;
u8 link_next_header;
__be16 destination_lid;
__be16 packet_length;
__be16 source_lid;
};
struct ib_unpacked_grh {
u8 ip_version;
u8 traffic_class;
__be32 flow_label;
__be16 payload_length;
u8 next_header;
u8 hop_limit;
union ib_gid source_gid;
union ib_gid destination_gid;
};
struct ib_unpacked_bth {
u8 opcode;
u8 solicited_event;
u8 mig_req;
u8 pad_count;
u8 transport_header_version;
__be16 pkey;
__be32 destination_qpn;
u8 ack_req;
__be32 psn;
};
struct ib_unpacked_deth {
__be32 qkey;
__be32 source_qpn;
};
struct ib_unpacked_eth {
u8 dmac_h[4];
u8 dmac_l[2];
u8 smac_h[2];
u8 smac_l[4];
__be16 type;
};
struct ib_unpacked_ip4 {
u8 ver;
u8 hdr_len;
u8 tos;
__be16 tot_len;
__be16 id;
__be16 frag_off;
u8 ttl;
u8 protocol;
__sum16 check;
__be32 saddr;
__be32 daddr;
};
struct ib_unpacked_udp {
__be16 sport;
__be16 dport;
__be16 length;
__be16 csum;
};
struct ib_unpacked_vlan {
__be16 tag;
__be16 type;
};
struct ib_ud_header {
int lrh_present;
struct ib_unpacked_lrh lrh;
int eth_present;
struct ib_unpacked_eth eth;
int vlan_present;
struct ib_unpacked_vlan vlan;
int grh_present;
struct ib_unpacked_grh grh;
int ipv4_present;
struct ib_unpacked_ip4 ip4;
int udp_present;
struct ib_unpacked_udp udp;
struct ib_unpacked_bth bth;
struct ib_unpacked_deth deth;
int immediate_present;
__be32 immediate_data;
};
void ib_pack(const struct ib_field *desc,
int desc_len,
void *structure,
void *buf);
void ib_unpack(const struct ib_field *desc,
int desc_len,
void *buf,
void *structure);
__sum16 ib_ud_ip4_csum(struct ib_ud_header *header);
int ib_ud_header_init(int payload_bytes,
int lrh_present,
int eth_present,
int vlan_present,
int grh_present,
int ip_version,
int udp_present,
int immediate_present,
struct ib_ud_header *header);
int ib_ud_header_pack(struct ib_ud_header *header,
void *buf);
int ib_ud_header_unpack(void *buf,
struct ib_ud_header *header);
#endif /* IB_PACK_H */
+157
View File
@@ -0,0 +1,157 @@
/*
* Copyright (c) 2006, 2007, 2008, 2009, 2010 QLogic Corporation.
* All rights reserved.
* Copyright (c) 2005, 2006 PathScale, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_PMA_H)
#define IB_PMA_H
#include <rdma/ib_mad.h>
/*
* PMA class portinfo capability mask bits
*/
#define IB_PMA_CLASS_CAP_ALLPORTSELECT cpu_to_be16(1 << 8)
#define IB_PMA_CLASS_CAP_EXT_WIDTH cpu_to_be16(1 << 9)
#define IB_PMA_CLASS_CAP_EXT_WIDTH_NOIETF cpu_to_be16(1 << 10)
#define IB_PMA_CLASS_CAP_XMIT_WAIT cpu_to_be16(1 << 12)
#define IB_PMA_CLASS_PORT_INFO cpu_to_be16(0x0001)
#define IB_PMA_PORT_SAMPLES_CONTROL cpu_to_be16(0x0010)
#define IB_PMA_PORT_SAMPLES_RESULT cpu_to_be16(0x0011)
#define IB_PMA_PORT_COUNTERS cpu_to_be16(0x0012)
#define IB_PMA_PORT_COUNTERS_EXT cpu_to_be16(0x001D)
#define IB_PMA_PORT_SAMPLES_RESULT_EXT cpu_to_be16(0x001E)
struct ib_pma_mad {
struct ib_mad_hdr mad_hdr;
u8 reserved[40];
u8 data[192];
} __packed;
struct ib_pma_portsamplescontrol {
u8 opcode;
u8 port_select;
u8 tick;
u8 counter_width; /* resv: 7:3, counter width: 2:0 */
__be32 counter_mask0_9; /* 2, 10 3-bit fields */
__be16 counter_mask10_14; /* 1, 5 3-bit fields */
u8 sample_mechanisms;
u8 sample_status; /* only lower 2 bits */
__be64 option_mask;
__be64 vendor_mask;
__be32 sample_start;
__be32 sample_interval;
__be16 tag;
__be16 counter_select[15];
__be32 reserved1;
__be64 samples_only_option_mask;
__be32 reserved2[28];
};
struct ib_pma_portsamplesresult {
__be16 tag;
__be16 sample_status; /* only lower 2 bits */
__be32 counter[15];
};
struct ib_pma_portsamplesresult_ext {
__be16 tag;
__be16 sample_status; /* only lower 2 bits */
__be32 extended_width; /* only upper 2 bits */
__be64 counter[15];
};
struct ib_pma_portcounters {
u8 reserved;
u8 port_select;
__be16 counter_select;
__be16 symbol_error_counter;
u8 link_error_recovery_counter;
u8 link_downed_counter;
__be16 port_rcv_errors;
__be16 port_rcv_remphys_errors;
__be16 port_rcv_switch_relay_errors;
__be16 port_xmit_discards;
u8 port_xmit_constraint_errors;
u8 port_rcv_constraint_errors;
u8 reserved1;
u8 link_overrun_errors; /* LocalLink: 7:4, BufferOverrun: 3:0 */
__be16 reserved2;
__be16 vl15_dropped;
__be32 port_xmit_data;
__be32 port_rcv_data;
__be32 port_xmit_packets;
__be32 port_rcv_packets;
__be32 port_xmit_wait;
} __packed;
#define IB_PMA_SEL_SYMBOL_ERROR cpu_to_be16(0x0001)
#define IB_PMA_SEL_LINK_ERROR_RECOVERY cpu_to_be16(0x0002)
#define IB_PMA_SEL_LINK_DOWNED cpu_to_be16(0x0004)
#define IB_PMA_SEL_PORT_RCV_ERRORS cpu_to_be16(0x0008)
#define IB_PMA_SEL_PORT_RCV_REMPHYS_ERRORS cpu_to_be16(0x0010)
#define IB_PMA_SEL_PORT_XMIT_DISCARDS cpu_to_be16(0x0040)
#define IB_PMA_SEL_LOCAL_LINK_INTEGRITY_ERRORS cpu_to_be16(0x0200)
#define IB_PMA_SEL_EXCESSIVE_BUFFER_OVERRUNS cpu_to_be16(0x0400)
#define IB_PMA_SEL_PORT_VL15_DROPPED cpu_to_be16(0x0800)
#define IB_PMA_SEL_PORT_XMIT_DATA cpu_to_be16(0x1000)
#define IB_PMA_SEL_PORT_RCV_DATA cpu_to_be16(0x2000)
#define IB_PMA_SEL_PORT_XMIT_PACKETS cpu_to_be16(0x4000)
#define IB_PMA_SEL_PORT_RCV_PACKETS cpu_to_be16(0x8000)
struct ib_pma_portcounters_ext {
u8 reserved;
u8 port_select;
__be16 counter_select;
__be32 reserved1;
__be64 port_xmit_data;
__be64 port_rcv_data;
__be64 port_xmit_packets;
__be64 port_rcv_packets;
__be64 port_unicast_xmit_packets;
__be64 port_unicast_rcv_packets;
__be64 port_multicast_xmit_packets;
__be64 port_multicast_rcv_packets;
} __packed;
#define IB_PMA_SELX_PORT_XMIT_DATA cpu_to_be16(0x0001)
#define IB_PMA_SELX_PORT_RCV_DATA cpu_to_be16(0x0002)
#define IB_PMA_SELX_PORT_XMIT_PACKETS cpu_to_be16(0x0004)
#define IB_PMA_SELX_PORT_RCV_PACKETS cpu_to_be16(0x0008)
#define IB_PMA_SELX_PORT_UNI_XMIT_PACKETS cpu_to_be16(0x0010)
#define IB_PMA_SELX_PORT_UNI_RCV_PACKETS cpu_to_be16(0x0020)
#define IB_PMA_SELX_PORT_MULTI_XMIT_PACKETS cpu_to_be16(0x0040)
#define IB_PMA_SELX_PORT_MULTI_RCV_PACKETS cpu_to_be16(0x0080)
#endif /* IB_PMA_H */
+668
View File
@@ -0,0 +1,668 @@
/*
* Copyright (c) 2004 Topspin Communications. All rights reserved.
* Copyright (c) 2005 Voltaire, Inc. All rights reserved.
* Copyright (c) 2006 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef IB_SA_H
#define IB_SA_H
#include <linux/completion.h>
#include <linux/compiler.h>
#include <linux/atomic.h>
#include <linux/netdevice.h>
#include <rdma/ib_verbs.h>
#include <rdma/ib_mad.h>
#include <rdma/ib_addr.h>
#include <rdma/opa_addr.h>
enum {
IB_SA_CLASS_VERSION = 2, /* IB spec version 1.1/1.2 */
IB_SA_METHOD_GET_TABLE = 0x12,
IB_SA_METHOD_GET_TABLE_RESP = 0x92,
IB_SA_METHOD_DELETE = 0x15,
IB_SA_METHOD_DELETE_RESP = 0x95,
IB_SA_METHOD_GET_MULTI = 0x14,
IB_SA_METHOD_GET_MULTI_RESP = 0x94,
IB_SA_METHOD_GET_TRACE_TBL = 0x13
};
#define OPA_SA_CLASS_VERSION 0x80
enum {
IB_SA_ATTR_CLASS_PORTINFO = 0x01,
IB_SA_ATTR_NOTICE = 0x02,
IB_SA_ATTR_INFORM_INFO = 0x03,
IB_SA_ATTR_NODE_REC = 0x11,
IB_SA_ATTR_PORT_INFO_REC = 0x12,
IB_SA_ATTR_SL2VL_REC = 0x13,
IB_SA_ATTR_SWITCH_REC = 0x14,
IB_SA_ATTR_LINEAR_FDB_REC = 0x15,
IB_SA_ATTR_RANDOM_FDB_REC = 0x16,
IB_SA_ATTR_MCAST_FDB_REC = 0x17,
IB_SA_ATTR_SM_INFO_REC = 0x18,
IB_SA_ATTR_LINK_REC = 0x20,
IB_SA_ATTR_GUID_INFO_REC = 0x30,
IB_SA_ATTR_SERVICE_REC = 0x31,
IB_SA_ATTR_PARTITION_REC = 0x33,
IB_SA_ATTR_PATH_REC = 0x35,
IB_SA_ATTR_VL_ARB_REC = 0x36,
IB_SA_ATTR_MC_MEMBER_REC = 0x38,
IB_SA_ATTR_TRACE_REC = 0x39,
IB_SA_ATTR_MULTI_PATH_REC = 0x3a,
IB_SA_ATTR_SERVICE_ASSOC_REC = 0x3b,
IB_SA_ATTR_INFORM_INFO_REC = 0xf3
};
enum ib_sa_selector {
IB_SA_GT = 0,
IB_SA_LT = 1,
IB_SA_EQ = 2,
/*
* The meaning of "best" depends on the attribute: for
* example, for MTU best will return the largest available
* MTU, while for packet life time, best will return the
* smallest available life time.
*/
IB_SA_BEST = 3
};
/*
* There are 4 types of join states:
* FullMember, NonMember, SendOnlyNonMember, SendOnlyFullMember.
* The order corresponds to JoinState bits in MCMemberRecord.
*/
enum ib_sa_mc_join_states {
FULLMEMBER_JOIN,
NONMEMBER_JOIN,
SENDONLY_NONMEBER_JOIN,
SENDONLY_FULLMEMBER_JOIN,
NUM_JOIN_MEMBERSHIP_TYPES,
};
#define IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT BIT(12)
/*
* Structures for SA records are named "struct ib_sa_xxx_rec." No
* attempt is made to pack structures to match the physical layout of
* SA records in SA MADs; all packing and unpacking is handled by the
* SA query code.
*
* For a record with structure ib_sa_xxx_rec, the naming convention
* for the component mask value for field yyy is IB_SA_XXX_REC_YYY (we
* never use different abbreviations or otherwise change the spelling
* of xxx/yyy between ib_sa_xxx_rec.yyy and IB_SA_XXX_REC_YYY).
*
* Reserved rows are indicated with comments to help maintainability.
*/
#define IB_SA_PATH_REC_SERVICE_ID (IB_SA_COMP_MASK( 0) |\
IB_SA_COMP_MASK( 1))
#define IB_SA_PATH_REC_DGID IB_SA_COMP_MASK( 2)
#define IB_SA_PATH_REC_SGID IB_SA_COMP_MASK( 3)
#define IB_SA_PATH_REC_DLID IB_SA_COMP_MASK( 4)
#define IB_SA_PATH_REC_SLID IB_SA_COMP_MASK( 5)
#define IB_SA_PATH_REC_RAW_TRAFFIC IB_SA_COMP_MASK( 6)
/* reserved: 7 */
#define IB_SA_PATH_REC_FLOW_LABEL IB_SA_COMP_MASK( 8)
#define IB_SA_PATH_REC_HOP_LIMIT IB_SA_COMP_MASK( 9)
#define IB_SA_PATH_REC_TRAFFIC_CLASS IB_SA_COMP_MASK(10)
#define IB_SA_PATH_REC_REVERSIBLE IB_SA_COMP_MASK(11)
#define IB_SA_PATH_REC_NUMB_PATH IB_SA_COMP_MASK(12)
#define IB_SA_PATH_REC_PKEY IB_SA_COMP_MASK(13)
#define IB_SA_PATH_REC_QOS_CLASS IB_SA_COMP_MASK(14)
#define IB_SA_PATH_REC_SL IB_SA_COMP_MASK(15)
#define IB_SA_PATH_REC_MTU_SELECTOR IB_SA_COMP_MASK(16)
#define IB_SA_PATH_REC_MTU IB_SA_COMP_MASK(17)
#define IB_SA_PATH_REC_RATE_SELECTOR IB_SA_COMP_MASK(18)
#define IB_SA_PATH_REC_RATE IB_SA_COMP_MASK(19)
#define IB_SA_PATH_REC_PACKET_LIFE_TIME_SELECTOR IB_SA_COMP_MASK(20)
#define IB_SA_PATH_REC_PACKET_LIFE_TIME IB_SA_COMP_MASK(21)
#define IB_SA_PATH_REC_PREFERENCE IB_SA_COMP_MASK(22)
enum sa_path_rec_type {
SA_PATH_REC_TYPE_IB,
SA_PATH_REC_TYPE_ROCE_V1,
SA_PATH_REC_TYPE_ROCE_V2,
SA_PATH_REC_TYPE_OPA
};
struct sa_path_rec_ib {
__be16 dlid;
__be16 slid;
u8 raw_traffic;
};
/**
* struct sa_path_rec_roce - RoCE specific portion of the path record entry
* @route_resolved: When set, it indicates that this route is already
* resolved for this path record entry.
* @dmac: Destination mac address for the given DGID entry
* of the path record entry.
*/
struct sa_path_rec_roce {
bool route_resolved;
u8 dmac[ETH_ALEN];
};
struct sa_path_rec_opa {
__be32 dlid;
__be32 slid;
u8 raw_traffic;
u8 l2_8B;
u8 l2_10B;
u8 l2_9B;
u8 l2_16B;
u8 qos_type;
u8 qos_priority;
};
struct sa_path_rec {
union ib_gid dgid;
union ib_gid sgid;
__be64 service_id;
/* reserved */
__be32 flow_label;
u8 hop_limit;
u8 traffic_class;
u8 reversible;
u8 numb_path;
__be16 pkey;
__be16 qos_class;
u8 sl;
u8 mtu_selector;
u8 mtu;
u8 rate_selector;
u8 rate;
u8 packet_life_time_selector;
u8 packet_life_time;
u8 preference;
union {
struct sa_path_rec_ib ib;
struct sa_path_rec_roce roce;
struct sa_path_rec_opa opa;
};
enum sa_path_rec_type rec_type;
};
static inline enum ib_gid_type
sa_conv_pathrec_to_gid_type(struct sa_path_rec *rec)
{
switch (rec->rec_type) {
case SA_PATH_REC_TYPE_ROCE_V1:
return IB_GID_TYPE_ROCE;
case SA_PATH_REC_TYPE_ROCE_V2:
return IB_GID_TYPE_ROCE_UDP_ENCAP;
default:
return IB_GID_TYPE_IB;
}
}
static inline enum sa_path_rec_type
sa_conv_gid_to_pathrec_type(enum ib_gid_type type)
{
switch (type) {
case IB_GID_TYPE_ROCE:
return SA_PATH_REC_TYPE_ROCE_V1;
case IB_GID_TYPE_ROCE_UDP_ENCAP:
return SA_PATH_REC_TYPE_ROCE_V2;
default:
return SA_PATH_REC_TYPE_IB;
}
}
static inline void path_conv_opa_to_ib(struct sa_path_rec *ib,
struct sa_path_rec *opa)
{
if ((be32_to_cpu(opa->opa.dlid) >=
be16_to_cpu(IB_MULTICAST_LID_BASE)) ||
(be32_to_cpu(opa->opa.slid) >=
be16_to_cpu(IB_MULTICAST_LID_BASE))) {
/* Create OPA GID and zero out the LID */
ib->dgid.global.interface_id
= OPA_MAKE_ID(be32_to_cpu(opa->opa.dlid));
ib->dgid.global.subnet_prefix
= opa->dgid.global.subnet_prefix;
ib->sgid.global.interface_id
= OPA_MAKE_ID(be32_to_cpu(opa->opa.slid));
ib->dgid.global.subnet_prefix
= opa->dgid.global.subnet_prefix;
ib->ib.dlid = 0;
ib->ib.slid = 0;
} else {
ib->ib.dlid = htons(ntohl(opa->opa.dlid));
ib->ib.slid = htons(ntohl(opa->opa.slid));
}
ib->service_id = opa->service_id;
ib->ib.raw_traffic = opa->opa.raw_traffic;
}
static inline void path_conv_ib_to_opa(struct sa_path_rec *opa,
struct sa_path_rec *ib)
{
__be32 slid, dlid;
if ((ib_is_opa_gid(&ib->sgid)) ||
(ib_is_opa_gid(&ib->dgid))) {
slid = htonl(opa_get_lid_from_gid(&ib->sgid));
dlid = htonl(opa_get_lid_from_gid(&ib->dgid));
} else {
slid = htonl(ntohs(ib->ib.slid));
dlid = htonl(ntohs(ib->ib.dlid));
}
opa->opa.slid = slid;
opa->opa.dlid = dlid;
opa->service_id = ib->service_id;
opa->opa.raw_traffic = ib->ib.raw_traffic;
}
/* Convert from OPA to IB path record */
static inline void sa_convert_path_opa_to_ib(struct sa_path_rec *dest,
struct sa_path_rec *src)
{
if (src->rec_type != SA_PATH_REC_TYPE_OPA)
return;
*dest = *src;
dest->rec_type = SA_PATH_REC_TYPE_IB;
path_conv_opa_to_ib(dest, src);
}
/* Convert from IB to OPA path record */
static inline void sa_convert_path_ib_to_opa(struct sa_path_rec *dest,
struct sa_path_rec *src)
{
if (src->rec_type != SA_PATH_REC_TYPE_IB)
return;
/* Do a structure copy and overwrite the relevant fields */
*dest = *src;
dest->rec_type = SA_PATH_REC_TYPE_OPA;
path_conv_ib_to_opa(dest, src);
}
#define IB_SA_MCMEMBER_REC_MGID IB_SA_COMP_MASK( 0)
#define IB_SA_MCMEMBER_REC_PORT_GID IB_SA_COMP_MASK( 1)
#define IB_SA_MCMEMBER_REC_QKEY IB_SA_COMP_MASK( 2)
#define IB_SA_MCMEMBER_REC_MLID IB_SA_COMP_MASK( 3)
#define IB_SA_MCMEMBER_REC_MTU_SELECTOR IB_SA_COMP_MASK( 4)
#define IB_SA_MCMEMBER_REC_MTU IB_SA_COMP_MASK( 5)
#define IB_SA_MCMEMBER_REC_TRAFFIC_CLASS IB_SA_COMP_MASK( 6)
#define IB_SA_MCMEMBER_REC_PKEY IB_SA_COMP_MASK( 7)
#define IB_SA_MCMEMBER_REC_RATE_SELECTOR IB_SA_COMP_MASK( 8)
#define IB_SA_MCMEMBER_REC_RATE IB_SA_COMP_MASK( 9)
#define IB_SA_MCMEMBER_REC_PACKET_LIFE_TIME_SELECTOR IB_SA_COMP_MASK(10)
#define IB_SA_MCMEMBER_REC_PACKET_LIFE_TIME IB_SA_COMP_MASK(11)
#define IB_SA_MCMEMBER_REC_SL IB_SA_COMP_MASK(12)
#define IB_SA_MCMEMBER_REC_FLOW_LABEL IB_SA_COMP_MASK(13)
#define IB_SA_MCMEMBER_REC_HOP_LIMIT IB_SA_COMP_MASK(14)
#define IB_SA_MCMEMBER_REC_SCOPE IB_SA_COMP_MASK(15)
#define IB_SA_MCMEMBER_REC_JOIN_STATE IB_SA_COMP_MASK(16)
#define IB_SA_MCMEMBER_REC_PROXY_JOIN IB_SA_COMP_MASK(17)
struct ib_sa_mcmember_rec {
union ib_gid mgid;
union ib_gid port_gid;
__be32 qkey;
__be16 mlid;
u8 mtu_selector;
u8 mtu;
u8 traffic_class;
__be16 pkey;
u8 rate_selector;
u8 rate;
u8 packet_life_time_selector;
u8 packet_life_time;
u8 sl;
__be32 flow_label;
u8 hop_limit;
u8 scope;
u8 join_state;
u8 proxy_join;
};
/* Service Record Component Mask Sec 15.2.5.14 Ver 1.1 */
#define IB_SA_SERVICE_REC_SERVICE_ID IB_SA_COMP_MASK( 0)
#define IB_SA_SERVICE_REC_SERVICE_GID IB_SA_COMP_MASK( 1)
#define IB_SA_SERVICE_REC_SERVICE_PKEY IB_SA_COMP_MASK( 2)
/* reserved: 3 */
#define IB_SA_SERVICE_REC_SERVICE_LEASE IB_SA_COMP_MASK( 4)
#define IB_SA_SERVICE_REC_SERVICE_KEY IB_SA_COMP_MASK( 5)
#define IB_SA_SERVICE_REC_SERVICE_NAME IB_SA_COMP_MASK( 6)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_0 IB_SA_COMP_MASK( 7)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_1 IB_SA_COMP_MASK( 8)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_2 IB_SA_COMP_MASK( 9)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_3 IB_SA_COMP_MASK(10)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_4 IB_SA_COMP_MASK(11)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_5 IB_SA_COMP_MASK(12)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_6 IB_SA_COMP_MASK(13)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_7 IB_SA_COMP_MASK(14)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_8 IB_SA_COMP_MASK(15)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_9 IB_SA_COMP_MASK(16)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_10 IB_SA_COMP_MASK(17)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_11 IB_SA_COMP_MASK(18)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_12 IB_SA_COMP_MASK(19)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_13 IB_SA_COMP_MASK(20)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_14 IB_SA_COMP_MASK(21)
#define IB_SA_SERVICE_REC_SERVICE_DATA8_15 IB_SA_COMP_MASK(22)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_0 IB_SA_COMP_MASK(23)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_1 IB_SA_COMP_MASK(24)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_2 IB_SA_COMP_MASK(25)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_3 IB_SA_COMP_MASK(26)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_4 IB_SA_COMP_MASK(27)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_5 IB_SA_COMP_MASK(28)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_6 IB_SA_COMP_MASK(29)
#define IB_SA_SERVICE_REC_SERVICE_DATA16_7 IB_SA_COMP_MASK(30)
#define IB_SA_SERVICE_REC_SERVICE_DATA32_0 IB_SA_COMP_MASK(31)
#define IB_SA_SERVICE_REC_SERVICE_DATA32_1 IB_SA_COMP_MASK(32)
#define IB_SA_SERVICE_REC_SERVICE_DATA32_2 IB_SA_COMP_MASK(33)
#define IB_SA_SERVICE_REC_SERVICE_DATA32_3 IB_SA_COMP_MASK(34)
#define IB_SA_SERVICE_REC_SERVICE_DATA64_0 IB_SA_COMP_MASK(35)
#define IB_SA_SERVICE_REC_SERVICE_DATA64_1 IB_SA_COMP_MASK(36)
#define IB_DEFAULT_SERVICE_LEASE 0xFFFFFFFF
struct ib_sa_service_rec {
u64 id;
union ib_gid gid;
__be16 pkey;
/* reserved */
u32 lease;
u8 key[16];
u8 name[64];
u8 data8[16];
u16 data16[8];
u32 data32[4];
u64 data64[2];
};
#define IB_SA_GUIDINFO_REC_LID IB_SA_COMP_MASK(0)
#define IB_SA_GUIDINFO_REC_BLOCK_NUM IB_SA_COMP_MASK(1)
#define IB_SA_GUIDINFO_REC_RES1 IB_SA_COMP_MASK(2)
#define IB_SA_GUIDINFO_REC_RES2 IB_SA_COMP_MASK(3)
#define IB_SA_GUIDINFO_REC_GID0 IB_SA_COMP_MASK(4)
#define IB_SA_GUIDINFO_REC_GID1 IB_SA_COMP_MASK(5)
#define IB_SA_GUIDINFO_REC_GID2 IB_SA_COMP_MASK(6)
#define IB_SA_GUIDINFO_REC_GID3 IB_SA_COMP_MASK(7)
#define IB_SA_GUIDINFO_REC_GID4 IB_SA_COMP_MASK(8)
#define IB_SA_GUIDINFO_REC_GID5 IB_SA_COMP_MASK(9)
#define IB_SA_GUIDINFO_REC_GID6 IB_SA_COMP_MASK(10)
#define IB_SA_GUIDINFO_REC_GID7 IB_SA_COMP_MASK(11)
struct ib_sa_guidinfo_rec {
__be16 lid;
u8 block_num;
/* reserved */
u8 res1;
__be32 res2;
u8 guid_info_list[64];
};
struct ib_sa_client {
atomic_t users;
struct completion comp;
};
/**
* ib_sa_register_client - Register an SA client.
*/
void ib_sa_register_client(struct ib_sa_client *client);
/**
* ib_sa_unregister_client - Deregister an SA client.
* @client: Client object to deregister.
*/
void ib_sa_unregister_client(struct ib_sa_client *client);
struct ib_sa_query;
void ib_sa_cancel_query(int id, struct ib_sa_query *query);
int ib_sa_path_rec_get(struct ib_sa_client *client,
struct ib_device *device, u8 port_num,
struct sa_path_rec *rec,
ib_sa_comp_mask comp_mask,
int timeout_ms, gfp_t gfp_mask,
void (*callback)(int status,
struct sa_path_rec *resp,
void *context),
void *context,
struct ib_sa_query **query);
int ib_sa_service_rec_query(struct ib_sa_client *client,
struct ib_device *device, u8 port_num,
u8 method,
struct ib_sa_service_rec *rec,
ib_sa_comp_mask comp_mask,
int timeout_ms, gfp_t gfp_mask,
void (*callback)(int status,
struct ib_sa_service_rec *resp,
void *context),
void *context,
struct ib_sa_query **sa_query);
struct ib_sa_multicast {
struct ib_sa_mcmember_rec rec;
ib_sa_comp_mask comp_mask;
int (*callback)(int status,
struct ib_sa_multicast *multicast);
void *context;
};
/**
* ib_sa_join_multicast - Initiates a join request to the specified multicast
* group.
* @client: SA client
* @device: Device associated with the multicast group.
* @port_num: Port on the specified device to associate with the multicast
* group.
* @rec: SA multicast member record specifying group attributes.
* @comp_mask: Component mask indicating which group attributes of %rec are
* valid.
* @gfp_mask: GFP mask for memory allocations.
* @callback: User callback invoked once the join operation completes.
* @context: User specified context stored with the ib_sa_multicast structure.
*
* This call initiates a multicast join request with the SA for the specified
* multicast group. If the join operation is started successfully, it returns
* an ib_sa_multicast structure that is used to track the multicast operation.
* Users must free this structure by calling ib_free_multicast, even if the
* join operation later fails. (The callback status is non-zero.)
*
* If the join operation fails; status will be non-zero, with the following
* failures possible:
* -ETIMEDOUT: The request timed out.
* -EIO: An error occurred sending the query.
* -EINVAL: The MCMemberRecord values differed from the existing group's.
* -ENETRESET: Indicates that an fatal error has occurred on the multicast
* group, and the user must rejoin the group to continue using it.
*/
struct ib_sa_multicast *ib_sa_join_multicast(struct ib_sa_client *client,
struct ib_device *device, u8 port_num,
struct ib_sa_mcmember_rec *rec,
ib_sa_comp_mask comp_mask, gfp_t gfp_mask,
int (*callback)(int status,
struct ib_sa_multicast
*multicast),
void *context);
/**
* ib_free_multicast - Frees the multicast tracking structure, and releases
* any reference on the multicast group.
* @multicast: Multicast tracking structure allocated by ib_join_multicast.
*
* This call blocks until the multicast identifier is destroyed. It may
* not be called from within the multicast callback; however, returning a non-
* zero value from the callback will result in destroying the multicast
* tracking structure.
*/
void ib_sa_free_multicast(struct ib_sa_multicast *multicast);
/**
* ib_get_mcmember_rec - Looks up a multicast member record by its MGID and
* returns it if found.
* @device: Device associated with the multicast group.
* @port_num: Port on the specified device to associate with the multicast
* group.
* @mgid: MGID of multicast group.
* @rec: Location to copy SA multicast member record.
*/
int ib_sa_get_mcmember_rec(struct ib_device *device, u8 port_num,
union ib_gid *mgid, struct ib_sa_mcmember_rec *rec);
/**
* ib_init_ah_from_mcmember - Initialize address handle attributes based on
* an SA multicast member record.
*/
int ib_init_ah_from_mcmember(struct ib_device *device, u8 port_num,
struct ib_sa_mcmember_rec *rec,
struct net_device *ndev,
enum ib_gid_type gid_type,
struct rdma_ah_attr *ah_attr);
int ib_init_ah_attr_from_path(struct ib_device *device, u8 port_num,
struct sa_path_rec *rec,
struct rdma_ah_attr *ah_attr,
const struct ib_gid_attr *sgid_attr);
/**
* ib_sa_pack_path - Conert a path record from struct ib_sa_path_rec
* to IB MAD wire format.
*/
void ib_sa_pack_path(struct sa_path_rec *rec, void *attribute);
/**
* ib_sa_unpack_path - Convert a path record from MAD format to struct
* ib_sa_path_rec.
*/
void ib_sa_unpack_path(void *attribute, struct sa_path_rec *rec);
/* Support GuidInfoRecord */
int ib_sa_guid_info_rec_query(struct ib_sa_client *client,
struct ib_device *device, u8 port_num,
struct ib_sa_guidinfo_rec *rec,
ib_sa_comp_mask comp_mask, u8 method,
int timeout_ms, gfp_t gfp_mask,
void (*callback)(int status,
struct ib_sa_guidinfo_rec *resp,
void *context),
void *context,
struct ib_sa_query **sa_query);
bool ib_sa_sendonly_fullmem_support(struct ib_sa_client *client,
struct ib_device *device,
u8 port_num);
static inline bool sa_path_is_roce(struct sa_path_rec *rec)
{
return ((rec->rec_type == SA_PATH_REC_TYPE_ROCE_V1) ||
(rec->rec_type == SA_PATH_REC_TYPE_ROCE_V2));
}
static inline bool sa_path_is_opa(struct sa_path_rec *rec)
{
return (rec->rec_type == SA_PATH_REC_TYPE_OPA);
}
static inline void sa_path_set_slid(struct sa_path_rec *rec, u32 slid)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
rec->ib.slid = cpu_to_be16(slid);
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
rec->opa.slid = cpu_to_be32(slid);
}
static inline void sa_path_set_dlid(struct sa_path_rec *rec, u32 dlid)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
rec->ib.dlid = cpu_to_be16(dlid);
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
rec->opa.dlid = cpu_to_be32(dlid);
}
static inline void sa_path_set_raw_traffic(struct sa_path_rec *rec,
u8 raw_traffic)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
rec->ib.raw_traffic = raw_traffic;
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
rec->opa.raw_traffic = raw_traffic;
}
static inline __be32 sa_path_get_slid(struct sa_path_rec *rec)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
return htonl(ntohs(rec->ib.slid));
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
return rec->opa.slid;
return 0;
}
static inline __be32 sa_path_get_dlid(struct sa_path_rec *rec)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
return htonl(ntohs(rec->ib.dlid));
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
return rec->opa.dlid;
return 0;
}
static inline u8 sa_path_get_raw_traffic(struct sa_path_rec *rec)
{
if (rec->rec_type == SA_PATH_REC_TYPE_IB)
return rec->ib.raw_traffic;
else if (rec->rec_type == SA_PATH_REC_TYPE_OPA)
return rec->opa.raw_traffic;
return 0;
}
static inline void sa_path_set_dmac(struct sa_path_rec *rec, u8 *dmac)
{
if (sa_path_is_roce(rec))
memcpy(rec->roce.dmac, dmac, ETH_ALEN);
}
static inline void sa_path_set_dmac_zero(struct sa_path_rec *rec)
{
if (sa_path_is_roce(rec))
eth_zero_addr(rec->roce.dmac);
}
static inline u8 *sa_path_get_dmac(struct sa_path_rec *rec)
{
if (sa_path_is_roce(rec))
return rec->roce.dmac;
return NULL;
}
#endif /* IB_SA_H */
+175
View File
@@ -0,0 +1,175 @@
/*
* Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved.
* Copyright (c) 2004 Infinicon Corporation. All rights reserved.
* Copyright (c) 2004 Intel Corporation. All rights reserved.
* Copyright (c) 2004 Topspin Corporation. All rights reserved.
* Copyright (c) 2004 Voltaire Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(IB_SMI_H)
#define IB_SMI_H
#include <rdma/ib_mad.h>
#define IB_SMP_DATA_SIZE 64
#define IB_SMP_MAX_PATH_HOPS 64
struct ib_smp {
u8 base_version;
u8 mgmt_class;
u8 class_version;
u8 method;
__be16 status;
u8 hop_ptr;
u8 hop_cnt;
__be64 tid;
__be16 attr_id;
__be16 resv;
__be32 attr_mod;
__be64 mkey;
__be16 dr_slid;
__be16 dr_dlid;
u8 reserved[28];
u8 data[IB_SMP_DATA_SIZE];
u8 initial_path[IB_SMP_MAX_PATH_HOPS];
u8 return_path[IB_SMP_MAX_PATH_HOPS];
} __attribute__ ((packed));
#define IB_SMP_DIRECTION cpu_to_be16(0x8000)
/* Subnet management attributes */
#define IB_SMP_ATTR_NOTICE cpu_to_be16(0x0002)
#define IB_SMP_ATTR_NODE_DESC cpu_to_be16(0x0010)
#define IB_SMP_ATTR_NODE_INFO cpu_to_be16(0x0011)
#define IB_SMP_ATTR_SWITCH_INFO cpu_to_be16(0x0012)
#define IB_SMP_ATTR_GUID_INFO cpu_to_be16(0x0014)
#define IB_SMP_ATTR_PORT_INFO cpu_to_be16(0x0015)
#define IB_SMP_ATTR_PKEY_TABLE cpu_to_be16(0x0016)
#define IB_SMP_ATTR_SL_TO_VL_TABLE cpu_to_be16(0x0017)
#define IB_SMP_ATTR_VL_ARB_TABLE cpu_to_be16(0x0018)
#define IB_SMP_ATTR_LINEAR_FORWARD_TABLE cpu_to_be16(0x0019)
#define IB_SMP_ATTR_RANDOM_FORWARD_TABLE cpu_to_be16(0x001A)
#define IB_SMP_ATTR_MCAST_FORWARD_TABLE cpu_to_be16(0x001B)
#define IB_SMP_ATTR_SM_INFO cpu_to_be16(0x0020)
#define IB_SMP_ATTR_VENDOR_DIAG cpu_to_be16(0x0030)
#define IB_SMP_ATTR_LED_INFO cpu_to_be16(0x0031)
#define IB_SMP_ATTR_VENDOR_MASK cpu_to_be16(0xFF00)
struct ib_port_info {
__be64 mkey;
__be64 gid_prefix;
__be16 lid;
__be16 sm_lid;
__be32 cap_mask;
__be16 diag_code;
__be16 mkey_lease_period;
u8 local_port_num;
u8 link_width_enabled;
u8 link_width_supported;
u8 link_width_active;
u8 linkspeed_portstate; /* 4 bits, 4 bits */
u8 portphysstate_linkdown; /* 4 bits, 4 bits */
u8 mkeyprot_resv_lmc; /* 2 bits, 3, 3 */
u8 linkspeedactive_enabled; /* 4 bits, 4 bits */
u8 neighbormtu_mastersmsl; /* 4 bits, 4 bits */
u8 vlcap_inittype; /* 4 bits, 4 bits */
u8 vl_high_limit;
u8 vl_arb_high_cap;
u8 vl_arb_low_cap;
u8 inittypereply_mtucap; /* 4 bits, 4 bits */
u8 vlstallcnt_hoqlife; /* 3 bits, 5 bits */
u8 operationalvl_pei_peo_fpi_fpo; /* 4 bits, 1, 1, 1, 1 */
__be16 mkey_violations;
__be16 pkey_violations;
__be16 qkey_violations;
u8 guid_cap;
u8 clientrereg_resv_subnetto; /* 1 bit, 2 bits, 5 */
u8 resv_resptimevalue; /* 3 bits, 5 bits */
u8 localphyerrors_overrunerrors; /* 4 bits, 4 bits */
__be16 max_credit_hint;
u8 resv;
u8 link_roundtrip_latency[3];
};
struct ib_node_info {
u8 base_version;
u8 class_version;
u8 node_type;
u8 num_ports;
__be64 sys_guid;
__be64 node_guid;
__be64 port_guid;
__be16 partition_cap;
__be16 device_id;
__be32 revision;
u8 local_port_num;
u8 vendor_id[3];
} __packed;
struct ib_vl_weight_elem {
u8 vl; /* IB: VL is low 4 bits, upper 4 bits reserved */
/* OPA: VL is low 5 bits, upper 3 bits reserved */
u8 weight;
};
static inline u8
ib_get_smp_direction(struct ib_smp *smp)
{
return ((smp->status & IB_SMP_DIRECTION) == IB_SMP_DIRECTION);
}
/*
* SM Trap/Notice numbers
*/
#define IB_NOTICE_TRAP_LLI_THRESH cpu_to_be16(129)
#define IB_NOTICE_TRAP_EBO_THRESH cpu_to_be16(130)
#define IB_NOTICE_TRAP_FLOW_UPDATE cpu_to_be16(131)
#define IB_NOTICE_TRAP_CAP_MASK_CHG cpu_to_be16(144)
#define IB_NOTICE_TRAP_SYS_GUID_CHG cpu_to_be16(145)
#define IB_NOTICE_TRAP_BAD_MKEY cpu_to_be16(256)
#define IB_NOTICE_TRAP_BAD_PKEY cpu_to_be16(257)
#define IB_NOTICE_TRAP_BAD_QKEY cpu_to_be16(258)
/*
* Other local changes flags (trap 144).
*/
#define IB_NOTICE_TRAP_LSE_CHG 0x04 /* Link Speed Enable changed */
#define IB_NOTICE_TRAP_LWE_CHG 0x02 /* Link Width Enable changed */
#define IB_NOTICE_TRAP_NODE_DESC_CHG 0x01
/*
* M_Key volation flags in dr_trunc_hop (trap 256).
*/
#define IB_NOTICE_TRAP_DR_NOTICE 0x80
#define IB_NOTICE_TRAP_DR_TRUNC 0x40
#endif /* IB_SMI_H */
+108
View File
@@ -0,0 +1,108 @@
/*
* Copyright (c) 2007 Cisco Systems. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef IB_UMEM_H
#define IB_UMEM_H
#include <linux/list.h>
#include <linux/scatterlist.h>
#include <linux/workqueue.h>
struct ib_ucontext;
struct ib_umem_odp;
struct ib_umem {
struct ib_ucontext *context;
size_t length;
unsigned long address;
int page_shift;
int writable;
int hugetlb;
struct work_struct work;
struct mm_struct *mm;
unsigned long diff;
struct ib_umem_odp *odp_data;
struct sg_table sg_head;
int nmap;
int npages;
};
/* Returns the offset of the umem start relative to the first page. */
static inline int ib_umem_offset(struct ib_umem *umem)
{
return umem->address & (BIT(umem->page_shift) - 1);
}
/* Returns the first page of an ODP umem. */
static inline unsigned long ib_umem_start(struct ib_umem *umem)
{
return umem->address - ib_umem_offset(umem);
}
/* Returns the address of the page after the last one of an ODP umem. */
static inline unsigned long ib_umem_end(struct ib_umem *umem)
{
return ALIGN(umem->address + umem->length, BIT(umem->page_shift));
}
static inline size_t ib_umem_num_pages(struct ib_umem *umem)
{
return (ib_umem_end(umem) - ib_umem_start(umem)) >> umem->page_shift;
}
#ifdef CONFIG_INFINIBAND_USER_MEM
struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
size_t size, int access, int dmasync);
void ib_umem_release(struct ib_umem *umem);
int ib_umem_page_count(struct ib_umem *umem);
int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
size_t length);
#else /* CONFIG_INFINIBAND_USER_MEM */
#include <linux/err.h>
static inline struct ib_umem *ib_umem_get(struct ib_ucontext *context,
unsigned long addr, size_t size,
int access, int dmasync) {
return ERR_PTR(-EINVAL);
}
static inline void ib_umem_release(struct ib_umem *umem) { }
static inline int ib_umem_page_count(struct ib_umem *umem) { return 0; }
static inline int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
size_t length) {
return -EINVAL;
}
#endif /* CONFIG_INFINIBAND_USER_MEM */
#endif /* IB_UMEM_H */
+175
View File
@@ -0,0 +1,175 @@
/*
* Copyright (c) 2014 Mellanox Technologies. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef IB_UMEM_ODP_H
#define IB_UMEM_ODP_H
#include <rdma/ib_umem.h>
#include <rdma/ib_verbs.h>
#include <linux/interval_tree.h>
struct umem_odp_node {
u64 __subtree_last;
struct rb_node rb;
};
struct ib_umem_odp {
/*
* An array of the pages included in the on-demand paging umem.
* Indices of pages that are currently not mapped into the device will
* contain NULL.
*/
struct page **page_list;
/*
* An array of the same size as page_list, with DMA addresses mapped
* for pages the pages in page_list. The lower two bits designate
* access permissions. See ODP_READ_ALLOWED_BIT and
* ODP_WRITE_ALLOWED_BIT.
*/
dma_addr_t *dma_list;
/*
* The umem_mutex protects the page_list and dma_list fields of an ODP
* umem, allowing only a single thread to map/unmap pages. The mutex
* also protects access to the mmu notifier counters.
*/
struct mutex umem_mutex;
void *private; /* for the HW driver to use. */
/* When false, use the notifier counter in the ucontext struct. */
bool mn_counters_active;
int notifiers_seq;
int notifiers_count;
/* A linked list of umems that don't have private mmu notifier
* counters yet. */
struct list_head no_private_counters;
struct ib_umem *umem;
/* Tree tracking */
struct umem_odp_node interval_tree;
struct completion notifier_completion;
int dying;
struct work_struct work;
};
#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem,
int access);
struct ib_umem *ib_alloc_odp_umem(struct ib_ucontext *context,
unsigned long addr,
size_t size);
void ib_umem_odp_release(struct ib_umem *umem);
/*
* The lower 2 bits of the DMA address signal the R/W permissions for
* the entry. To upgrade the permissions, provide the appropriate
* bitmask to the map_dma_pages function.
*
* Be aware that upgrading a mapped address might result in change of
* the DMA address for the page.
*/
#define ODP_READ_ALLOWED_BIT (1<<0ULL)
#define ODP_WRITE_ALLOWED_BIT (1<<1ULL)
#define ODP_DMA_ADDR_MASK (~(ODP_READ_ALLOWED_BIT | ODP_WRITE_ALLOWED_BIT))
int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 start_offset, u64 bcnt,
u64 access_mask, unsigned long current_seq);
void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 start_offset,
u64 bound);
typedef int (*umem_call_back)(struct ib_umem *item, u64 start, u64 end,
void *cookie);
/*
* Call the callback on each ib_umem in the range. Returns the logical or of
* the return values of the functions called.
*/
int rbt_ib_umem_for_each_in_range(struct rb_root_cached *root,
u64 start, u64 end,
umem_call_back cb,
bool blockable, void *cookie);
/*
* Find first region intersecting with address range.
* Return NULL if not found
*/
struct ib_umem_odp *rbt_ib_umem_lookup(struct rb_root_cached *root,
u64 addr, u64 length);
static inline int ib_umem_mmu_notifier_retry(struct ib_umem *item,
unsigned long mmu_seq)
{
/*
* This code is strongly based on the KVM code from
* mmu_notifier_retry. Should be called with
* the relevant locks taken (item->odp_data->umem_mutex
* and the ucontext umem_mutex semaphore locked for read).
*/
/* Do not allow page faults while the new ib_umem hasn't seen a state
* with zero notifiers yet, and doesn't have its own valid set of
* private counters. */
if (!item->odp_data->mn_counters_active)
return 1;
if (unlikely(item->odp_data->notifiers_count))
return 1;
if (item->odp_data->notifiers_seq != mmu_seq)
return 1;
return 0;
}
#else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
static inline int ib_umem_odp_get(struct ib_ucontext *context,
struct ib_umem *umem,
int access)
{
return -EINVAL;
}
static inline struct ib_umem *ib_alloc_odp_umem(struct ib_ucontext *context,
unsigned long addr,
size_t size)
{
return ERR_PTR(-EINVAL);
}
static inline void ib_umem_odp_release(struct ib_umem *umem) {}
#endif /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
#endif /* IB_UMEM_ODP_H */
File diff suppressed because it is too large. Load diff
+262
View File
@@ -0,0 +1,262 @@
/*
* Copyright (c) 2005 Network Appliance, Inc. All rights reserved.
* Copyright (c) 2005 Open Grid Computing, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef IW_CM_H
#define IW_CM_H
#include <linux/in.h>
#include <rdma/ib_cm.h>
struct iw_cm_id;
enum iw_cm_event_type {
IW_CM_EVENT_CONNECT_REQUEST = 1, /* connect request received */
IW_CM_EVENT_CONNECT_REPLY, /* reply from active connect request */
IW_CM_EVENT_ESTABLISHED, /* passive side accept successful */
IW_CM_EVENT_DISCONNECT, /* orderly shutdown */
IW_CM_EVENT_CLOSE /* close complete */
};
struct iw_cm_event {
enum iw_cm_event_type event;
int status;
struct sockaddr_storage local_addr;
struct sockaddr_storage remote_addr;
void *private_data;
void *provider_data;
u8 private_data_len;
u8 ord;
u8 ird;
};
/**
* iw_cm_handler - Function to be called by the IW CM when delivering events
* to the client.
*
* @cm_id: The IW CM identifier associated with the event.
* @event: Pointer to the event structure.
*/
typedef int (*iw_cm_handler)(struct iw_cm_id *cm_id,
struct iw_cm_event *event);
/**
* iw_event_handler - Function called by the provider when delivering provider
* events to the IW CM. Returns either 0 indicating the event was processed
* or -errno if the event could not be processed.
*
* @cm_id: The IW CM identifier associated with the event.
* @event: Pointer to the event structure.
*/
typedef int (*iw_event_handler)(struct iw_cm_id *cm_id,
struct iw_cm_event *event);
struct iw_cm_id {
iw_cm_handler cm_handler; /* client callback function */
void *context; /* client cb context */
struct ib_device *device;
struct sockaddr_storage local_addr; /* local addr */
struct sockaddr_storage remote_addr;
struct sockaddr_storage m_local_addr; /* nmapped local addr */
struct sockaddr_storage m_remote_addr; /* nmapped rem addr */
void *provider_data; /* provider private data */
iw_event_handler event_handler; /* cb for provider
events */
/* Used by provider to add and remove refs on IW cm_id */
void (*add_ref)(struct iw_cm_id *);
void (*rem_ref)(struct iw_cm_id *);
u8 tos;
bool mapped;
};
struct iw_cm_conn_param {
const void *private_data;
u16 private_data_len;
u32 ord;
u32 ird;
u32 qpn;
};
struct iw_cm_verbs {
void (*add_ref)(struct ib_qp *qp);
void (*rem_ref)(struct ib_qp *qp);
struct ib_qp * (*get_qp)(struct ib_device *device,
int qpn);
int (*connect)(struct iw_cm_id *cm_id,
struct iw_cm_conn_param *conn_param);
int (*accept)(struct iw_cm_id *cm_id,
struct iw_cm_conn_param *conn_param);
int (*reject)(struct iw_cm_id *cm_id,
const void *pdata, u8 pdata_len);
int (*create_listen)(struct iw_cm_id *cm_id,
int backlog);
int (*destroy_listen)(struct iw_cm_id *cm_id);
char ifname[IFNAMSIZ];
};
/**
* iw_create_cm_id - Create an IW CM identifier.
*
* @device: The IB device on which to create the IW CM identier.
* @event_handler: User callback invoked to report events associated with the
* returned IW CM identifier.
* @context: User specified context associated with the id.
*/
struct iw_cm_id *iw_create_cm_id(struct ib_device *device,
iw_cm_handler cm_handler, void *context);
/**
* iw_destroy_cm_id - Destroy an IW CM identifier.
*
* @cm_id: The previously created IW CM identifier to destroy.
*
* The client can assume that no events will be delivered for the CM ID after
* this function returns.
*/
void iw_destroy_cm_id(struct iw_cm_id *cm_id);
/**
* iw_cm_bind_qp - Unbind the specified IW CM identifier and QP
*
* @cm_id: The IW CM idenfier to unbind from the QP.
* @qp: The QP
*
* This is called by the provider when destroying the QP to ensure
* that any references held by the IWCM are released. It may also
* be called by the IWCM when destroying a CM_ID to that any
* references held by the provider are released.
*/
void iw_cm_unbind_qp(struct iw_cm_id *cm_id, struct ib_qp *qp);
/**
* iw_cm_get_qp - Return the ib_qp associated with a QPN
*
* @ib_device: The IB device
* @qpn: The queue pair number
*/
struct ib_qp *iw_cm_get_qp(struct ib_device *device, int qpn);
/**
* iw_cm_listen - Listen for incoming connection requests on the
* specified IW CM id.
*
* @cm_id: The IW CM identifier.
* @backlog: The maximum number of outstanding un-accepted inbound listen
* requests to queue.
*
* The source address and port number are specified in the IW CM identifier
* structure.
*/
int iw_cm_listen(struct iw_cm_id *cm_id, int backlog);
/**
* iw_cm_accept - Called to accept an incoming connect request.
*
* @cm_id: The IW CM identifier associated with the connection request.
* @iw_param: Pointer to a structure containing connection establishment
* parameters.
*
* The specified cm_id will have been provided in the event data for a
* CONNECT_REQUEST event. Subsequent events related to this connection will be
* delivered to the specified IW CM identifier prior and may occur prior to
* the return of this function. If this function returns a non-zero value, the
* client can assume that no events will be delivered to the specified IW CM
* identifier.
*/
int iw_cm_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *iw_param);
/**
* iw_cm_reject - Reject an incoming connection request.
*
* @cm_id: Connection identifier associated with the request.
* @private_daa: Pointer to data to deliver to the remote peer as part of the
* reject message.
* @private_data_len: The number of bytes in the private_data parameter.
*
* The client can assume that no events will be delivered to the specified IW
* CM identifier following the return of this function. The private_data
* buffer is available for reuse when this function returns.
*/
int iw_cm_reject(struct iw_cm_id *cm_id, const void *private_data,
u8 private_data_len);
/**
* iw_cm_connect - Called to request a connection to a remote peer.
*
* @cm_id: The IW CM identifier for the connection.
* @iw_param: Pointer to a structure containing connection establishment
* parameters.
*
* Events may be delivered to the specified IW CM identifier prior to the
* return of this function. If this function returns a non-zero value, the
* client can assume that no events will be delivered to the specified IW CM
* identifier.
*/
int iw_cm_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *iw_param);
/**
* iw_cm_disconnect - Close the specified connection.
*
* @cm_id: The IW CM identifier to close.
* @abrupt: If 0, the connection will be closed gracefully, otherwise, the
* connection will be reset.
*
* The IW CM identifier is still active until the IW_CM_EVENT_CLOSE event is
* delivered.
*/
int iw_cm_disconnect(struct iw_cm_id *cm_id, int abrupt);
/**
* iw_cm_init_qp_attr - Called to initialize the attributes of the QP
* associated with a IW CM identifier.
*
* @cm_id: The IW CM identifier associated with the QP
* @qp_attr: Pointer to the QP attributes structure.
* @qp_attr_mask: Pointer to a bit vector specifying which QP attributes are
* valid.
*/
int iw_cm_init_qp_attr(struct iw_cm_id *cm_id, struct ib_qp_attr *qp_attr,
int *qp_attr_mask);
/**
* iwcm_reject_msg - return a pointer to a reject message string.
* @reason: Value returned in the REJECT event status field.
*/
const char *__attribute_const__ iwcm_reject_msg(int reason);
#endif /* IW_CM_H */
+224
View File
@@ -0,0 +1,224 @@
/*
* Copyright (c) 2014 Intel Corporation. All rights reserved.
* Copyright (c) 2014 Chelsio, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _IW_PORTMAP_H
#define _IW_PORTMAP_H
#define IWPM_ULIBNAME_SIZE 32
#define IWPM_DEVNAME_SIZE 32
#define IWPM_IFNAME_SIZE 16
#define IWPM_IPADDR_SIZE 16
enum {
IWPM_INVALID_NLMSG_ERR = 10,
IWPM_CREATE_MAPPING_ERR,
IWPM_DUPLICATE_MAPPING_ERR,
IWPM_UNKNOWN_MAPPING_ERR,
IWPM_CLIENT_DEV_INFO_ERR,
IWPM_USER_LIB_INFO_ERR,
IWPM_REMOTE_QUERY_REJECT
};
struct iwpm_dev_data {
char dev_name[IWPM_DEVNAME_SIZE];
char if_name[IWPM_IFNAME_SIZE];
};
struct iwpm_sa_data {
struct sockaddr_storage loc_addr;
struct sockaddr_storage mapped_loc_addr;
struct sockaddr_storage rem_addr;
struct sockaddr_storage mapped_rem_addr;
};
/**
* iwpm_init - Allocate resources for the iwarp port mapper
*
* Should be called when network interface goes up.
*/
int iwpm_init(u8);
/**
* iwpm_exit - Deallocate resources for the iwarp port mapper
*
* Should be called when network interface goes down.
*/
int iwpm_exit(u8);
/**
* iwpm_valid_pid - Check if the userspace iwarp port mapper pid is valid
*
* Returns true if the pid is greater than zero, otherwise returns false
*/
int iwpm_valid_pid(void);
/**
* iwpm_register_pid - Send a netlink query to userspace
* to get the iwarp port mapper pid
* @pm_msg: Contains driver info to send to the userspace port mapper
* @nl_client: The index of the netlink client
*/
int iwpm_register_pid(struct iwpm_dev_data *pm_msg, u8 nl_client);
/**
* iwpm_add_mapping - Send a netlink add mapping request to
* the userspace port mapper
* @pm_msg: Contains the local ip/tcp address info to send
* @nl_client: The index of the netlink client
*
* If the request is successful, the pm_msg stores
* the port mapper response (mapped address info)
*/
int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client);
/**
* iwpm_add_and_query_mapping - Send a netlink add and query mapping request
* to the userspace port mapper
* @pm_msg: Contains the local and remote ip/tcp address info to send
* @nl_client: The index of the netlink client
*
* If the request is successful, the pm_msg stores the
* port mapper response (mapped local and remote address info)
*/
int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client);
/**
* iwpm_remove_mapping - Send a netlink remove mapping request
* to the userspace port mapper
*
* @local_addr: Local ip/tcp address to remove
* @nl_client: The index of the netlink client
*/
int iwpm_remove_mapping(struct sockaddr_storage *local_addr, u8 nl_client);
/**
* iwpm_register_pid_cb - Process the port mapper response to
* iwpm_register_pid query
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* If successful, the function receives the userspace port mapper pid
* which is used in future communication with the port mapper
*/
int iwpm_register_pid_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_add_mapping_cb - Process the port mapper response to
* iwpm_add_mapping request
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_add_mapping_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_add_and_query_mapping_cb - Process the port mapper response to
* iwpm_add_and_query_mapping request
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_add_and_query_mapping_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_remote_info_cb - Process remote connecting peer address info, which
* the port mapper has received from the connecting peer
*
* @cb: Contains the received message (payload and netlink header)
*
* Stores the IPv4/IPv6 address info in a hash table
*/
int iwpm_remote_info_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_mapping_error_cb - Process port mapper notification for error
*
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_mapping_error_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_mapping_info_cb - Process a notification that the userspace
* port mapper daemon is started
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* Using the received port mapper pid, send all the local mapping
* info records to the userspace port mapper
*/
int iwpm_mapping_info_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_ack_mapping_info_cb - Process the port mapper ack for
* the provided local mapping info records
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_ack_mapping_info_cb(struct sk_buff *, struct netlink_callback *);
/**
* iwpm_get_remote_info - Get the remote connecting peer address info
*
* @mapped_loc_addr: Mapped local address of the listening peer
* @mapped_rem_addr: Mapped remote address of the connecting peer
* @remote_addr: To store the remote address of the connecting peer
* @nl_client: The index of the netlink client
*
* The remote address info is retrieved and provided to the client in
* the remote_addr. After that it is removed from the hash table
*/
int iwpm_get_remote_info(struct sockaddr_storage *mapped_loc_addr,
struct sockaddr_storage *mapped_rem_addr,
struct sockaddr_storage *remote_addr, u8 nl_client);
/**
* iwpm_create_mapinfo - Store local and mapped IPv4/IPv6 address
* info in a hash table
* @local_addr: Local ip/tcp address
* @mapped_addr: Mapped local ip/tcp address
* @nl_client: The index of the netlink client
*/
int iwpm_create_mapinfo(struct sockaddr_storage *local_addr,
struct sockaddr_storage *mapped_addr, u8 nl_client);
/**
* iwpm_remove_mapinfo - Remove local and mapped IPv4/IPv6 address
* info from the hash table
* @local_addr: Local ip/tcp address
* @mapped_addr: Mapped local ip/tcp address
*
* Returns err code if mapping info is not found in the hash table,
* otherwise returns 0
*/
int iwpm_remove_mapinfo(struct sockaddr_storage *local_addr,
struct sockaddr_storage *mapped_addr);
#endif /* _IW_PORTMAP_H */
+25
View File
@@ -0,0 +1,25 @@
/*
* Copyright (c) 2016 HGST, a Western Digital Company.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*/
#ifndef _RDMA_MR_POOL_H
#define _RDMA_MR_POOL_H 1
#include <rdma/ib_verbs.h>
struct ib_mr *ib_mr_pool_get(struct ib_qp *qp, struct list_head *list);
void ib_mr_pool_put(struct ib_qp *qp, struct list_head *list, struct ib_mr *mr);
int ib_mr_pool_init(struct ib_qp *qp, struct list_head *list, int nr,
enum ib_mr_type type, u32 max_num_sg);
void ib_mr_pool_destroy(struct ib_qp *qp, struct list_head *list);
#endif /* _RDMA_MR_POOL_H */
+133
View File
@@ -0,0 +1,133 @@
/*
* Copyright(c) 2017 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#ifndef OPA_ADDR_H
#define OPA_ADDR_H
#include <rdma/opa_smi.h>
#define OPA_SPECIAL_OUI (0x00066AULL)
#define OPA_MAKE_ID(x) (cpu_to_be64(OPA_SPECIAL_OUI << 40 | (x)))
#define OPA_TO_IB_UCAST_LID(x) (((x) >= be16_to_cpu(IB_MULTICAST_LID_BASE)) \
? 0 : x)
#define OPA_GID_INDEX 0x1
/**
* 0xF8 - 4 bits of multicast range and 1 bit for collective range
* Example: For 24 bit LID space,
* Multicast range: 0xF00000 to 0xF7FFFF
* Collective range: 0xF80000 to 0xFFFFFE
*/
#define OPA_MCAST_NR 0x4 /* Number of top bits set */
#define OPA_COLLECTIVE_NR 0x1 /* Number of bits after MCAST_NR */
/**
* ib_is_opa_gid: Returns true if the top 24 bits of the gid
* contains the OPA_STL_OUI identifier. This identifies that
* the provided gid is a special purpose GID meant to carry
* extended LID information.
*
* @gid: The Global identifier
*/
static inline bool ib_is_opa_gid(const union ib_gid *gid)
{
return ((be64_to_cpu(gid->global.interface_id) >> 40) ==
OPA_SPECIAL_OUI);
}
/**
* opa_get_lid_from_gid: Returns the last 32 bits of the gid.
* OPA devices use one of the gids in the gid table to also
* store the lid.
*
* @gid: The Global identifier
*/
static inline u32 opa_get_lid_from_gid(const union ib_gid *gid)
{
return be64_to_cpu(gid->global.interface_id) & 0xFFFFFFFF;
}
/**
* opa_is_extended_lid: Returns true if dlid or slid are
* extended.
*
* @dlid: The DLID
* @slid: The SLID
*/
static inline bool opa_is_extended_lid(__be32 dlid, __be32 slid)
{
if ((be32_to_cpu(dlid) >=
be16_to_cpu(IB_MULTICAST_LID_BASE)) ||
(be32_to_cpu(slid) >=
be16_to_cpu(IB_MULTICAST_LID_BASE)))
return true;
return false;
}
/* Get multicast lid base */
static inline u32 opa_get_mcast_base(u32 nr_top_bits)
{
return (be32_to_cpu(OPA_LID_PERMISSIVE) << (32 - nr_top_bits));
}
/* Check for a valid unicast LID for non-SM traffic types */
static inline bool rdma_is_valid_unicast_lid(struct rdma_ah_attr *attr)
{
if (attr->type == RDMA_AH_ATTR_TYPE_IB) {
if (!rdma_ah_get_dlid(attr) ||
rdma_ah_get_dlid(attr) >=
be16_to_cpu(IB_MULTICAST_LID_BASE))
return false;
} else if (attr->type == RDMA_AH_ATTR_TYPE_OPA) {
if (!rdma_ah_get_dlid(attr) ||
rdma_ah_get_dlid(attr) >=
opa_get_mcast_base(OPA_MCAST_NR))
return false;
}
return true;
}
#endif /* OPA_ADDR_H */
+418
View File
@@ -0,0 +1,418 @@
/*
* Copyright (c) 2014-2017 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(OPA_PORT_INFO_H)
#define OPA_PORT_INFO_H
#define OPA_PORT_LINK_MODE_NOP 0 /* No change */
#define OPA_PORT_LINK_MODE_OPA 4 /* Port mode is OPA */
#define OPA_PORT_PACKET_FORMAT_NOP 0 /* No change */
#define OPA_PORT_PACKET_FORMAT_8B 1 /* Format 8B */
#define OPA_PORT_PACKET_FORMAT_9B 2 /* Format 9B */
#define OPA_PORT_PACKET_FORMAT_10B 4 /* Format 10B */
#define OPA_PORT_PACKET_FORMAT_16B 8 /* Format 16B */
#define OPA_PORT_LTP_CRC_MODE_NONE 0 /* No change */
#define OPA_PORT_LTP_CRC_MODE_14 1 /* 14-bit LTP CRC mode (optional) */
#define OPA_PORT_LTP_CRC_MODE_16 2 /* 16-bit LTP CRC mode */
#define OPA_PORT_LTP_CRC_MODE_48 4 /* 48-bit LTP CRC mode (optional) */
#define OPA_PORT_LTP_CRC_MODE_PER_LANE 8 /* 12/16-bit per lane LTP CRC mode */
/* Link Down / Neighbor Link Down Reason; indicated as follows: */
#define OPA_LINKDOWN_REASON_NONE 0 /* No specified reason */
#define OPA_LINKDOWN_REASON_RCV_ERROR_0 1
#define OPA_LINKDOWN_REASON_BAD_PKT_LEN 2
#define OPA_LINKDOWN_REASON_PKT_TOO_LONG 3
#define OPA_LINKDOWN_REASON_PKT_TOO_SHORT 4
#define OPA_LINKDOWN_REASON_BAD_SLID 5
#define OPA_LINKDOWN_REASON_BAD_DLID 6
#define OPA_LINKDOWN_REASON_BAD_L2 7
#define OPA_LINKDOWN_REASON_BAD_SC 8
#define OPA_LINKDOWN_REASON_RCV_ERROR_8 9
#define OPA_LINKDOWN_REASON_BAD_MID_TAIL 10
#define OPA_LINKDOWN_REASON_RCV_ERROR_10 11
#define OPA_LINKDOWN_REASON_PREEMPT_ERROR 12
#define OPA_LINKDOWN_REASON_PREEMPT_VL15 13
#define OPA_LINKDOWN_REASON_BAD_VL_MARKER 14
#define OPA_LINKDOWN_REASON_RCV_ERROR_14 15
#define OPA_LINKDOWN_REASON_RCV_ERROR_15 16
#define OPA_LINKDOWN_REASON_BAD_HEAD_DIST 17
#define OPA_LINKDOWN_REASON_BAD_TAIL_DIST 18
#define OPA_LINKDOWN_REASON_BAD_CTRL_DIST 19
#define OPA_LINKDOWN_REASON_BAD_CREDIT_ACK 20
#define OPA_LINKDOWN_REASON_UNSUPPORTED_VL_MARKER 21
#define OPA_LINKDOWN_REASON_BAD_PREEMPT 22
#define OPA_LINKDOWN_REASON_BAD_CONTROL_FLIT 23
#define OPA_LINKDOWN_REASON_EXCEED_MULTICAST_LIMIT 24
#define OPA_LINKDOWN_REASON_RCV_ERROR_24 25
#define OPA_LINKDOWN_REASON_RCV_ERROR_25 26
#define OPA_LINKDOWN_REASON_RCV_ERROR_26 27
#define OPA_LINKDOWN_REASON_RCV_ERROR_27 28
#define OPA_LINKDOWN_REASON_RCV_ERROR_28 29
#define OPA_LINKDOWN_REASON_RCV_ERROR_29 30
#define OPA_LINKDOWN_REASON_RCV_ERROR_30 31
#define OPA_LINKDOWN_REASON_EXCESSIVE_BUFFER_OVERRUN 32
#define OPA_LINKDOWN_REASON_UNKNOWN 33
/* 34 -reserved */
#define OPA_LINKDOWN_REASON_REBOOT 35
#define OPA_LINKDOWN_REASON_NEIGHBOR_UNKNOWN 36
/* 37-38 reserved */
#define OPA_LINKDOWN_REASON_FM_BOUNCE 39
#define OPA_LINKDOWN_REASON_SPEED_POLICY 40
#define OPA_LINKDOWN_REASON_WIDTH_POLICY 41
/* 42-48 reserved */
#define OPA_LINKDOWN_REASON_DISCONNECTED 49
#define OPA_LINKDOWN_REASON_LOCAL_MEDIA_NOT_INSTALLED 50
#define OPA_LINKDOWN_REASON_NOT_INSTALLED 51
#define OPA_LINKDOWN_REASON_CHASSIS_CONFIG 52
/* 53 reserved */
#define OPA_LINKDOWN_REASON_END_TO_END_NOT_INSTALLED 54
/* 55 reserved */
#define OPA_LINKDOWN_REASON_POWER_POLICY 56
#define OPA_LINKDOWN_REASON_LINKSPEED_POLICY 57
#define OPA_LINKDOWN_REASON_LINKWIDTH_POLICY 58
/* 59 reserved */
#define OPA_LINKDOWN_REASON_SWITCH_MGMT 60
#define OPA_LINKDOWN_REASON_SMA_DISABLED 61
/* 62 reserved */
#define OPA_LINKDOWN_REASON_TRANSIENT 63
/* 64-255 reserved */
/* OPA Link Init reason; indicated as follows: */
/* 3-7; 11-15 reserved; 8-15 cleared on Polling->LinkUp */
#define OPA_LINKINIT_REASON_NOP 0
#define OPA_LINKINIT_REASON_LINKUP (1 << 4)
#define OPA_LINKINIT_REASON_FLAPPING (2 << 4)
#define OPA_LINKINIT_REASON_CLEAR (8 << 4)
#define OPA_LINKINIT_OUTSIDE_POLICY (8 << 4)
#define OPA_LINKINIT_QUARANTINED (9 << 4)
#define OPA_LINKINIT_INSUFIC_CAPABILITY (10 << 4)
#define OPA_LINK_SPEED_NOP 0x0000 /* Reserved (1-5 Gbps) */
#define OPA_LINK_SPEED_12_5G 0x0001 /* 12.5 Gbps */
#define OPA_LINK_SPEED_25G 0x0002 /* 25.78125? Gbps (EDR) */
#define OPA_LINK_WIDTH_1X 0x0001
#define OPA_LINK_WIDTH_2X 0x0002
#define OPA_LINK_WIDTH_3X 0x0004
#define OPA_LINK_WIDTH_4X 0x0008
#define OPA_CAP_MASK3_IsEthOnFabricSupported (1 << 13)
#define OPA_CAP_MASK3_IsSnoopSupported (1 << 7)
#define OPA_CAP_MASK3_IsAsyncSC2VLSupported (1 << 6)
#define OPA_CAP_MASK3_IsAddrRangeConfigSupported (1 << 5)
#define OPA_CAP_MASK3_IsPassThroughSupported (1 << 4)
#define OPA_CAP_MASK3_IsSharedSpaceSupported (1 << 3)
/* reserved (1 << 2) */
#define OPA_CAP_MASK3_IsVLMarkerSupported (1 << 1)
#define OPA_CAP_MASK3_IsVLrSupported (1 << 0)
/**
* new MTU values
*/
enum {
OPA_MTU_8192 = 6,
OPA_MTU_10240 = 7,
};
enum {
OPA_PORT_PHYS_CONF_DISCONNECTED = 0,
OPA_PORT_PHYS_CONF_STANDARD = 1,
OPA_PORT_PHYS_CONF_FIXED = 2,
OPA_PORT_PHYS_CONF_VARIABLE = 3,
OPA_PORT_PHYS_CONF_SI_PHOTO = 4
};
enum port_info_field_masks {
/* vl.cap */
OPA_PI_MASK_VL_CAP = 0x1F,
/* port_states.ledenable_offlinereason */
OPA_PI_MASK_OFFLINE_REASON = 0x0F,
OPA_PI_MASK_LED_ENABLE = 0x40,
/* port_states.unsleepstate_downdefstate */
OPA_PI_MASK_UNSLEEP_STATE = 0xF0,
OPA_PI_MASK_DOWNDEF_STATE = 0x0F,
/* port_states.portphysstate_portstate */
OPA_PI_MASK_PORT_PHYSICAL_STATE = 0xF0,
OPA_PI_MASK_PORT_STATE = 0x0F,
/* port_phys_conf */
OPA_PI_MASK_PORT_PHYSICAL_CONF = 0x0F,
/* collectivemask_multicastmask */
OPA_PI_MASK_COLLECT_MASK = 0x38,
OPA_PI_MASK_MULTICAST_MASK = 0x07,
/* mkeyprotect_lmc */
OPA_PI_MASK_MKEY_PROT_BIT = 0xC0,
OPA_PI_MASK_LMC = 0x0F,
/* smsl */
OPA_PI_MASK_SMSL = 0x1F,
/* partenforce_filterraw */
/* Filter Raw In/Out bits 1 and 2 were removed */
OPA_PI_MASK_LINKINIT_REASON = 0xF0,
OPA_PI_MASK_PARTITION_ENFORCE_IN = 0x08,
OPA_PI_MASK_PARTITION_ENFORCE_OUT = 0x04,
/* operational_vls */
OPA_PI_MASK_OPERATIONAL_VL = 0x1F,
/* sa_qp */
OPA_PI_MASK_SA_QP = 0x00FFFFFF,
/* sm_trap_qp */
OPA_PI_MASK_SM_TRAP_QP = 0x00FFFFFF,
/* localphy_overrun_errors */
OPA_PI_MASK_LOCAL_PHY_ERRORS = 0xF0,
OPA_PI_MASK_OVERRUN_ERRORS = 0x0F,
/* clientrereg_subnettimeout */
OPA_PI_MASK_CLIENT_REREGISTER = 0x80,
OPA_PI_MASK_SUBNET_TIMEOUT = 0x1F,
/* port_link_mode */
OPA_PI_MASK_PORT_LINK_SUPPORTED = (0x001F << 10),
OPA_PI_MASK_PORT_LINK_ENABLED = (0x001F << 5),
OPA_PI_MASK_PORT_LINK_ACTIVE = (0x001F << 0),
/* port_link_crc_mode */
OPA_PI_MASK_PORT_LINK_CRC_SUPPORTED = 0x0F00,
OPA_PI_MASK_PORT_LINK_CRC_ENABLED = 0x00F0,
OPA_PI_MASK_PORT_LINK_CRC_ACTIVE = 0x000F,
/* port_mode */
OPA_PI_MASK_PORT_MODE_SECURITY_CHECK = 0x0001,
OPA_PI_MASK_PORT_MODE_16B_TRAP_QUERY = 0x0002,
OPA_PI_MASK_PORT_MODE_PKEY_CONVERT = 0x0004,
OPA_PI_MASK_PORT_MODE_SC2SC_MAPPING = 0x0008,
OPA_PI_MASK_PORT_MODE_VL_MARKER = 0x0010,
OPA_PI_MASK_PORT_PASS_THROUGH = 0x0020,
OPA_PI_MASK_PORT_ACTIVE_OPTOMIZE = 0x0040,
/* flit_control.interleave */
OPA_PI_MASK_INTERLEAVE_DIST_SUP = (0x0003 << 12),
OPA_PI_MASK_INTERLEAVE_DIST_ENABLE = (0x0003 << 10),
OPA_PI_MASK_INTERLEAVE_MAX_NEST_TX = (0x001F << 5),
OPA_PI_MASK_INTERLEAVE_MAX_NEST_RX = (0x001F << 0),
/* port_error_action */
OPA_PI_MASK_EX_BUFFER_OVERRUN = 0x80000000,
/* 7 bits reserved */
OPA_PI_MASK_FM_CFG_ERR_EXCEED_MULTICAST_LIMIT = 0x00800000,
OPA_PI_MASK_FM_CFG_BAD_CONTROL_FLIT = 0x00400000,
OPA_PI_MASK_FM_CFG_BAD_PREEMPT = 0x00200000,
OPA_PI_MASK_FM_CFG_UNSUPPORTED_VL_MARKER = 0x00100000,
OPA_PI_MASK_FM_CFG_BAD_CRDT_ACK = 0x00080000,
OPA_PI_MASK_FM_CFG_BAD_CTRL_DIST = 0x00040000,
OPA_PI_MASK_FM_CFG_BAD_TAIL_DIST = 0x00020000,
OPA_PI_MASK_FM_CFG_BAD_HEAD_DIST = 0x00010000,
/* 2 bits reserved */
OPA_PI_MASK_PORT_RCV_BAD_VL_MARKER = 0x00002000,
OPA_PI_MASK_PORT_RCV_PREEMPT_VL15 = 0x00001000,
OPA_PI_MASK_PORT_RCV_PREEMPT_ERROR = 0x00000800,
/* 1 bit reserved */
OPA_PI_MASK_PORT_RCV_BAD_MidTail = 0x00000200,
/* 1 bit reserved */
OPA_PI_MASK_PORT_RCV_BAD_SC = 0x00000080,
OPA_PI_MASK_PORT_RCV_BAD_L2 = 0x00000040,
OPA_PI_MASK_PORT_RCV_BAD_DLID = 0x00000020,
OPA_PI_MASK_PORT_RCV_BAD_SLID = 0x00000010,
OPA_PI_MASK_PORT_RCV_PKTLEN_TOOSHORT = 0x00000008,
OPA_PI_MASK_PORT_RCV_PKTLEN_TOOLONG = 0x00000004,
OPA_PI_MASK_PORT_RCV_BAD_PKTLEN = 0x00000002,
OPA_PI_MASK_PORT_RCV_BAD_LT = 0x00000001,
/* pass_through.res_drctl */
OPA_PI_MASK_PASS_THROUGH_DR_CONTROL = 0x01,
/* buffer_units */
OPA_PI_MASK_BUF_UNIT_VL15_INIT = (0x00000FFF << 11),
OPA_PI_MASK_BUF_UNIT_VL15_CREDIT_RATE = (0x0000001F << 6),
OPA_PI_MASK_BUF_UNIT_CREDIT_ACK = (0x00000003 << 3),
OPA_PI_MASK_BUF_UNIT_BUF_ALLOC = (0x00000003 << 0),
/* neigh_mtu.pvlx_to_mtu */
OPA_PI_MASK_NEIGH_MTU_PVL0 = 0xF0,
OPA_PI_MASK_NEIGH_MTU_PVL1 = 0x0F,
/* neigh_mtu.vlstall_hoq_life */
OPA_PI_MASK_VL_STALL = (0x03 << 5),
OPA_PI_MASK_HOQ_LIFE = (0x1F << 0),
/* port_neigh_mode */
OPA_PI_MASK_NEIGH_MGMT_ALLOWED = (0x01 << 3),
OPA_PI_MASK_NEIGH_FW_AUTH_BYPASS = (0x01 << 2),
OPA_PI_MASK_NEIGH_NODE_TYPE = (0x03 << 0),
/* resptime_value */
OPA_PI_MASK_RESPONSE_TIME_VALUE = 0x1F,
/* mtucap */
OPA_PI_MASK_MTU_CAP = 0x0F,
};
struct opa_port_states {
u8 reserved;
u8 ledenable_offlinereason; /* 1 res, 1 bit, 6 bits */
u8 reserved2;
u8 portphysstate_portstate; /* 4 bits, 4 bits */
};
struct opa_port_state_info {
struct opa_port_states port_states;
__be16 link_width_downgrade_tx_active;
__be16 link_width_downgrade_rx_active;
};
struct opa_port_info {
__be32 lid;
__be32 flow_control_mask;
struct {
u8 res; /* was inittype */
u8 cap; /* 3 res, 5 bits */
__be16 high_limit;
__be16 preempt_limit;
u8 arb_high_cap;
u8 arb_low_cap;
} vl;
struct opa_port_states port_states;
u8 port_phys_conf; /* 4 res, 4 bits */
u8 collectivemask_multicastmask; /* 2 res, 3, 3 */
u8 mkeyprotect_lmc; /* 2 bits, 2 res, 4 bits */
u8 smsl; /* 3 res, 5 bits */
u8 partenforce_filterraw; /* bit fields */
u8 operational_vls; /* 3 res, 5 bits */
__be16 pkey_8b;
__be16 pkey_10b;
__be16 mkey_violations;
__be16 pkey_violations;
__be16 qkey_violations;
__be32 sm_trap_qp; /* 8 bits, 24 bits */
__be32 sa_qp; /* 8 bits, 24 bits */
u8 neigh_port_num;
u8 link_down_reason;
u8 neigh_link_down_reason;
u8 clientrereg_subnettimeout; /* 1 bit, 2 bits, 5 */
struct {
__be16 supported;
__be16 enabled;
__be16 active;
} link_speed;
struct {
__be16 supported;
__be16 enabled;
__be16 active;
} link_width;
struct {
__be16 supported;
__be16 enabled;
__be16 tx_active;
__be16 rx_active;
} link_width_downgrade;
__be16 port_link_mode; /* 1 res, 5 bits, 5 bits, 5 bits */
__be16 port_ltp_crc_mode; /* 4 res, 4 bits, 4 bits, 4 bits */
__be16 port_mode; /* 9 res, bit fields */
struct {
__be16 supported;
__be16 enabled;
} port_packet_format;
struct {
__be16 interleave; /* 2 res, 2,2,5,5 */
struct {
__be16 min_initial;
__be16 min_tail;
u8 large_pkt_limit;
u8 small_pkt_limit;
u8 max_small_pkt_limit;
u8 preemption_limit;
} preemption;
} flit_control;
__be32 reserved4;
__be32 port_error_action; /* bit field */
struct {
u8 egress_port;
u8 res_drctl; /* 7 res, 1 */
} pass_through;
__be16 mkey_lease_period;
__be32 buffer_units; /* 9 res, 12, 5, 3, 3 */
__be32 reserved5;
__be32 sm_lid;
__be64 mkey;
__be64 subnet_prefix;
struct {
u8 pvlx_to_mtu[OPA_MAX_VLS/2]; /* 4 bits, 4 bits */
} neigh_mtu;
struct {
u8 vlstall_hoqlife; /* 3 bits, 5 bits */
} xmit_q[OPA_MAX_VLS];
struct {
u8 addr[16];
} ipaddr_ipv6;
struct {
u8 addr[4];
} ipaddr_ipv4;
u32 reserved6;
u32 reserved7;
u32 reserved8;
__be64 neigh_node_guid;
__be32 ib_cap_mask;
__be16 reserved9; /* was ib_cap_mask2 */
__be16 opa_cap_mask;
__be32 reserved10; /* was link_roundtrip_latency */
__be16 overall_buffer_space;
__be16 reserved11; /* was max_credit_hint */
__be16 diag_code;
struct {
u8 buffer;
u8 wire;
} replay_depth;
u8 port_neigh_mode;
u8 mtucap; /* 4 res, 4 bits */
u8 resptimevalue; /* 3 res, 5 bits */
u8 local_port_num;
u8 reserved12;
u8 reserved13; /* was guid_cap */
} __attribute__ ((packed));
#endif /* OPA_PORT_INFO_H */
+151
View File
@@ -0,0 +1,151 @@
/*
* Copyright (c) 2014 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(OPA_SMI_H)
#define OPA_SMI_H
#include <rdma/ib_mad.h>
#include <rdma/ib_smi.h>
#define OPA_SMP_LID_DATA_SIZE 2016
#define OPA_SMP_DR_DATA_SIZE 1872
#define OPA_SMP_MAX_PATH_HOPS 64
#define OPA_MAX_VLS 32
#define OPA_MAX_SLS 32
#define OPA_MAX_SCS 32
#define OPA_LID_PERMISSIVE cpu_to_be32(0xFFFFFFFF)
struct opa_smp {
u8 base_version;
u8 mgmt_class;
u8 class_version;
u8 method;
__be16 status;
u8 hop_ptr;
u8 hop_cnt;
__be64 tid;
__be16 attr_id;
__be16 resv;
__be32 attr_mod;
__be64 mkey;
union {
struct {
uint8_t data[OPA_SMP_LID_DATA_SIZE];
} lid;
struct {
__be32 dr_slid;
__be32 dr_dlid;
u8 initial_path[OPA_SMP_MAX_PATH_HOPS];
u8 return_path[OPA_SMP_MAX_PATH_HOPS];
u8 reserved[8];
u8 data[OPA_SMP_DR_DATA_SIZE];
} dr;
} route;
} __packed;
/* Subnet management attributes */
/* ... */
#define OPA_ATTRIB_ID_NODE_DESCRIPTION cpu_to_be16(0x0010)
#define OPA_ATTRIB_ID_NODE_INFO cpu_to_be16(0x0011)
#define OPA_ATTRIB_ID_PORT_INFO cpu_to_be16(0x0015)
#define OPA_ATTRIB_ID_PARTITION_TABLE cpu_to_be16(0x0016)
#define OPA_ATTRIB_ID_SL_TO_SC_MAP cpu_to_be16(0x0017)
#define OPA_ATTRIB_ID_VL_ARBITRATION cpu_to_be16(0x0018)
#define OPA_ATTRIB_ID_SM_INFO cpu_to_be16(0x0020)
#define OPA_ATTRIB_ID_CABLE_INFO cpu_to_be16(0x0032)
#define OPA_ATTRIB_ID_AGGREGATE cpu_to_be16(0x0080)
#define OPA_ATTRIB_ID_SC_TO_SL_MAP cpu_to_be16(0x0082)
#define OPA_ATTRIB_ID_SC_TO_VLR_MAP cpu_to_be16(0x0083)
#define OPA_ATTRIB_ID_SC_TO_VLT_MAP cpu_to_be16(0x0084)
#define OPA_ATTRIB_ID_SC_TO_VLNT_MAP cpu_to_be16(0x0085)
/* ... */
#define OPA_ATTRIB_ID_PORT_STATE_INFO cpu_to_be16(0x0087)
/* ... */
#define OPA_ATTRIB_ID_BUFFER_CONTROL_TABLE cpu_to_be16(0x008A)
/* ... */
struct opa_node_description {
u8 data[64];
} __attribute__ ((packed));
struct opa_node_info {
u8 base_version;
u8 class_version;
u8 node_type;
u8 num_ports;
__be32 reserved;
__be64 system_image_guid;
__be64 node_guid;
__be64 port_guid;
__be16 partition_cap;
__be16 device_id;
__be32 revision;
u8 local_port_num;
u8 vendor_id[3]; /* network byte order */
} __attribute__ ((packed));
#define OPA_PARTITION_TABLE_BLK_SIZE 32
static inline u8
opa_get_smp_direction(struct opa_smp *smp)
{
return ib_get_smp_direction((struct ib_smp *)smp);
}
static inline u8 *opa_get_smp_data(struct opa_smp *smp)
{
if (smp->mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)
return smp->route.dr.data;
return smp->route.lid.data;
}
static inline size_t opa_get_smp_data_size(struct opa_smp *smp)
{
if (smp->mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)
return sizeof(smp->route.dr.data);
return sizeof(smp->route.lid.data);
}
static inline size_t opa_get_smp_header_size(struct opa_smp *smp)
{
if (smp->mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)
return sizeof(*smp) - sizeof(smp->route.dr.data);
return sizeof(*smp) - sizeof(smp->route.lid.data);
}
#endif /* OPA_SMI_H */
+138
View File
@@ -0,0 +1,138 @@
#ifndef _OPA_VNIC_H
#define _OPA_VNIC_H
/*
* Copyright(c) 2017 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
/*
* This file contains Intel Omni-Path (OPA) Virtual Network Interface
* Controller (VNIC) specific declarations.
*/
#include <rdma/ib_verbs.h>
/* 16 header bytes + 2 reserved bytes */
#define OPA_VNIC_L2_HDR_LEN (16 + 2)
#define OPA_VNIC_L4_HDR_LEN 2
#define OPA_VNIC_HDR_LEN (OPA_VNIC_L2_HDR_LEN + \
OPA_VNIC_L4_HDR_LEN)
#define OPA_VNIC_L4_ETHR 0x78
#define OPA_VNIC_ICRC_LEN 4
#define OPA_VNIC_TAIL_LEN 1
#define OPA_VNIC_ICRC_TAIL_LEN (OPA_VNIC_ICRC_LEN + OPA_VNIC_TAIL_LEN)
#define OPA_VNIC_SKB_MDATA_LEN 4
#define OPA_VNIC_SKB_MDATA_ENCAP_ERR 0x1
/* opa vnic rdma netdev's private data structure */
struct opa_vnic_rdma_netdev {
struct rdma_netdev rn; /* keep this first */
/* followed by device private data */
char *dev_priv[0];
};
static inline void *opa_vnic_priv(const struct net_device *dev)
{
struct rdma_netdev *rn = netdev_priv(dev);
return rn->clnt_priv;
}
static inline void *opa_vnic_dev_priv(const struct net_device *dev)
{
struct opa_vnic_rdma_netdev *oparn = netdev_priv(dev);
return oparn->dev_priv;
}
/* opa_vnic skb meta data structrue */
struct opa_vnic_skb_mdata {
u8 vl;
u8 entropy;
u8 flags;
u8 rsvd;
} __packed;
/* OPA VNIC group statistics */
struct opa_vnic_grp_stats {
u64 unicast;
u64 mcastbcast;
u64 untagged;
u64 vlan;
u64 s_64;
u64 s_65_127;
u64 s_128_255;
u64 s_256_511;
u64 s_512_1023;
u64 s_1024_1518;
u64 s_1519_max;
};
struct opa_vnic_stats {
/* standard netdev statistics */
struct rtnl_link_stats64 netstats;
/* OPA VNIC statistics */
struct opa_vnic_grp_stats tx_grp;
struct opa_vnic_grp_stats rx_grp;
u64 tx_dlid_zero;
u64 tx_drop_state;
u64 rx_drop_state;
u64 rx_runt;
u64 rx_oversize;
};
static inline bool rdma_cap_opa_vnic(struct ib_device *device)
{
return !!(device->attrs.device_cap_flags &
IB_DEVICE_RDMA_NETDEV_OPA_VNIC);
}
#endif /* _OPA_VNIC_H */
+426
View File
@@ -0,0 +1,426 @@
/*
* Copyright (c) 2005 Voltaire Inc. All rights reserved.
* Copyright (c) 2005 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(RDMA_CM_H)
#define RDMA_CM_H
#include <linux/socket.h>
#include <linux/in6.h>
#include <rdma/ib_addr.h>
#include <rdma/ib_sa.h>
#include <uapi/rdma/rdma_user_cm.h>
/*
* Upon receiving a device removal event, users must destroy the associated
* RDMA identifier and release all resources allocated with the device.
*/
enum rdma_cm_event_type {
RDMA_CM_EVENT_ADDR_RESOLVED,
RDMA_CM_EVENT_ADDR_ERROR,
RDMA_CM_EVENT_ROUTE_RESOLVED,
RDMA_CM_EVENT_ROUTE_ERROR,
RDMA_CM_EVENT_CONNECT_REQUEST,
RDMA_CM_EVENT_CONNECT_RESPONSE,
RDMA_CM_EVENT_CONNECT_ERROR,
RDMA_CM_EVENT_UNREACHABLE,
RDMA_CM_EVENT_REJECTED,
RDMA_CM_EVENT_ESTABLISHED,
RDMA_CM_EVENT_DISCONNECTED,
RDMA_CM_EVENT_DEVICE_REMOVAL,
RDMA_CM_EVENT_MULTICAST_JOIN,
RDMA_CM_EVENT_MULTICAST_ERROR,
RDMA_CM_EVENT_ADDR_CHANGE,
RDMA_CM_EVENT_TIMEWAIT_EXIT
};
const char *__attribute_const__ rdma_event_msg(enum rdma_cm_event_type event);
#define RDMA_IB_IP_PS_MASK 0xFFFFFFFFFFFF0000ULL
#define RDMA_IB_IP_PS_TCP 0x0000000001060000ULL
#define RDMA_IB_IP_PS_UDP 0x0000000001110000ULL
#define RDMA_IB_IP_PS_IB 0x00000000013F0000ULL
struct rdma_addr {
struct sockaddr_storage src_addr;
struct sockaddr_storage dst_addr;
struct rdma_dev_addr dev_addr;
};
struct rdma_route {
struct rdma_addr addr;
struct sa_path_rec *path_rec;
int num_paths;
};
struct rdma_conn_param {
const void *private_data;
u8 private_data_len;
u8 responder_resources;
u8 initiator_depth;
u8 flow_control;
u8 retry_count; /* ignored when accepting */
u8 rnr_retry_count;
/* Fields below ignored if a QP is created on the rdma_cm_id. */
u8 srq;
u32 qp_num;
u32 qkey;
};
struct rdma_ud_param {
const void *private_data;
u8 private_data_len;
struct rdma_ah_attr ah_attr;
u32 qp_num;
u32 qkey;
};
struct rdma_cm_event {
enum rdma_cm_event_type event;
int status;
union {
struct rdma_conn_param conn;
struct rdma_ud_param ud;
} param;
};
struct rdma_cm_id;
/**
* rdma_cm_event_handler - Callback used to report user events.
*
* Notes: Users may not call rdma_destroy_id from this callback to destroy
* the passed in id, or a corresponding listen id. Returning a
* non-zero value from the callback will destroy the passed in id.
*/
typedef int (*rdma_cm_event_handler)(struct rdma_cm_id *id,
struct rdma_cm_event *event);
struct rdma_cm_id {
struct ib_device *device;
void *context;
struct ib_qp *qp;
rdma_cm_event_handler event_handler;
struct rdma_route route;
enum rdma_ucm_port_space ps;
enum ib_qp_type qp_type;
u8 port_num;
};
struct rdma_cm_id *__rdma_create_id(struct net *net,
rdma_cm_event_handler event_handler,
void *context, enum rdma_ucm_port_space ps,
enum ib_qp_type qp_type,
const char *caller);
/**
* rdma_create_id - Create an RDMA identifier.
*
* @net: The network namespace in which to create the new id.
* @event_handler: User callback invoked to report events associated with the
* returned rdma_id.
* @context: User specified context associated with the id.
* @ps: RDMA port space.
* @qp_type: type of queue pair associated with the id.
*
* The id holds a reference on the network namespace until it is destroyed.
*/
#define rdma_create_id(net, event_handler, context, ps, qp_type) \
__rdma_create_id((net), (event_handler), (context), (ps), (qp_type), \
KBUILD_MODNAME)
/**
* rdma_destroy_id - Destroys an RDMA identifier.
*
* @id: RDMA identifier.
*
* Note: calling this function has the effect of canceling in-flight
* asynchronous operations associated with the id.
*/
void rdma_destroy_id(struct rdma_cm_id *id);
/**
* rdma_bind_addr - Bind an RDMA identifier to a source address and
* associated RDMA device, if needed.
*
* @id: RDMA identifier.
* @addr: Local address information. Wildcard values are permitted.
*
* This associates a source address with the RDMA identifier before calling
* rdma_listen. If a specific local address is given, the RDMA identifier will
* be bound to a local RDMA device.
*/
int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr);
/**
* rdma_resolve_addr - Resolve destination and optional source addresses
* from IP addresses to an RDMA address. If successful, the specified
* rdma_cm_id will be bound to a local device.
*
* @id: RDMA identifier.
* @src_addr: Source address information. This parameter may be NULL.
* @dst_addr: Destination address information.
* @timeout_ms: Time to wait for resolution to complete.
*/
int rdma_resolve_addr(struct rdma_cm_id *id, struct sockaddr *src_addr,
const struct sockaddr *dst_addr, int timeout_ms);
/**
* rdma_resolve_route - Resolve the RDMA address bound to the RDMA identifier
* into route information needed to establish a connection.
*
* This is called on the client side of a connection.
* Users must have first called rdma_resolve_addr to resolve a dst_addr
* into an RDMA address before calling this routine.
*/
int rdma_resolve_route(struct rdma_cm_id *id, int timeout_ms);
/**
* rdma_create_qp - Allocate a QP and associate it with the specified RDMA
* identifier.
*
* QPs allocated to an rdma_cm_id will automatically be transitioned by the CMA
* through their states.
*/
int rdma_create_qp(struct rdma_cm_id *id, struct ib_pd *pd,
struct ib_qp_init_attr *qp_init_attr);
/**
* rdma_destroy_qp - Deallocate the QP associated with the specified RDMA
* identifier.
*
* Users must destroy any QP associated with an RDMA identifier before
* destroying the RDMA ID.
*/
void rdma_destroy_qp(struct rdma_cm_id *id);
/**
* rdma_init_qp_attr - Initializes the QP attributes for use in transitioning
* to a specified QP state.
* @id: Communication identifier associated with the QP attributes to
* initialize.
* @qp_attr: On input, specifies the desired QP state. On output, the
* mandatory and desired optional attributes will be set in order to
* modify the QP to the specified state.
* @qp_attr_mask: The QP attribute mask that may be used to transition the
* QP to the specified state.
*
* Users must set the @qp_attr->qp_state to the desired QP state. This call
* will set all required attributes for the given transition, along with
* known optional attributes. Users may override the attributes returned from
* this call before calling ib_modify_qp.
*
* Users that wish to have their QP automatically transitioned through its
* states can associate a QP with the rdma_cm_id by calling rdma_create_qp().
*/
int rdma_init_qp_attr(struct rdma_cm_id *id, struct ib_qp_attr *qp_attr,
int *qp_attr_mask);
/**
* rdma_connect - Initiate an active connection request.
* @id: Connection identifier to connect.
* @conn_param: Connection information used for connected QPs.
*
* Users must have resolved a route for the rdma_cm_id to connect with
* by having called rdma_resolve_route before calling this routine.
*
* This call will either connect to a remote QP or obtain remote QP
* information for unconnected rdma_cm_id's. The actual operation is
* based on the rdma_cm_id's port space.
*/
int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param);
/**
* rdma_listen - This function is called by the passive side to
* listen for incoming connection requests.
*
* Users must have bound the rdma_cm_id to a local address by calling
* rdma_bind_addr before calling this routine.
*/
int rdma_listen(struct rdma_cm_id *id, int backlog);
int __rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param,
const char *caller);
/**
* rdma_accept - Called to accept a connection request or response.
* @id: Connection identifier associated with the request.
* @conn_param: Information needed to establish the connection. This must be
* provided if accepting a connection request. If accepting a connection
* response, this parameter must be NULL.
*
* Typically, this routine is only called by the listener to accept a connection
* request. It must also be called on the active side of a connection if the
* user is performing their own QP transitions.
*
* In the case of error, a reject message is sent to the remote side and the
* state of the qp associated with the id is modified to error, such that any
* previously posted receive buffers would be flushed.
*/
#define rdma_accept(id, conn_param) \
__rdma_accept((id), (conn_param), KBUILD_MODNAME)
/**
* rdma_notify - Notifies the RDMA CM of an asynchronous event that has
* occurred on the connection.
* @id: Connection identifier to transition to established.
* @event: Asynchronous event.
*
* This routine should be invoked by users to notify the CM of relevant
* communication events. Events that should be reported to the CM and
* when to report them are:
*
* IB_EVENT_COMM_EST - Used when a message is received on a connected
* QP before an RTU has been received.
*/
int rdma_notify(struct rdma_cm_id *id, enum ib_event_type event);
/**
* rdma_reject - Called to reject a connection request or response.
*/
int rdma_reject(struct rdma_cm_id *id, const void *private_data,
u8 private_data_len);
/**
* rdma_disconnect - This function disconnects the associated QP and
* transitions it into the error state.
*/
int rdma_disconnect(struct rdma_cm_id *id);
/**
* rdma_join_multicast - Join the multicast group specified by the given
* address.
* @id: Communication identifier associated with the request.
* @addr: Multicast address identifying the group to join.
* @join_state: Multicast JoinState bitmap requested by port.
* Bitmap is based on IB_SA_MCMEMBER_REC_JOIN_STATE bits.
* @context: User-defined context associated with the join request, returned
* to the user through the private_data pointer in multicast events.
*/
int rdma_join_multicast(struct rdma_cm_id *id, struct sockaddr *addr,
u8 join_state, void *context);
/**
* rdma_leave_multicast - Leave the multicast group specified by the given
* address.
*/
void rdma_leave_multicast(struct rdma_cm_id *id, struct sockaddr *addr);
/**
* rdma_set_service_type - Set the type of service associated with a
* connection identifier.
* @id: Communication identifier to associated with service type.
* @tos: Type of service.
*
* The type of service is interpretted as a differentiated service
* field (RFC 2474). The service type should be specified before
* performing route resolution, as existing communication on the
* connection identifier may be unaffected. The type of service
* requested may not be supported by the network to all destinations.
*/
void rdma_set_service_type(struct rdma_cm_id *id, int tos);
/**
* rdma_set_reuseaddr - Allow the reuse of local addresses when binding
* the rdma_cm_id.
* @id: Communication identifier to configure.
* @reuse: Value indicating if the bound address is reusable.
*
* Reuse must be set before an address is bound to the id.
*/
int rdma_set_reuseaddr(struct rdma_cm_id *id, int reuse);
/**
* rdma_set_afonly - Specify that listens are restricted to the
* bound address family only.
* @id: Communication identifer to configure.
* @afonly: Value indicating if listens are restricted.
*
* Must be set before identifier is in the listening state.
*/
int rdma_set_afonly(struct rdma_cm_id *id, int afonly);
/**
* rdma_get_service_id - Return the IB service ID for a specified address.
* @id: Communication identifier associated with the address.
* @addr: Address for the service ID.
*/
__be64 rdma_get_service_id(struct rdma_cm_id *id, struct sockaddr *addr);
/**
* rdma_reject_msg - return a pointer to a reject message string.
* @id: Communication identifier that received the REJECT event.
* @reason: Value returned in the REJECT event status field.
*/
const char *__attribute_const__ rdma_reject_msg(struct rdma_cm_id *id,
int reason);
/**
* rdma_is_consumer_reject - return true if the consumer rejected the connect
* request.
* @id: Communication identifier that received the REJECT event.
* @reason: Value returned in the REJECT event status field.
*/
bool rdma_is_consumer_reject(struct rdma_cm_id *id, int reason);
/**
* rdma_consumer_reject_data - return the consumer reject private data and
* length, if any.
* @id: Communication identifier that received the REJECT event.
* @ev: RDMA CM reject event.
* @data_len: Pointer to the resulting length of the consumer data.
*/
const void *rdma_consumer_reject_data(struct rdma_cm_id *id,
struct rdma_cm_event *ev, u8 *data_len);
/**
* rdma_read_gids - Return the SGID and DGID used for establishing
* connection. This can be used after rdma_resolve_addr()
* on client side. This can be use on new connection
* on server side. This is applicable to IB, RoCE, iWarp.
* If cm_id is not bound yet to the RDMA device, it doesn't
* copy and SGID or DGID to the given pointers.
* @id: Communication identifier whose GIDs are queried.
* @sgid: Pointer to SGID where SGID will be returned. It is optional.
* @dgid: Pointer to DGID where DGID will be returned. It is optional.
* Note: This API should not be used by any new ULPs or new code.
* Instead, users interested in querying GIDs should refer to path record
* of the rdma_cm_id to query the GIDs.
* This API is provided for compatibility for existing users.
*/
void rdma_read_gids(struct rdma_cm_id *cm_id, union ib_gid *sgid,
union ib_gid *dgid);
struct iw_cm_id *rdma_iw_cm_id(struct rdma_cm_id *cm_id);
struct rdma_cm_id *rdma_res_to_id(struct rdma_restrack_entry *res);
#endif /* RDMA_CM_H */
+54
View File
@@ -0,0 +1,54 @@
/*
* Copyright (c) 2006 Intel Corporation. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#if !defined(RDMA_CM_IB_H)
#define RDMA_CM_IB_H
#include <rdma/rdma_cm.h>
/**
* rdma_set_ib_path - Manually sets the path record used to establish a
* connection.
* @id: Connection identifier associated with the request.
* @path_rec: Reference to the path record
*
* This call permits a user to specify routing information for rdma_cm_id's
* bound to InfiniBand devices. It is called on the client side of a
* connection and replaces the call to rdma_resolve_route.
*/
int rdma_set_ib_path(struct rdma_cm_id *id,
struct sa_path_rec *path_rec);
/* Global qkey for UDP QPs and multicast groups. */
#define RDMA_UDP_QKEY 0x01234567
#endif /* RDMA_CM_IB_H */
+102
View File
@@ -0,0 +1,102 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _RDMA_NETLINK_H
#define _RDMA_NETLINK_H
#include <linux/netlink.h>
#include <uapi/rdma/rdma_netlink.h>
struct rdma_nl_cbs {
int (*doit)(struct sk_buff *skb, struct nlmsghdr *nlh,
struct netlink_ext_ack *extack);
int (*dump)(struct sk_buff *skb, struct netlink_callback *nlcb);
u8 flags;
};
enum rdma_nl_flags {
/* Require CAP_NET_ADMIN */
RDMA_NL_ADMIN_PERM = 1 << 0,
};
/* Define this module as providing netlink services for NETLINK_RDMA, with
* index _index. Since the client indexes were setup in a uapi header as an
* enum and we do no want to change that, the user must supply the expanded
* constant as well and the compiler checks they are the same.
*/
#define MODULE_ALIAS_RDMA_NETLINK(_index, _val) \
static inline void __chk_##_index(void) \
{ \
BUILD_BUG_ON(_index != _val); \
} \
MODULE_ALIAS("rdma-netlink-subsys-" __stringify(_val))
/**
* Register client in RDMA netlink.
* @index: Index of the added client
* @cb_table: A table for op->callback
*/
void rdma_nl_register(unsigned int index,
const struct rdma_nl_cbs cb_table[]);
/**
* Remove a client from IB netlink.
* @index: Index of the removed IB client.
*/
void rdma_nl_unregister(unsigned int index);
/**
* Put a new message in a supplied skb.
* @skb: The netlink skb.
* @nlh: Pointer to put the header of the new netlink message.
* @seq: The message sequence number.
* @len: The requested message length to allocate.
* @client: Calling IB netlink client.
* @op: message content op.
* Returns the allocated buffer on success and NULL on failure.
*/
void *ibnl_put_msg(struct sk_buff *skb, struct nlmsghdr **nlh, int seq,
int len, int client, int op, int flags);
/**
* Put a new attribute in a supplied skb.
* @skb: The netlink skb.
* @nlh: Header of the netlink message to append the attribute to.
* @len: The length of the attribute data.
* @data: The attribute data to put.
* @type: The attribute type.
* Returns the 0 and a negative error code on failure.
*/
int ibnl_put_attr(struct sk_buff *skb, struct nlmsghdr *nlh,
int len, void *data, int type);
/**
* Send the supplied skb to a specific userspace PID.
* @skb: The netlink skb
* @pid: Userspace netlink process ID
* Returns 0 on success or a negative error code.
*/
int rdma_nl_unicast(struct sk_buff *skb, u32 pid);
/**
* Send, with wait/1 retry, the supplied skb to a specific userspace PID.
* @skb: The netlink skb
* @pid: Userspace netlink process ID
* Returns 0 on success or a negative error code.
*/
int rdma_nl_unicast_wait(struct sk_buff *skb, __u32 pid);
/**
* Send the supplied skb to a netlink group.
* @skb: The netlink skb
* @group: Netlink group ID
* @flags: allocation flags
* Returns 0 on success or a negative error code.
*/
int rdma_nl_multicast(struct sk_buff *skb, unsigned int group, gfp_t flags);
/**
* Check if there are any listeners to the netlink group
* @group: the netlink group ID
* Returns 0 on success or a negative for no listeners.
*/
int rdma_nl_chk_listeners(unsigned int group);
#endif /* _RDMA_NETLINK_H */
+558
View File
@@ -0,0 +1,558 @@
#ifndef DEF_RDMA_VT_H
#define DEF_RDMA_VT_H
/*
* Copyright(c) 2016 - 2018 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
/*
* Structure that low level drivers will populate in order to register with the
* rdmavt layer.
*/
#include <linux/spinlock.h>
#include <linux/list.h>
#include <linux/hash.h>
#include <rdma/ib_verbs.h>
#include <rdma/ib_mad.h>
#include <rdma/rdmavt_mr.h>
#include <rdma/rdmavt_qp.h>
#define RVT_MAX_PKEY_VALUES 16
#define RVT_MAX_TRAP_LEN 100 /* Limit pending trap list */
#define RVT_MAX_TRAP_LISTS 5 /*((IB_NOTICE_TYPE_INFO & 0x0F) + 1)*/
#define RVT_TRAP_TIMEOUT 4096 /* 4.096 usec */
struct trap_list {
u32 list_len;
struct list_head list;
};
struct rvt_ibport {
struct rvt_qp __rcu *qp[2];
struct ib_mad_agent *send_agent; /* agent for SMI (traps) */
struct rb_root mcast_tree;
spinlock_t lock; /* protect changes in this struct */
/* non-zero when timer is set */
unsigned long mkey_lease_timeout;
unsigned long trap_timeout;
__be64 gid_prefix; /* in network order */
__be64 mkey;
u64 tid;
u32 port_cap_flags;
u16 port_cap3_flags;
u32 pma_sample_start;
u32 pma_sample_interval;
__be16 pma_counter_select[5];
u16 pma_tag;
u16 mkey_lease_period;
u32 sm_lid;
u8 sm_sl;
u8 mkeyprot;
u8 subnet_timeout;
u8 vl_high_limit;
/*
* Driver is expected to keep these up to date. These
* counters are informational only and not required to be
* completely accurate.
*/
u64 n_rc_resends;
u64 n_seq_naks;
u64 n_rdma_seq;
u64 n_rnr_naks;
u64 n_other_naks;
u64 n_loop_pkts;
u64 n_pkt_drops;
u64 n_vl15_dropped;
u64 n_rc_timeouts;
u64 n_dmawait;
u64 n_unaligned;
u64 n_rc_dupreq;
u64 n_rc_seqnak;
u16 pkey_violations;
u16 qkey_violations;
u16 mkey_violations;
/* Hot-path per CPU counters to avoid cacheline trading to update */
u64 z_rc_acks;
u64 z_rc_qacks;
u64 z_rc_delayed_comp;
u64 __percpu *rc_acks;
u64 __percpu *rc_qacks;
u64 __percpu *rc_delayed_comp;
void *priv; /* driver private data */
/*
* The pkey table is allocated and maintained by the driver. Drivers
* need to have access to this before registering with rdmav. However
* rdmavt will need access to it so drivers need to proviee this during
* the attach port API call.
*/
u16 *pkey_table;
struct rvt_ah *sm_ah;
/*
* Keep a list of traps that have not been repressed. They will be
* resent based on trap_timer.
*/
struct trap_list trap_lists[RVT_MAX_TRAP_LISTS];
struct timer_list trap_timer;
};
#define RVT_CQN_MAX 16 /* maximum length of cq name */
/*
* Things that are driver specific, module parameters in hfi1 and qib
*/
struct rvt_driver_params {
struct ib_device_attr props;
/*
* Anything driver specific that is not covered by props
* For instance special module parameters. Goes here.
*/
unsigned int lkey_table_size;
unsigned int qp_table_size;
int qpn_start;
int qpn_inc;
int qpn_res_start;
int qpn_res_end;
int nports;
int npkeys;
int node;
int psn_mask;
int psn_shift;
int psn_modify_mask;
u32 core_cap_flags;
u32 max_mad_size;
u8 qos_shift;
u8 max_rdma_atomic;
u8 reserved_operations;
};
/* Protection domain */
struct rvt_pd {
struct ib_pd ibpd;
bool user;
};
/* Address handle */
struct rvt_ah {
struct ib_ah ibah;
struct rdma_ah_attr attr;
atomic_t refcount;
u8 vl;
u8 log_pmtu;
};
struct rvt_dev_info;
struct rvt_swqe;
struct rvt_driver_provided {
/*
* Which functions are required depends on which verbs rdmavt is
* providing and which verbs the driver is overriding. See
* check_support() for details.
*/
/* hot path calldowns in a single cacheline */
/*
* Give the driver a notice that there is send work to do. It is up to
* the driver to generally push the packets out, this just queues the
* work with the driver. There are two variants here. The no_lock
* version requires the s_lock not to be held. The other assumes the
* s_lock is held.
*/
void (*schedule_send)(struct rvt_qp *qp);
void (*schedule_send_no_lock)(struct rvt_qp *qp);
/* Driver specific work request checking */
int (*check_send_wqe)(struct rvt_qp *qp, struct rvt_swqe *wqe);
/*
* Sometimes rdmavt needs to kick the driver's send progress. That is
* done by this call back.
*/
void (*do_send)(struct rvt_qp *qp);
/* Passed to ib core registration. Callback to create syfs files */
int (*port_callback)(struct ib_device *, u8, struct kobject *);
/*
* Returns a pointer to the undelying hardware's PCI device. This is
* used to display information as to what hardware is being referenced
* in an output message
*/
struct pci_dev * (*get_pci_dev)(struct rvt_dev_info *rdi);
/*
* Allocate a private queue pair data structure for driver specific
* information which is opaque to rdmavt. Errors are returned via
* ERR_PTR(err). The driver is free to return NULL or a valid
* pointer.
*/
void * (*qp_priv_alloc)(struct rvt_dev_info *rdi, struct rvt_qp *qp);
/*
* Free the driver's private qp structure.
*/
void (*qp_priv_free)(struct rvt_dev_info *rdi, struct rvt_qp *qp);
/*
* Inform the driver the particular qp in quesiton has been reset so
* that it can clean up anything it needs to.
*/
void (*notify_qp_reset)(struct rvt_qp *qp);
/*
* Get a path mtu from the driver based on qp attributes.
*/
int (*get_pmtu_from_attr)(struct rvt_dev_info *rdi, struct rvt_qp *qp,
struct ib_qp_attr *attr);
/*
* Notify driver that it needs to flush any outstanding IO requests that
* are waiting on a qp.
*/
void (*flush_qp_waiters)(struct rvt_qp *qp);
/*
* Notify driver to stop its queue of sending packets. Nothing else
* should be posted to the queue pair after this has been called.
*/
void (*stop_send_queue)(struct rvt_qp *qp);
/*
* Have the drivr drain any in progress operations
*/
void (*quiesce_qp)(struct rvt_qp *qp);
/*
* Inform the driver a qp has went to error state.
*/
void (*notify_error_qp)(struct rvt_qp *qp);
/*
* Get an MTU for a qp.
*/
u32 (*mtu_from_qp)(struct rvt_dev_info *rdi, struct rvt_qp *qp,
u32 pmtu);
/*
* Convert an mtu to a path mtu
*/
int (*mtu_to_path_mtu)(u32 mtu);
/*
* Get the guid of a port in big endian byte order
*/
int (*get_guid_be)(struct rvt_dev_info *rdi, struct rvt_ibport *rvp,
int guid_index, __be64 *guid);
/*
* Query driver for the state of the port.
*/
int (*query_port_state)(struct rvt_dev_info *rdi, u8 port_num,
struct ib_port_attr *props);
/*
* Tell driver to shutdown a port
*/
int (*shut_down_port)(struct rvt_dev_info *rdi, u8 port_num);
/* Tell driver to send a trap for changed port capabilities */
void (*cap_mask_chg)(struct rvt_dev_info *rdi, u8 port_num);
/*
* The following functions can be safely ignored completely. Any use of
* these is checked for NULL before blindly calling. Rdmavt should also
* be functional if drivers omit these.
*/
/* Called to inform the driver that all qps should now be freed. */
unsigned (*free_all_qps)(struct rvt_dev_info *rdi);
/* Driver specific AH validation */
int (*check_ah)(struct ib_device *, struct rdma_ah_attr *);
/* Inform the driver a new AH has been created */
void (*notify_new_ah)(struct ib_device *, struct rdma_ah_attr *,
struct rvt_ah *);
/* Let the driver pick the next queue pair number*/
int (*alloc_qpn)(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt,
enum ib_qp_type type, u8 port_num);
/* Determine if its safe or allowed to modify the qp */
int (*check_modify_qp)(struct rvt_qp *qp, struct ib_qp_attr *attr,
int attr_mask, struct ib_udata *udata);
/* Driver specific QP modification/notification-of */
void (*modify_qp)(struct rvt_qp *qp, struct ib_qp_attr *attr,
int attr_mask, struct ib_udata *udata);
/* Notify driver a mad agent has been created */
void (*notify_create_mad_agent)(struct rvt_dev_info *rdi, int port_idx);
/* Notify driver a mad agent has been removed */
void (*notify_free_mad_agent)(struct rvt_dev_info *rdi, int port_idx);
/* Notify driver to restart rc */
void (*notify_restart_rc)(struct rvt_qp *qp, u32 psn, int wait);
/* Get and return CPU to pin CQ processing thread */
int (*comp_vect_cpu_lookup)(struct rvt_dev_info *rdi, int comp_vect);
};
struct rvt_dev_info {
struct ib_device ibdev; /* Keep this first. Nothing above here */
/*
* Prior to calling for registration the driver will be responsible for
* allocating space for this structure.
*
* The driver will also be responsible for filling in certain members of
* dparms.props. The driver needs to fill in dparms exactly as it would
* want values reported to a ULP. This will be returned to the caller
* in rdmavt's device. The driver should also therefore refrain from
* modifying this directly after registration with rdmavt.
*/
/* Driver specific properties */
struct rvt_driver_params dparms;
/* post send table */
const struct rvt_operation_params *post_parms;
/* Driver specific helper functions */
struct rvt_driver_provided driver_f;
struct rvt_mregion __rcu *dma_mr;
struct rvt_lkey_table lkey_table;
/* Internal use */
int n_pds_allocated;
spinlock_t n_pds_lock; /* Protect pd allocated count */
int n_ahs_allocated;
spinlock_t n_ahs_lock; /* Protect ah allocated count */
u32 n_srqs_allocated;
spinlock_t n_srqs_lock; /* Protect srqs allocated count */
int flags;
struct rvt_ibport **ports;
/* QP */
struct rvt_qp_ibdev *qp_dev;
u32 n_qps_allocated; /* number of QPs allocated for device */
u32 n_rc_qps; /* number of RC QPs allocated for device */
u32 busy_jiffies; /* timeout scaling based on RC QP count */
spinlock_t n_qps_lock; /* protect qps, rc qps and busy jiffy counts */
/* memory maps */
struct list_head pending_mmaps;
spinlock_t mmap_offset_lock; /* protect mmap_offset */
u32 mmap_offset;
spinlock_t pending_lock; /* protect pending mmap list */
/* CQ */
u32 n_cqs_allocated; /* number of CQs allocated for device */
spinlock_t n_cqs_lock; /* protect count of in use cqs */
/* Multicast */
u32 n_mcast_grps_allocated; /* number of mcast groups allocated */
spinlock_t n_mcast_grps_lock;
};
/**
* rvt_set_ibdev_name - Craft an IB device name from client info
* @rdi: pointer to the client rvt_dev_info structure
* @name: client specific name
* @unit: client specific unit number.
*/
static inline void rvt_set_ibdev_name(struct rvt_dev_info *rdi,
const char *fmt, const char *name,
const int unit)
{
snprintf(rdi->ibdev.name, sizeof(rdi->ibdev.name), fmt, name, unit);
}
/**
* rvt_get_ibdev_name - return the IB name
* @rdi: rdmavt device
*
* Return the registered name of the device.
*/
static inline const char *rvt_get_ibdev_name(const struct rvt_dev_info *rdi)
{
return rdi->ibdev.name;
}
static inline struct rvt_pd *ibpd_to_rvtpd(struct ib_pd *ibpd)
{
return container_of(ibpd, struct rvt_pd, ibpd);
}
static inline struct rvt_ah *ibah_to_rvtah(struct ib_ah *ibah)
{
return container_of(ibah, struct rvt_ah, ibah);
}
static inline struct rvt_dev_info *ib_to_rvt(struct ib_device *ibdev)
{
return container_of(ibdev, struct rvt_dev_info, ibdev);
}
static inline struct rvt_srq *ibsrq_to_rvtsrq(struct ib_srq *ibsrq)
{
return container_of(ibsrq, struct rvt_srq, ibsrq);
}
static inline struct rvt_qp *ibqp_to_rvtqp(struct ib_qp *ibqp)
{
return container_of(ibqp, struct rvt_qp, ibqp);
}
static inline unsigned rvt_get_npkeys(struct rvt_dev_info *rdi)
{
/*
* All ports have same number of pkeys.
*/
return rdi->dparms.npkeys;
}
/*
* Return the max atomic suitable for determining
* the size of the ack ring buffer in a QP.
*/
static inline unsigned int rvt_max_atomic(struct rvt_dev_info *rdi)
{
return rdi->dparms.max_rdma_atomic + 1;
}
/*
* Return the indexed PKEY from the port PKEY table.
*/
static inline u16 rvt_get_pkey(struct rvt_dev_info *rdi,
int port_index,
unsigned index)
{
if (index >= rvt_get_npkeys(rdi))
return 0;
else
return rdi->ports[port_index]->pkey_table[index];
}
/**
* rvt_lookup_qpn - return the QP with the given QPN
* @ibp: the ibport
* @qpn: the QP number to look up
*
* The caller must hold the rcu_read_lock(), and keep the lock until
* the returned qp is no longer in use.
*/
/* TODO: Remove this and put in rdmavt/qp.h when no longer needed by drivers */
static inline struct rvt_qp *rvt_lookup_qpn(struct rvt_dev_info *rdi,
struct rvt_ibport *rvp,
u32 qpn) __must_hold(RCU)
{
struct rvt_qp *qp = NULL;
if (unlikely(qpn <= 1)) {
qp = rcu_dereference(rvp->qp[qpn]);
} else {
u32 n = hash_32(qpn, rdi->qp_dev->qp_table_bits);
for (qp = rcu_dereference(rdi->qp_dev->qp_table[n]); qp;
qp = rcu_dereference(qp->next))
if (qp->ibqp.qp_num == qpn)
break;
}
return qp;
}
/**
* rvt_mod_retry_timer - mod a retry timer
* @qp - the QP
* Modify a potentially already running retry timer
*/
static inline void rvt_mod_retry_timer(struct rvt_qp *qp)
{
struct ib_qp *ibqp = &qp->ibqp;
struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
lockdep_assert_held(&qp->s_lock);
qp->s_flags |= RVT_S_TIMER;
/* 4.096 usec. * (1 << qp->timeout) */
mod_timer(&qp->s_timer, jiffies + qp->timeout_jiffies +
rdi->busy_jiffies);
}
struct rvt_dev_info *rvt_alloc_device(size_t size, int nports);
void rvt_dealloc_device(struct rvt_dev_info *rdi);
int rvt_register_device(struct rvt_dev_info *rvd, u32 driver_id);
void rvt_unregister_device(struct rvt_dev_info *rvd);
int rvt_check_ah(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr);
int rvt_init_port(struct rvt_dev_info *rdi, struct rvt_ibport *port,
int port_index, u16 *pkey_table);
int rvt_fast_reg_mr(struct rvt_qp *qp, struct ib_mr *ibmr, u32 key,
int access);
int rvt_invalidate_rkey(struct rvt_qp *qp, u32 rkey);
int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge,
u32 len, u64 vaddr, u32 rkey, int acc);
int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd,
struct rvt_sge *isge, struct rvt_sge *last_sge,
struct ib_sge *sge, int acc);
struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid,
u16 lid);
#endif /* DEF_RDMA_VT_H */
+100
View File
@@ -0,0 +1,100 @@
#ifndef DEF_RDMAVT_INCCQ_H
#define DEF_RDMAVT_INCCQ_H
/*
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* Copyright(c) 2016 - 2018 Intel Corporation.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Copyright(c) 2015 Intel Corporation.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#include <linux/kthread.h>
#include <rdma/ib_user_verbs.h>
/*
* Define an ib_cq_notify value that is not valid so we know when CQ
* notifications are armed.
*/
#define RVT_CQ_NONE (IB_CQ_NEXT_COMP + 1)
/*
* This structure is used to contain the head pointer, tail pointer,
* and completion queue entries as a single memory allocation so
* it can be mmap'ed into user space.
*/
struct rvt_cq_wc {
u32 head; /* index of next entry to fill */
u32 tail; /* index of next ib_poll_cq() entry */
union {
/* these are actually size ibcq.cqe + 1 */
struct ib_uverbs_wc uqueue[0];
struct ib_wc kqueue[0];
};
};
/*
* The completion queue structure.
*/
struct rvt_cq {
struct ib_cq ibcq;
struct work_struct comptask;
spinlock_t lock; /* protect changes in this struct */
u8 notify;
u8 triggered;
int comp_vector_cpu;
struct rvt_dev_info *rdi;
struct rvt_cq_wc *queue;
struct rvt_mmap_info *ip;
};
static inline struct rvt_cq *ibcq_to_rvtcq(struct ib_cq *ibcq)
{
return container_of(ibcq, struct rvt_cq, ibcq);
}
void rvt_cq_enter(struct rvt_cq *cq, struct ib_wc *entry, bool solicited);
#endif /* DEF_RDMAVT_INCCQH */
+197
View File
@@ -0,0 +1,197 @@
#ifndef DEF_RDMAVT_INCMR_H
#define DEF_RDMAVT_INCMR_H
/*
* Copyright(c) 2016 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
/*
* For Memory Regions. This stuff should probably be moved into rdmavt/mr.h once
* drivers no longer need access to the MR directly.
*/
#include <linux/percpu-refcount.h>
/*
* A segment is a linear region of low physical memory.
* Used by the verbs layer.
*/
struct rvt_seg {
void *vaddr;
size_t length;
};
/* The number of rvt_segs that fit in a page. */
#define RVT_SEGSZ (PAGE_SIZE / sizeof(struct rvt_seg))
struct rvt_segarray {
struct rvt_seg segs[RVT_SEGSZ];
};
struct rvt_mregion {
struct ib_pd *pd; /* shares refcnt of ibmr.pd */
u64 user_base; /* User's address for this region */
u64 iova; /* IB start address of this region */
size_t length;
u32 lkey;
u32 offset; /* offset (bytes) to start of region */
int access_flags;
u32 max_segs; /* number of rvt_segs in all the arrays */
u32 mapsz; /* size of the map array */
atomic_t lkey_invalid; /* true if current lkey is invalid */
u8 page_shift; /* 0 - non unform/non powerof2 sizes */
u8 lkey_published; /* in global table */
struct percpu_ref refcount;
struct completion comp; /* complete when refcount goes to zero */
struct rvt_segarray *map[0]; /* the segments */
};
#define RVT_MAX_LKEY_TABLE_BITS 23
struct rvt_lkey_table {
/* read mostly fields */
u32 max; /* size of the table */
u32 shift; /* lkey/rkey shift */
struct rvt_mregion __rcu **table;
/* writeable fields */
/* protect changes in this struct */
spinlock_t lock ____cacheline_aligned_in_smp;
u32 next; /* next unused index (speeds search) */
u32 gen; /* generation count */
};
/*
* These keep track of the copy progress within a memory region.
* Used by the verbs layer.
*/
struct rvt_sge {
struct rvt_mregion *mr;
void *vaddr; /* kernel virtual address of segment */
u32 sge_length; /* length of the SGE */
u32 length; /* remaining length of the segment */
u16 m; /* current index: mr->map[m] */
u16 n; /* current index: mr->map[m]->segs[n] */
};
struct rvt_sge_state {
struct rvt_sge *sg_list; /* next SGE to be used if any */
struct rvt_sge sge; /* progress state for the current SGE */
u32 total_len;
u8 num_sge;
};
static inline void rvt_put_mr(struct rvt_mregion *mr)
{
percpu_ref_put(&mr->refcount);
}
static inline void rvt_get_mr(struct rvt_mregion *mr)
{
percpu_ref_get(&mr->refcount);
}
static inline void rvt_put_ss(struct rvt_sge_state *ss)
{
while (ss->num_sge) {
rvt_put_mr(ss->sge.mr);
if (--ss->num_sge)
ss->sge = *ss->sg_list++;
}
}
static inline u32 rvt_get_sge_length(struct rvt_sge *sge, u32 length)
{
u32 len = sge->length;
if (len > length)
len = length;
if (len > sge->sge_length)
len = sge->sge_length;
return len;
}
static inline void rvt_update_sge(struct rvt_sge_state *ss, u32 length,
bool release)
{
struct rvt_sge *sge = &ss->sge;
sge->vaddr += length;
sge->length -= length;
sge->sge_length -= length;
if (sge->sge_length == 0) {
if (release)
rvt_put_mr(sge->mr);
if (--ss->num_sge)
*sge = *ss->sg_list++;
} else if (sge->length == 0 && sge->mr->lkey) {
if (++sge->n >= RVT_SEGSZ) {
if (++sge->m >= sge->mr->mapsz)
return;
sge->n = 0;
}
sge->vaddr = sge->mr->map[sge->m]->segs[sge->n].vaddr;
sge->length = sge->mr->map[sge->m]->segs[sge->n].length;
}
}
static inline void rvt_skip_sge(struct rvt_sge_state *ss, u32 length,
bool release)
{
struct rvt_sge *sge = &ss->sge;
while (length) {
u32 len = rvt_get_sge_length(sge, length);
WARN_ON_ONCE(len == 0);
rvt_update_sge(ss, len, release);
length -= len;
}
}
bool rvt_ss_has_lkey(struct rvt_sge_state *ss, u32 lkey);
bool rvt_mr_has_lkey(struct rvt_mregion *mr, u32 lkey);
#endif /* DEF_RDMAVT_INCMRH */
+711
View File
@@ -0,0 +1,711 @@
#ifndef DEF_RDMAVT_INCQP_H
#define DEF_RDMAVT_INCQP_H
/*
* Copyright(c) 2016 - 2018 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* BSD LICENSE
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* - Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* - Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#include <rdma/rdma_vt.h>
#include <rdma/ib_pack.h>
#include <rdma/ib_verbs.h>
#include <rdma/rdmavt_cq.h>
/*
* Atomic bit definitions for r_aflags.
*/
#define RVT_R_WRID_VALID 0
#define RVT_R_REWIND_SGE 1
/*
* Bit definitions for r_flags.
*/
#define RVT_R_REUSE_SGE 0x01
#define RVT_R_RDMAR_SEQ 0x02
#define RVT_R_RSP_NAK 0x04
#define RVT_R_RSP_SEND 0x08
#define RVT_R_COMM_EST 0x10
/*
* Bit definitions for s_flags.
*
* RVT_S_SIGNAL_REQ_WR - set if QP send WRs contain completion signaled
* RVT_S_BUSY - send tasklet is processing the QP
* RVT_S_TIMER - the RC retry timer is active
* RVT_S_ACK_PENDING - an ACK is waiting to be sent after RDMA read/atomics
* RVT_S_WAIT_FENCE - waiting for all prior RDMA read or atomic SWQEs
* before processing the next SWQE
* RVT_S_WAIT_RDMAR - waiting for a RDMA read or atomic SWQE to complete
* before processing the next SWQE
* RVT_S_WAIT_RNR - waiting for RNR timeout
* RVT_S_WAIT_SSN_CREDIT - waiting for RC credits to process next SWQE
* RVT_S_WAIT_DMA - waiting for send DMA queue to drain before generating
* next send completion entry not via send DMA
* RVT_S_WAIT_PIO - waiting for a send buffer to be available
* RVT_S_WAIT_PIO_DRAIN - waiting for a qp to drain pio packets
* RVT_S_WAIT_TX - waiting for a struct verbs_txreq to be available
* RVT_S_WAIT_DMA_DESC - waiting for DMA descriptors to be available
* RVT_S_WAIT_KMEM - waiting for kernel memory to be available
* RVT_S_WAIT_PSN - waiting for a packet to exit the send DMA queue
* RVT_S_WAIT_ACK - waiting for an ACK packet before sending more requests
* RVT_S_SEND_ONE - send one packet, request ACK, then wait for ACK
* RVT_S_ECN - a BECN was queued to the send engine
* RVT_S_MAX_BIT_MASK - The max bit that can be used by rdmavt
*/
#define RVT_S_SIGNAL_REQ_WR 0x0001
#define RVT_S_BUSY 0x0002
#define RVT_S_TIMER 0x0004
#define RVT_S_RESP_PENDING 0x0008
#define RVT_S_ACK_PENDING 0x0010
#define RVT_S_WAIT_FENCE 0x0020
#define RVT_S_WAIT_RDMAR 0x0040
#define RVT_S_WAIT_RNR 0x0080
#define RVT_S_WAIT_SSN_CREDIT 0x0100
#define RVT_S_WAIT_DMA 0x0200
#define RVT_S_WAIT_PIO 0x0400
#define RVT_S_WAIT_TX 0x0800
#define RVT_S_WAIT_DMA_DESC 0x1000
#define RVT_S_WAIT_KMEM 0x2000
#define RVT_S_WAIT_PSN 0x4000
#define RVT_S_WAIT_ACK 0x8000
#define RVT_S_SEND_ONE 0x10000
#define RVT_S_UNLIMITED_CREDIT 0x20000
#define RVT_S_ECN 0x40000
#define RVT_S_MAX_BIT_MASK 0x800000
/*
* Drivers should use s_flags starting with bit 31 down to the bit next to
* RVT_S_MAX_BIT_MASK
*/
/*
* Wait flags that would prevent any packet type from being sent.
*/
#define RVT_S_ANY_WAIT_IO \
(RVT_S_WAIT_PIO | RVT_S_WAIT_TX | \
RVT_S_WAIT_DMA_DESC | RVT_S_WAIT_KMEM)
/*
* Wait flags that would prevent send work requests from making progress.
*/
#define RVT_S_ANY_WAIT_SEND (RVT_S_WAIT_FENCE | RVT_S_WAIT_RDMAR | \
RVT_S_WAIT_RNR | RVT_S_WAIT_SSN_CREDIT | RVT_S_WAIT_DMA | \
RVT_S_WAIT_PSN | RVT_S_WAIT_ACK)
#define RVT_S_ANY_WAIT (RVT_S_ANY_WAIT_IO | RVT_S_ANY_WAIT_SEND)
/* Number of bits to pay attention to in the opcode for checking qp type */
#define RVT_OPCODE_QP_MASK 0xE0
/* Flags for checking QP state (see ib_rvt_state_ops[]) */
#define RVT_POST_SEND_OK 0x01
#define RVT_POST_RECV_OK 0x02
#define RVT_PROCESS_RECV_OK 0x04
#define RVT_PROCESS_SEND_OK 0x08
#define RVT_PROCESS_NEXT_SEND_OK 0x10
#define RVT_FLUSH_SEND 0x20
#define RVT_FLUSH_RECV 0x40
#define RVT_PROCESS_OR_FLUSH_SEND \
(RVT_PROCESS_SEND_OK | RVT_FLUSH_SEND)
#define RVT_SEND_OR_FLUSH_OR_RECV_OK \
(RVT_PROCESS_SEND_OK | RVT_FLUSH_SEND | RVT_PROCESS_RECV_OK)
/*
* Internal send flags
*/
#define RVT_SEND_RESERVE_USED IB_SEND_RESERVED_START
#define RVT_SEND_COMPLETION_ONLY (IB_SEND_RESERVED_START << 1)
/*
* Send work request queue entry.
* The size of the sg_list is determined when the QP is created and stored
* in qp->s_max_sge.
*/
struct rvt_swqe {
union {
struct ib_send_wr wr; /* don't use wr.sg_list */
struct ib_ud_wr ud_wr;
struct ib_reg_wr reg_wr;
struct ib_rdma_wr rdma_wr;
struct ib_atomic_wr atomic_wr;
};
u32 psn; /* first packet sequence number */
u32 lpsn; /* last packet sequence number */
u32 ssn; /* send sequence number */
u32 length; /* total length of data in sg_list */
struct rvt_sge sg_list[0];
};
/*
* Receive work request queue entry.
* The size of the sg_list is determined when the QP (or SRQ) is created
* and stored in qp->r_rq.max_sge (or srq->rq.max_sge).
*/
struct rvt_rwqe {
u64 wr_id;
u8 num_sge;
struct ib_sge sg_list[0];
};
/*
* This structure is used to contain the head pointer, tail pointer,
* and receive work queue entries as a single memory allocation so
* it can be mmap'ed into user space.
* Note that the wq array elements are variable size so you can't
* just index into the array to get the N'th element;
* use get_rwqe_ptr() instead.
*/
struct rvt_rwq {
u32 head; /* new work requests posted to the head */
u32 tail; /* receives pull requests from here. */
struct rvt_rwqe wq[0];
};
struct rvt_rq {
struct rvt_rwq *wq;
u32 size; /* size of RWQE array */
u8 max_sge;
/* protect changes in this struct */
spinlock_t lock ____cacheline_aligned_in_smp;
};
/*
* This structure is used by rvt_mmap() to validate an offset
* when an mmap() request is made. The vm_area_struct then uses
* this as its vm_private_data.
*/
struct rvt_mmap_info {
struct list_head pending_mmaps;
struct ib_ucontext *context;
void *obj;
__u64 offset;
struct kref ref;
unsigned size;
};
/*
* This structure holds the information that the send tasklet needs
* to send a RDMA read response or atomic operation.
*/
struct rvt_ack_entry {
struct rvt_sge rdma_sge;
u64 atomic_data;
u32 psn;
u32 lpsn;
u8 opcode;
u8 sent;
};
#define RC_QP_SCALING_INTERVAL 5
#define RVT_OPERATION_PRIV 0x00000001
#define RVT_OPERATION_ATOMIC 0x00000002
#define RVT_OPERATION_ATOMIC_SGE 0x00000004
#define RVT_OPERATION_LOCAL 0x00000008
#define RVT_OPERATION_USE_RESERVE 0x00000010
#define RVT_OPERATION_MAX (IB_WR_RESERVED10 + 1)
/**
* rvt_operation_params - op table entry
* @length - the length to copy into the swqe entry
* @qpt_support - a bit mask indicating QP type support
* @flags - RVT_OPERATION flags (see above)
*
* This supports table driven post send so that
* the driver can have differing an potentially
* different sets of operations.
*
**/
struct rvt_operation_params {
size_t length;
u32 qpt_support;
u32 flags;
};
/*
* Common variables are protected by both r_rq.lock and s_lock in that order
* which only happens in modify_qp() or changing the QP 'state'.
*/
struct rvt_qp {
struct ib_qp ibqp;
void *priv; /* Driver private data */
/* read mostly fields above and below */
struct rdma_ah_attr remote_ah_attr;
struct rdma_ah_attr alt_ah_attr;
struct rvt_qp __rcu *next; /* link list for QPN hash table */
struct rvt_swqe *s_wq; /* send work queue */
struct rvt_mmap_info *ip;
unsigned long timeout_jiffies; /* computed from timeout */
int srate_mbps; /* s_srate (below) converted to Mbit/s */
pid_t pid; /* pid for user mode QPs */
u32 remote_qpn;
u32 qkey; /* QKEY for this QP (for UD or RD) */
u32 s_size; /* send work queue size */
u16 pmtu; /* decoded from path_mtu */
u8 log_pmtu; /* shift for pmtu */
u8 state; /* QP state */
u8 allowed_ops; /* high order bits of allowed opcodes */
u8 qp_access_flags;
u8 alt_timeout; /* Alternate path timeout for this QP */
u8 timeout; /* Timeout for this QP */
u8 s_srate;
u8 s_mig_state;
u8 port_num;
u8 s_pkey_index; /* PKEY index to use */
u8 s_alt_pkey_index; /* Alternate path PKEY index to use */
u8 r_max_rd_atomic; /* max number of RDMA read/atomic to receive */
u8 s_max_rd_atomic; /* max number of RDMA read/atomic to send */
u8 s_retry_cnt; /* number of times to retry */
u8 s_rnr_retry_cnt;
u8 r_min_rnr_timer; /* retry timeout value for RNR NAKs */
u8 s_max_sge; /* size of s_wq->sg_list */
u8 s_draining;
/* start of read/write fields */
atomic_t refcount ____cacheline_aligned_in_smp;
wait_queue_head_t wait;
struct rvt_ack_entry *s_ack_queue;
struct rvt_sge_state s_rdma_read_sge;
spinlock_t r_lock ____cacheline_aligned_in_smp; /* used for APM */
u32 r_psn; /* expected rcv packet sequence number */
unsigned long r_aflags;
u64 r_wr_id; /* ID for current receive WQE */
u32 r_ack_psn; /* PSN for next ACK or atomic ACK */
u32 r_len; /* total length of r_sge */
u32 r_rcv_len; /* receive data len processed */
u32 r_msn; /* message sequence number */
u8 r_state; /* opcode of last packet received */
u8 r_flags;
u8 r_head_ack_queue; /* index into s_ack_queue[] */
u8 r_adefered; /* defered ack count */
struct list_head rspwait; /* link for waiting to respond */
struct rvt_sge_state r_sge; /* current receive data */
struct rvt_rq r_rq; /* receive work queue */
/* post send line */
spinlock_t s_hlock ____cacheline_aligned_in_smp;
u32 s_head; /* new entries added here */
u32 s_next_psn; /* PSN for next request */
u32 s_avail; /* number of entries avail */
u32 s_ssn; /* SSN of tail entry */
atomic_t s_reserved_used; /* reserved entries in use */
spinlock_t s_lock ____cacheline_aligned_in_smp;
u32 s_flags;
struct rvt_sge_state *s_cur_sge;
struct rvt_swqe *s_wqe;
struct rvt_sge_state s_sge; /* current send request data */
struct rvt_mregion *s_rdma_mr;
u32 s_len; /* total length of s_sge */
u32 s_rdma_read_len; /* total length of s_rdma_read_sge */
u32 s_last_psn; /* last response PSN processed */
u32 s_sending_psn; /* lowest PSN that is being sent */
u32 s_sending_hpsn; /* highest PSN that is being sent */
u32 s_psn; /* current packet sequence number */
u32 s_ack_rdma_psn; /* PSN for sending RDMA read responses */
u32 s_ack_psn; /* PSN for acking sends and RDMA writes */
u32 s_tail; /* next entry to process */
u32 s_cur; /* current work queue entry */
u32 s_acked; /* last un-ACK'ed entry */
u32 s_last; /* last completed entry */
u32 s_lsn; /* limit sequence number (credit) */
u32 s_ahgpsn; /* set to the psn in the copy of the header */
u16 s_cur_size; /* size of send packet in bytes */
u16 s_rdma_ack_cnt;
u8 s_hdrwords; /* size of s_hdr in 32 bit words */
s8 s_ahgidx;
u8 s_state; /* opcode of last packet sent */
u8 s_ack_state; /* opcode of packet to ACK */
u8 s_nak_state; /* non-zero if NAK is pending */
u8 r_nak_state; /* non-zero if NAK is pending */
u8 s_retry; /* requester retry counter */
u8 s_rnr_retry; /* requester RNR retry counter */
u8 s_num_rd_atomic; /* number of RDMA read/atomic pending */
u8 s_tail_ack_queue; /* index into s_ack_queue[] */
struct rvt_sge_state s_ack_rdma_sge;
struct timer_list s_timer;
struct hrtimer s_rnr_timer;
atomic_t local_ops_pending; /* number of fast_reg/local_inv reqs */
/*
* This sge list MUST be last. Do not add anything below here.
*/
struct rvt_sge r_sg_list[0] /* verified SGEs */
____cacheline_aligned_in_smp;
};
struct rvt_srq {
struct ib_srq ibsrq;
struct rvt_rq rq;
struct rvt_mmap_info *ip;
/* send signal when number of RWQEs < limit */
u32 limit;
};
#define RVT_QPN_MAX BIT(24)
#define RVT_QPNMAP_ENTRIES (RVT_QPN_MAX / PAGE_SIZE / BITS_PER_BYTE)
#define RVT_BITS_PER_PAGE (PAGE_SIZE * BITS_PER_BYTE)
#define RVT_BITS_PER_PAGE_MASK (RVT_BITS_PER_PAGE - 1)
#define RVT_QPN_MASK IB_QPN_MASK
/*
* QPN-map pages start out as NULL, they get allocated upon
* first use and are never deallocated. This way,
* large bitmaps are not allocated unless large numbers of QPs are used.
*/
struct rvt_qpn_map {
void *page;
};
struct rvt_qpn_table {
spinlock_t lock; /* protect changes to the qp table */
unsigned flags; /* flags for QP0/1 allocated for each port */
u32 last; /* last QP number allocated */
u32 nmaps; /* size of the map table */
u16 limit;
u8 incr;
/* bit map of free QP numbers other than 0/1 */
struct rvt_qpn_map map[RVT_QPNMAP_ENTRIES];
};
struct rvt_qp_ibdev {
u32 qp_table_size;
u32 qp_table_bits;
struct rvt_qp __rcu **qp_table;
spinlock_t qpt_lock; /* qptable lock */
struct rvt_qpn_table qpn_table;
};
/*
* There is one struct rvt_mcast for each multicast GID.
* All attached QPs are then stored as a list of
* struct rvt_mcast_qp.
*/
struct rvt_mcast_qp {
struct list_head list;
struct rvt_qp *qp;
};
struct rvt_mcast_addr {
union ib_gid mgid;
u16 lid;
};
struct rvt_mcast {
struct rb_node rb_node;
struct rvt_mcast_addr mcast_addr;
struct list_head qp_list;
wait_queue_head_t wait;
atomic_t refcount;
int n_attached;
};
/*
* Since struct rvt_swqe is not a fixed size, we can't simply index into
* struct rvt_qp.s_wq. This function does the array index computation.
*/
static inline struct rvt_swqe *rvt_get_swqe_ptr(struct rvt_qp *qp,
unsigned n)
{
return (struct rvt_swqe *)((char *)qp->s_wq +
(sizeof(struct rvt_swqe) +
qp->s_max_sge *
sizeof(struct rvt_sge)) * n);
}
/*
* Since struct rvt_rwqe is not a fixed size, we can't simply index into
* struct rvt_rwq.wq. This function does the array index computation.
*/
static inline struct rvt_rwqe *rvt_get_rwqe_ptr(struct rvt_rq *rq, unsigned n)
{
return (struct rvt_rwqe *)
((char *)rq->wq->wq +
(sizeof(struct rvt_rwqe) +
rq->max_sge * sizeof(struct ib_sge)) * n);
}
/**
* rvt_is_user_qp - return if this is user mode QP
* @qp - the target QP
*/
static inline bool rvt_is_user_qp(struct rvt_qp *qp)
{
return !!qp->pid;
}
/**
* rvt_get_qp - get a QP reference
* @qp - the QP to hold
*/
static inline void rvt_get_qp(struct rvt_qp *qp)
{
atomic_inc(&qp->refcount);
}
/**
* rvt_put_qp - release a QP reference
* @qp - the QP to release
*/
static inline void rvt_put_qp(struct rvt_qp *qp)
{
if (qp && atomic_dec_and_test(&qp->refcount))
wake_up(&qp->wait);
}
/**
* rvt_put_swqe - drop mr refs held by swqe
* @wqe - the send wqe
*
* This drops any mr references held by the swqe
*/
static inline void rvt_put_swqe(struct rvt_swqe *wqe)
{
int i;
for (i = 0; i < wqe->wr.num_sge; i++) {
struct rvt_sge *sge = &wqe->sg_list[i];
rvt_put_mr(sge->mr);
}
}
/**
* rvt_qp_wqe_reserve - reserve operation
* @qp - the rvt qp
* @wqe - the send wqe
*
* This routine used in post send to record
* a wqe relative reserved operation use.
*/
static inline void rvt_qp_wqe_reserve(
struct rvt_qp *qp,
struct rvt_swqe *wqe)
{
atomic_inc(&qp->s_reserved_used);
}
/**
* rvt_qp_wqe_unreserve - clean reserved operation
* @qp - the rvt qp
* @wqe - the send wqe
*
* This decrements the reserve use count.
*
* This call MUST precede the change to
* s_last to insure that post send sees a stable
* s_avail.
*
* An smp_mp__after_atomic() is used to insure
* the compiler does not juggle the order of the s_last
* ring index and the decrementing of s_reserved_used.
*/
static inline void rvt_qp_wqe_unreserve(
struct rvt_qp *qp,
struct rvt_swqe *wqe)
{
if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) {
atomic_dec(&qp->s_reserved_used);
/* insure no compiler re-order up to s_last change */
smp_mb__after_atomic();
}
}
extern const enum ib_wc_opcode ib_rvt_wc_opcode[];
/**
* rvt_qp_swqe_complete() - insert send completion
* @qp - the qp
* @wqe - the send wqe
* @status - completion status
*
* Insert a send completion into the completion
* queue if the qp indicates it should be done.
*
* See IBTA 10.7.3.1 for info on completion
* control.
*/
static inline void rvt_qp_swqe_complete(
struct rvt_qp *qp,
struct rvt_swqe *wqe,
enum ib_wc_opcode opcode,
enum ib_wc_status status)
{
if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED))
return;
if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) ||
(wqe->wr.send_flags & IB_SEND_SIGNALED) ||
status != IB_WC_SUCCESS) {
struct ib_wc wc;
memset(&wc, 0, sizeof(wc));
wc.wr_id = wqe->wr.wr_id;
wc.status = status;
wc.opcode = opcode;
wc.qp = &qp->ibqp;
wc.byte_len = wqe->length;
rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc,
status != IB_WC_SUCCESS);
}
}
/*
* Compare the lower 24 bits of the msn values.
* Returns an integer <, ==, or > than zero.
*/
static inline int rvt_cmp_msn(u32 a, u32 b)
{
return (((int)a) - ((int)b)) << 8;
}
/**
* rvt_compute_aeth - compute the AETH (syndrome + MSN)
* @qp: the queue pair to compute the AETH for
*
* Returns the AETH.
*/
__be32 rvt_compute_aeth(struct rvt_qp *qp);
/**
* rvt_get_credit - flush the send work queue of a QP
* @qp: the qp who's send work queue to flush
* @aeth: the Acknowledge Extended Transport Header
*
* The QP s_lock should be held.
*/
void rvt_get_credit(struct rvt_qp *qp, u32 aeth);
/**
* @qp - the qp pair
* @len - the length
*
* Perform a shift based mtu round up divide
*/
static inline u32 rvt_div_round_up_mtu(struct rvt_qp *qp, u32 len)
{
return (len + qp->pmtu - 1) >> qp->log_pmtu;
}
/**
* @qp - the qp pair
* @len - the length
*
* Perform a shift based mtu divide
*/
static inline u32 rvt_div_mtu(struct rvt_qp *qp, u32 len)
{
return len >> qp->log_pmtu;
}
/**
* rvt_timeout_to_jiffies - Convert a ULP timeout input into jiffies
* @timeout - timeout input(0 - 31).
*
* Return a timeout value in jiffies.
*/
static inline unsigned long rvt_timeout_to_jiffies(u8 timeout)
{
if (timeout > 31)
timeout = 31;
return usecs_to_jiffies(1U << timeout) * 4096UL / 1000UL;
}
extern const int ib_rvt_state_ops[];
struct rvt_dev_info;
int rvt_get_rwqe(struct rvt_qp *qp, bool wr_id_only);
void rvt_comm_est(struct rvt_qp *qp);
int rvt_error_qp(struct rvt_qp *qp, enum ib_wc_status err);
void rvt_rc_error(struct rvt_qp *qp, enum ib_wc_status err);
unsigned long rvt_rnr_tbl_to_usec(u32 index);
enum hrtimer_restart rvt_rc_rnr_retry(struct hrtimer *t);
void rvt_add_rnr_timer(struct rvt_qp *qp, u32 aeth);
void rvt_del_timers_sync(struct rvt_qp *qp);
void rvt_stop_rc_timers(struct rvt_qp *qp);
void rvt_add_retry_timer(struct rvt_qp *qp);
/**
* struct rvt_qp_iter - the iterator for QPs
* @qp - the current QP
*
* This structure defines the current iterator
* state for sequenced access to all QPs relative
* to an rvt_dev_info.
*/
struct rvt_qp_iter {
struct rvt_qp *qp;
/* private: backpointer */
struct rvt_dev_info *rdi;
/* private: callback routine */
void (*cb)(struct rvt_qp *qp, u64 v);
/* private: for arg to callback routine */
u64 v;
/* private: number of SMI,GSI QPs for device */
int specials;
/* private: current iterator index */
int n;
};
struct rvt_qp_iter *rvt_qp_iter_init(struct rvt_dev_info *rdi,
u64 v,
void (*cb)(struct rvt_qp *qp, u64 v));
int rvt_qp_iter_next(struct rvt_qp_iter *iter);
void rvt_qp_iter(struct rvt_dev_info *rdi,
u64 v,
void (*cb)(struct rvt_qp *qp, u64 v));
void rvt_qp_mr_clean(struct rvt_qp *qp, u32 lkey);
#endif /* DEF_RDMAVT_INCQP_H */
+197
View File
@@ -0,0 +1,197 @@
/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
/*
* Copyright (c) 2017-2018 Mellanox Technologies. All rights reserved.
*/
#ifndef _RDMA_RESTRACK_H_
#define _RDMA_RESTRACK_H_
#include <linux/typecheck.h>
#include <linux/rwsem.h>
#include <linux/sched.h>
#include <linux/kref.h>
#include <linux/completion.h>
#include <linux/sched/task.h>
#include <uapi/rdma/rdma_netlink.h>
/**
* enum rdma_restrack_type - HW objects to track
*/
enum rdma_restrack_type {
/**
* @RDMA_RESTRACK_PD: Protection domain (PD)
*/
RDMA_RESTRACK_PD,
/**
* @RDMA_RESTRACK_CQ: Completion queue (CQ)
*/
RDMA_RESTRACK_CQ,
/**
* @RDMA_RESTRACK_QP: Queue pair (QP)
*/
RDMA_RESTRACK_QP,
/**
* @RDMA_RESTRACK_CM_ID: Connection Manager ID (CM_ID)
*/
RDMA_RESTRACK_CM_ID,
/**
* @RDMA_RESTRACK_MR: Memory Region (MR)
*/
RDMA_RESTRACK_MR,
/**
* @RDMA_RESTRACK_MAX: Last entry, used for array dclarations
*/
RDMA_RESTRACK_MAX
};
#define RDMA_RESTRACK_HASH_BITS 8
struct rdma_restrack_entry;
/**
* struct rdma_restrack_root - main resource tracking management
* entity, per-device
*/
struct rdma_restrack_root {
/*
* @rwsem: Read/write lock to protect lists
*/
struct rw_semaphore rwsem;
/**
* @hash: global database for all resources per-device
*/
DECLARE_HASHTABLE(hash, RDMA_RESTRACK_HASH_BITS);
/**
* @fill_res_entry: driver-specific fill function
*
* Allows rdma drivers to add their own restrack attributes.
*/
int (*fill_res_entry)(struct sk_buff *msg,
struct rdma_restrack_entry *entry);
};
/**
* struct rdma_restrack_entry - metadata per-entry
*/
struct rdma_restrack_entry {
/**
* @valid: validity indicator
*
* The entries are filled during rdma_restrack_add,
* can be attempted to be free during rdma_restrack_del.
*
* As an example for that, see mlx5 QPs with type MLX5_IB_QPT_HW_GSI
*/
bool valid;
/*
* @kref: Protect destroy of the resource
*/
struct kref kref;
/*
* @comp: Signal that all consumers of resource are completed their work
*/
struct completion comp;
/**
* @task: owner of resource tracking entity
*
* There are two types of entities: created by user and created
* by kernel.
*
* This is relevant for the entities created by users.
* For the entities created by kernel, this pointer will be NULL.
*/
struct task_struct *task;
/**
* @kern_name: name of owner for the kernel created entities.
*/
const char *kern_name;
/**
* @node: hash table entry
*/
struct hlist_node node;
/**
* @type: various objects in restrack database
*/
enum rdma_restrack_type type;
};
/**
* rdma_restrack_init() - initialize resource tracking
* @res: resource tracking root
*/
void rdma_restrack_init(struct rdma_restrack_root *res);
/**
* rdma_restrack_clean() - clean resource tracking
* @res: resource tracking root
*/
void rdma_restrack_clean(struct rdma_restrack_root *res);
/**
* rdma_restrack_count() - the current usage of specific object
* @res: resource entry
* @type: actual type of object to operate
* @ns: PID namespace
*/
int rdma_restrack_count(struct rdma_restrack_root *res,
enum rdma_restrack_type type,
struct pid_namespace *ns);
/**
* rdma_restrack_add() - add object to the reource tracking database
* @res: resource entry
*/
void rdma_restrack_add(struct rdma_restrack_entry *res);
/**
* rdma_restrack_del() - delete object from the reource tracking database
* @res: resource entry
* @type: actual type of object to operate
*/
void rdma_restrack_del(struct rdma_restrack_entry *res);
/**
* rdma_is_kernel_res() - check the owner of resource
* @res: resource entry
*/
static inline bool rdma_is_kernel_res(struct rdma_restrack_entry *res)
{
return !res->task;
}
/**
* rdma_restrack_get() - grab to protect resource from release
* @res: resource entry
*/
int __must_check rdma_restrack_get(struct rdma_restrack_entry *res);
/**
* rdma_restrack_put() - release resource
* @res: resource entry
*/
int rdma_restrack_put(struct rdma_restrack_entry *res);
/**
* rdma_restrack_set_task() - set the task for this resource
* @res: resource entry
* @task: task struct
*/
static inline void rdma_restrack_set_task(struct rdma_restrack_entry *res,
struct task_struct *task)
{
if (res->task)
put_task_struct(res->task);
get_task_struct(task);
res->task = task;
}
/*
* Helper functions for rdma drivers when filling out
* nldev driver attributes.
*/
int rdma_nl_put_driver_u32(struct sk_buff *msg, const char *name, u32 value);
int rdma_nl_put_driver_u32_hex(struct sk_buff *msg, const char *name,
u32 value);
int rdma_nl_put_driver_u64(struct sk_buff *msg, const char *name, u64 value);
int rdma_nl_put_driver_u64_hex(struct sk_buff *msg, const char *name,
u64 value);
#endif /* _RDMA_RESTRACK_H_ */
+90
View File
@@ -0,0 +1,90 @@
/*
* Copyright (c) 2016 HGST, a Western Digital Company.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*/
#ifndef _RDMA_RW_H
#define _RDMA_RW_H
#include <linux/dma-mapping.h>
#include <linux/scatterlist.h>
#include <rdma/ib_verbs.h>
#include <rdma/rdma_cm.h>
#include <rdma/mr_pool.h>
struct rdma_rw_ctx {
/* number of RDMA READ/WRITE WRs (not counting MR WRs) */
u32 nr_ops;
/* tag for the union below: */
u8 type;
union {
/* for mapping a single SGE: */
struct {
struct ib_sge sge;
struct ib_rdma_wr wr;
} single;
/* for mapping of multiple SGEs: */
struct {
struct ib_sge *sges;
struct ib_rdma_wr *wrs;
} map;
/* for registering multiple WRs: */
struct rdma_rw_reg_ctx {
struct ib_sge sge;
struct ib_rdma_wr wr;
struct ib_reg_wr reg_wr;
struct ib_send_wr inv_wr;
struct ib_mr *mr;
} *reg;
struct {
struct rdma_rw_reg_ctx data;
struct rdma_rw_reg_ctx prot;
struct ib_send_wr sig_inv_wr;
struct ib_mr *sig_mr;
struct ib_sge sig_sge;
struct ib_sig_handover_wr sig_wr;
} *sig;
};
};
int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num,
struct scatterlist *sg, u32 sg_cnt, u32 sg_offset,
u64 remote_addr, u32 rkey, enum dma_data_direction dir);
void rdma_rw_ctx_destroy(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num,
struct scatterlist *sg, u32 sg_cnt,
enum dma_data_direction dir);
int rdma_rw_ctx_signature_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
u8 port_num, struct scatterlist *sg, u32 sg_cnt,
struct scatterlist *prot_sg, u32 prot_sg_cnt,
struct ib_sig_attrs *sig_attrs, u64 remote_addr, u32 rkey,
enum dma_data_direction dir);
void rdma_rw_ctx_destroy_signature(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
u8 port_num, struct scatterlist *sg, u32 sg_cnt,
struct scatterlist *prot_sg, u32 prot_sg_cnt,
enum dma_data_direction dir);
struct ib_send_wr *rdma_rw_ctx_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
u8 port_num, struct ib_cqe *cqe, struct ib_send_wr *chain_wr);
int rdma_rw_ctx_post(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num,
struct ib_cqe *cqe, struct ib_send_wr *chain_wr);
unsigned int rdma_rw_mr_factor(struct ib_device *device, u8 port_num,
unsigned int maxpages);
void rdma_rw_init_qp(struct ib_device *dev, struct ib_qp_init_attr *attr);
int rdma_rw_init_mrs(struct ib_qp *qp, struct ib_qp_init_attr *attr);
void rdma_rw_cleanup_mrs(struct ib_qp *qp);
#endif /* _RDMA_RW_H */
+636
View File
@@ -0,0 +1,636 @@
/*
* Copyright (c) 2017, Mellanox Technologies inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _UVERBS_IOCTL_
#define _UVERBS_IOCTL_
#include <rdma/uverbs_types.h>
#include <linux/uaccess.h>
#include <rdma/rdma_user_ioctl.h>
#include <rdma/ib_user_ioctl_verbs.h>
#include <rdma/ib_user_ioctl_cmds.h>
/*
* =======================================
* Verbs action specifications
* =======================================
*/
enum uverbs_attr_type {
UVERBS_ATTR_TYPE_NA,
UVERBS_ATTR_TYPE_PTR_IN,
UVERBS_ATTR_TYPE_PTR_OUT,
UVERBS_ATTR_TYPE_IDR,
UVERBS_ATTR_TYPE_FD,
UVERBS_ATTR_TYPE_ENUM_IN,
};
enum uverbs_obj_access {
UVERBS_ACCESS_READ,
UVERBS_ACCESS_WRITE,
UVERBS_ACCESS_NEW,
UVERBS_ACCESS_DESTROY
};
/* Specification of a single attribute inside the ioctl message */
/* good size 16 */
struct uverbs_attr_spec {
u8 type;
/*
* Support extending attributes by length. Allow the user to provide
* more bytes than ptr.len, but check that everything after is zero'd
* by the user.
*/
u8 zero_trailing:1;
/*
* Valid only for PTR_IN. Allocate and copy the data inside
* the parser
*/
u8 alloc_and_copy:1;
u8 mandatory:1;
union {
struct {
/* Current known size to kernel */
u16 len;
/* User isn't allowed to provide something < min_len */
u16 min_len;
} ptr;
struct {
/*
* higher bits mean the namespace and lower bits mean
* the type id within the namespace.
*/
u16 obj_type;
u8 access;
} obj;
struct {
u8 num_elems;
} enum_def;
} u;
/* This weird split of the enum lets us remove some padding */
union {
struct {
/*
* The enum attribute can select one of the attributes
* contained in the ids array. Currently only PTR_IN
* attributes are supported in the ids array.
*/
const struct uverbs_attr_spec *ids;
} enum_def;
} u2;
};
/*
* Information about the API is loaded into a radix tree. For IOCTL we start
* with a tuple of:
* object_id, attr_id, method_id
*
* Which is a 48 bit value, with most of the bits guaranteed to be zero. Based
* on the current kernel support this is compressed into 16 bit key for the
* radix tree. Since this compression is entirely internal to the kernel the
* below limits can be revised if the kernel gains additional data.
*
* With 64 leafs per node this is a 3 level radix tree.
*
* The tree encodes multiple types, and uses a scheme where OBJ_ID,0,0 returns
* the object slot, and OBJ_ID,METH_ID,0 and returns the method slot.
*/
enum uapi_radix_data {
UVERBS_API_NS_FLAG = 1U << UVERBS_ID_NS_SHIFT,
UVERBS_API_ATTR_KEY_BITS = 6,
UVERBS_API_ATTR_KEY_MASK = GENMASK(UVERBS_API_ATTR_KEY_BITS - 1, 0),
UVERBS_API_ATTR_BKEY_LEN = (1 << UVERBS_API_ATTR_KEY_BITS) - 1,
UVERBS_API_METHOD_KEY_BITS = 5,
UVERBS_API_METHOD_KEY_SHIFT = UVERBS_API_ATTR_KEY_BITS,
UVERBS_API_METHOD_KEY_NUM_CORE = 24,
UVERBS_API_METHOD_KEY_NUM_DRIVER = (1 << UVERBS_API_METHOD_KEY_BITS) -
UVERBS_API_METHOD_KEY_NUM_CORE,
UVERBS_API_METHOD_KEY_MASK = GENMASK(
UVERBS_API_METHOD_KEY_BITS + UVERBS_API_METHOD_KEY_SHIFT - 1,
UVERBS_API_METHOD_KEY_SHIFT),
UVERBS_API_OBJ_KEY_BITS = 5,
UVERBS_API_OBJ_KEY_SHIFT =
UVERBS_API_METHOD_KEY_BITS + UVERBS_API_METHOD_KEY_SHIFT,
UVERBS_API_OBJ_KEY_NUM_CORE = 24,
UVERBS_API_OBJ_KEY_NUM_DRIVER =
(1 << UVERBS_API_OBJ_KEY_BITS) - UVERBS_API_OBJ_KEY_NUM_CORE,
UVERBS_API_OBJ_KEY_MASK = GENMASK(31, UVERBS_API_OBJ_KEY_SHIFT),
/* This id guaranteed to not exist in the radix tree */
UVERBS_API_KEY_ERR = 0xFFFFFFFF,
};
static inline __attribute_const__ u32 uapi_key_obj(u32 id)
{
if (id & UVERBS_API_NS_FLAG) {
id &= ~UVERBS_API_NS_FLAG;
if (id >= UVERBS_API_OBJ_KEY_NUM_DRIVER)
return UVERBS_API_KEY_ERR;
id = id + UVERBS_API_OBJ_KEY_NUM_CORE;
} else {
if (id >= UVERBS_API_OBJ_KEY_NUM_CORE)
return UVERBS_API_KEY_ERR;
}
return id << UVERBS_API_OBJ_KEY_SHIFT;
}
static inline __attribute_const__ bool uapi_key_is_object(u32 key)
{
return (key & ~UVERBS_API_OBJ_KEY_MASK) == 0;
}
static inline __attribute_const__ u32 uapi_key_ioctl_method(u32 id)
{
if (id & UVERBS_API_NS_FLAG) {
id &= ~UVERBS_API_NS_FLAG;
if (id >= UVERBS_API_METHOD_KEY_NUM_DRIVER)
return UVERBS_API_KEY_ERR;
id = id + UVERBS_API_METHOD_KEY_NUM_CORE;
} else {
id++;
if (id >= UVERBS_API_METHOD_KEY_NUM_CORE)
return UVERBS_API_KEY_ERR;
}
return id << UVERBS_API_METHOD_KEY_SHIFT;
}
static inline __attribute_const__ u32 uapi_key_attr_to_method(u32 attr_key)
{
return attr_key &
(UVERBS_API_OBJ_KEY_MASK | UVERBS_API_METHOD_KEY_MASK);
}
static inline __attribute_const__ bool uapi_key_is_ioctl_method(u32 key)
{
return (key & UVERBS_API_METHOD_KEY_MASK) != 0 &&
(key & UVERBS_API_ATTR_KEY_MASK) == 0;
}
static inline __attribute_const__ u32 uapi_key_attrs_start(u32 ioctl_method_key)
{
/* 0 is the method slot itself */
return ioctl_method_key + 1;
}
static inline __attribute_const__ u32 uapi_key_attr(u32 id)
{
/*
* The attr is designed to fit in the typical single radix tree node
* of 64 entries. Since allmost all methods have driver attributes we
* organize things so that the driver and core attributes interleave to
* reduce the length of the attributes array in typical cases.
*/
if (id & UVERBS_API_NS_FLAG) {
id &= ~UVERBS_API_NS_FLAG;
id++;
if (id >= 1 << (UVERBS_API_ATTR_KEY_BITS - 1))
return UVERBS_API_KEY_ERR;
id = (id << 1) | 0;
} else {
if (id >= 1 << (UVERBS_API_ATTR_KEY_BITS - 1))
return UVERBS_API_KEY_ERR;
id = (id << 1) | 1;
}
return id;
}
static inline __attribute_const__ bool uapi_key_is_attr(u32 key)
{
return (key & UVERBS_API_METHOD_KEY_MASK) != 0 &&
(key & UVERBS_API_ATTR_KEY_MASK) != 0;
}
/*
* This returns a value in the range [0 to UVERBS_API_ATTR_BKEY_LEN),
* basically it undoes the reservation of 0 in the ID numbering. attr_key
* must already be masked with UVERBS_API_ATTR_KEY_MASK, or be the output of
* uapi_key_attr().
*/
static inline __attribute_const__ u32 uapi_bkey_attr(u32 attr_key)
{
return attr_key - 1;
}
/*
* =======================================
* Verbs definitions
* =======================================
*/
struct uverbs_attr_def {
u16 id;
struct uverbs_attr_spec attr;
};
struct uverbs_method_def {
u16 id;
/* Combination of bits from enum UVERBS_ACTION_FLAG_XXXX */
u32 flags;
size_t num_attrs;
const struct uverbs_attr_def * const (*attrs)[];
int (*handler)(struct ib_uverbs_file *ufile,
struct uverbs_attr_bundle *ctx);
};
struct uverbs_object_def {
u16 id;
const struct uverbs_obj_type *type_attrs;
size_t num_methods;
const struct uverbs_method_def * const (*methods)[];
};
struct uverbs_object_tree_def {
size_t num_objects;
const struct uverbs_object_def * const (*objects)[];
};
/*
* =======================================
* Attribute Specifications
* =======================================
*/
#define UVERBS_ATTR_SIZE(_min_len, _len) \
.u.ptr.min_len = _min_len, .u.ptr.len = _len
#define UVERBS_ATTR_NO_DATA() UVERBS_ATTR_SIZE(0, 0)
/*
* Specifies a uapi structure that cannot be extended. The user must always
* supply the whole structure and nothing more. The structure must be declared
* in a header under include/uapi/rdma.
*/
#define UVERBS_ATTR_TYPE(_type) \
.u.ptr.min_len = sizeof(_type), .u.ptr.len = sizeof(_type)
/*
* Specifies a uapi structure where the user must provide at least up to
* member 'last'. Anything after last and up until the end of the structure
* can be non-zero, anything longer than the end of the structure must be
* zero. The structure must be declared in a header under include/uapi/rdma.
*/
#define UVERBS_ATTR_STRUCT(_type, _last) \
.zero_trailing = 1, \
UVERBS_ATTR_SIZE(((uintptr_t)(&((_type *)0)->_last + 1)), \
sizeof(_type))
/*
* Specifies at least min_len bytes must be passed in, but the amount can be
* larger, up to the protocol maximum size. No check for zeroing is done.
*/
#define UVERBS_ATTR_MIN_SIZE(_min_len) UVERBS_ATTR_SIZE(_min_len, USHRT_MAX)
/* Must be used in the '...' of any UVERBS_ATTR */
#define UA_ALLOC_AND_COPY .alloc_and_copy = 1
#define UA_MANDATORY .mandatory = 1
#define UA_OPTIONAL .mandatory = 0
#define UVERBS_ATTR_IDR(_attr_id, _idr_type, _access, ...) \
(&(const struct uverbs_attr_def){ \
.id = _attr_id, \
.attr = { .type = UVERBS_ATTR_TYPE_IDR, \
.u.obj.obj_type = _idr_type, \
.u.obj.access = _access, \
__VA_ARGS__ } })
#define UVERBS_ATTR_FD(_attr_id, _fd_type, _access, ...) \
(&(const struct uverbs_attr_def){ \
.id = (_attr_id) + \
BUILD_BUG_ON_ZERO((_access) != UVERBS_ACCESS_NEW && \
(_access) != UVERBS_ACCESS_READ), \
.attr = { .type = UVERBS_ATTR_TYPE_FD, \
.u.obj.obj_type = _fd_type, \
.u.obj.access = _access, \
__VA_ARGS__ } })
#define UVERBS_ATTR_PTR_IN(_attr_id, _type, ...) \
(&(const struct uverbs_attr_def){ \
.id = _attr_id, \
.attr = { .type = UVERBS_ATTR_TYPE_PTR_IN, \
_type, \
__VA_ARGS__ } })
#define UVERBS_ATTR_PTR_OUT(_attr_id, _type, ...) \
(&(const struct uverbs_attr_def){ \
.id = _attr_id, \
.attr = { .type = UVERBS_ATTR_TYPE_PTR_OUT, \
_type, \
__VA_ARGS__ } })
/* _enum_arry should be a 'static const union uverbs_attr_spec[]' */
#define UVERBS_ATTR_ENUM_IN(_attr_id, _enum_arr, ...) \
(&(const struct uverbs_attr_def){ \
.id = _attr_id, \
.attr = { .type = UVERBS_ATTR_TYPE_ENUM_IN, \
.u2.enum_def.ids = _enum_arr, \
.u.enum_def.num_elems = ARRAY_SIZE(_enum_arr), \
__VA_ARGS__ }, \
})
/*
* An input value that is a bitwise combination of values of _enum_type.
* This permits the flag value to be passed as either a u32 or u64, it must
* be retrieved via uverbs_get_flag().
*/
#define UVERBS_ATTR_FLAGS_IN(_attr_id, _enum_type, ...) \
UVERBS_ATTR_PTR_IN( \
_attr_id, \
UVERBS_ATTR_SIZE(sizeof(u32) + BUILD_BUG_ON_ZERO( \
!sizeof(_enum_type *)), \
sizeof(u64)), \
__VA_ARGS__)
/*
* This spec is used in order to pass information to the hardware driver in a
* legacy way. Every verb that could get driver specific data should get this
* spec.
*/
#define UVERBS_ATTR_UHW() \
UVERBS_ATTR_PTR_IN(UVERBS_ATTR_UHW_IN, \
UVERBS_ATTR_MIN_SIZE(0), \
UA_OPTIONAL), \
UVERBS_ATTR_PTR_OUT(UVERBS_ATTR_UHW_OUT, \
UVERBS_ATTR_MIN_SIZE(0), \
UA_OPTIONAL)
/*
* =======================================
* Declaration helpers
* =======================================
*/
#define DECLARE_UVERBS_OBJECT_TREE(_name, ...) \
static const struct uverbs_object_def *const _name##_ptr[] = { \
__VA_ARGS__, \
}; \
static const struct uverbs_object_tree_def _name = { \
.num_objects = ARRAY_SIZE(_name##_ptr), \
.objects = &_name##_ptr, \
}
/* =================================================
* Parsing infrastructure
* =================================================
*/
struct uverbs_ptr_attr {
/*
* If UVERBS_ATTR_SPEC_F_ALLOC_AND_COPY is set then the 'ptr' is
* used.
*/
union {
void *ptr;
u64 data;
};
u16 len;
u16 uattr_idx;
u8 enum_id;
};
struct uverbs_obj_attr {
struct ib_uobject *uobject;
const struct uverbs_api_attr *attr_elm;
};
struct uverbs_attr {
union {
struct uverbs_ptr_attr ptr_attr;
struct uverbs_obj_attr obj_attr;
};
};
struct uverbs_attr_bundle {
struct ib_uverbs_file *ufile;
DECLARE_BITMAP(attr_present, UVERBS_API_ATTR_BKEY_LEN);
struct uverbs_attr attrs[];
};
static inline bool uverbs_attr_is_valid(const struct uverbs_attr_bundle *attrs_bundle,
unsigned int idx)
{
return test_bit(uapi_bkey_attr(uapi_key_attr(idx)),
attrs_bundle->attr_present);
}
#define IS_UVERBS_COPY_ERR(_ret) ((_ret) && (_ret) != -ENOENT)
static inline const struct uverbs_attr *uverbs_attr_get(const struct uverbs_attr_bundle *attrs_bundle,
u16 idx)
{
if (!uverbs_attr_is_valid(attrs_bundle, idx))
return ERR_PTR(-ENOENT);
return &attrs_bundle->attrs[uapi_bkey_attr(uapi_key_attr(idx))];
}
static inline int uverbs_attr_get_enum_id(const struct uverbs_attr_bundle *attrs_bundle,
u16 idx)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return PTR_ERR(attr);
return attr->ptr_attr.enum_id;
}
static inline void *uverbs_attr_get_obj(const struct uverbs_attr_bundle *attrs_bundle,
u16 idx)
{
const struct uverbs_attr *attr;
attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return ERR_CAST(attr);
return attr->obj_attr.uobject->object;
}
static inline struct ib_uobject *uverbs_attr_get_uobject(const struct uverbs_attr_bundle *attrs_bundle,
u16 idx)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return ERR_CAST(attr);
return attr->obj_attr.uobject;
}
static inline int
uverbs_attr_get_len(const struct uverbs_attr_bundle *attrs_bundle, u16 idx)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return PTR_ERR(attr);
return attr->ptr_attr.len;
}
static inline bool uverbs_attr_ptr_is_inline(const struct uverbs_attr *attr)
{
return attr->ptr_attr.len <= sizeof(attr->ptr_attr.data);
}
static inline void *uverbs_attr_get_alloced_ptr(
const struct uverbs_attr_bundle *attrs_bundle, u16 idx)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return (void *)attr;
return uverbs_attr_ptr_is_inline(attr) ? (void *)&attr->ptr_attr.data :
attr->ptr_attr.ptr;
}
static inline int _uverbs_copy_from(void *to,
const struct uverbs_attr_bundle *attrs_bundle,
size_t idx,
size_t size)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
if (IS_ERR(attr))
return PTR_ERR(attr);
/*
* Validation ensures attr->ptr_attr.len >= size. If the caller is
* using UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO then it must call
* uverbs_copy_from_or_zero.
*/
if (unlikely(size < attr->ptr_attr.len))
return -EINVAL;
if (uverbs_attr_ptr_is_inline(attr))
memcpy(to, &attr->ptr_attr.data, attr->ptr_attr.len);
else if (copy_from_user(to, u64_to_user_ptr(attr->ptr_attr.data),
attr->ptr_attr.len))
return -EFAULT;
return 0;
}
static inline int _uverbs_copy_from_or_zero(void *to,
const struct uverbs_attr_bundle *attrs_bundle,
size_t idx,
size_t size)
{
const struct uverbs_attr *attr = uverbs_attr_get(attrs_bundle, idx);
size_t min_size;
if (IS_ERR(attr))
return PTR_ERR(attr);
min_size = min_t(size_t, size, attr->ptr_attr.len);
if (uverbs_attr_ptr_is_inline(attr))
memcpy(to, &attr->ptr_attr.data, min_size);
else if (copy_from_user(to, u64_to_user_ptr(attr->ptr_attr.data),
min_size))
return -EFAULT;
if (size > min_size)
memset(to + min_size, 0, size - min_size);
return 0;
}
#define uverbs_copy_from(to, attrs_bundle, idx) \
_uverbs_copy_from(to, attrs_bundle, idx, sizeof(*to))
#define uverbs_copy_from_or_zero(to, attrs_bundle, idx) \
_uverbs_copy_from_or_zero(to, attrs_bundle, idx, sizeof(*to))
#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
int uverbs_get_flags64(u64 *to, const struct uverbs_attr_bundle *attrs_bundle,
size_t idx, u64 allowed_bits);
int uverbs_get_flags32(u32 *to, const struct uverbs_attr_bundle *attrs_bundle,
size_t idx, u64 allowed_bits);
int uverbs_copy_to(const struct uverbs_attr_bundle *attrs_bundle, size_t idx,
const void *from, size_t size);
__malloc void *_uverbs_alloc(struct uverbs_attr_bundle *bundle, size_t size,
gfp_t flags);
static inline __malloc void *uverbs_alloc(struct uverbs_attr_bundle *bundle,
size_t size)
{
return _uverbs_alloc(bundle, size, GFP_KERNEL);
}
static inline __malloc void *uverbs_zalloc(struct uverbs_attr_bundle *bundle,
size_t size)
{
return _uverbs_alloc(bundle, size, GFP_KERNEL | __GFP_ZERO);
}
#else
static inline int
uverbs_get_flags64(u64 *to, const struct uverbs_attr_bundle *attrs_bundle,
size_t idx, u64 allowed_bits)
{
return -EINVAL;
}
static inline int
uverbs_get_flags32(u32 *to, const struct uverbs_attr_bundle *attrs_bundle,
size_t idx, u64 allowed_bits)
{
return -EINVAL;
}
static inline int uverbs_copy_to(const struct uverbs_attr_bundle *attrs_bundle,
size_t idx, const void *from, size_t size)
{
return -EINVAL;
}
static inline __malloc void *uverbs_alloc(struct uverbs_attr_bundle *bundle,
size_t size)
{
return ERR_PTR(-EINVAL);
}
static inline __malloc void *uverbs_zalloc(struct uverbs_attr_bundle *bundle,
size_t size)
{
return ERR_PTR(-EINVAL);
}
#endif
#endif
+131
View File
@@ -0,0 +1,131 @@
/*
* Copyright (c) 2018, Mellanox Technologies inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _UVERBS_NAMED_IOCTL_
#define _UVERBS_NAMED_IOCTL_
#include <rdma/uverbs_ioctl.h>
#ifndef UVERBS_MODULE_NAME
#error "Please #define UVERBS_MODULE_NAME before including rdma/uverbs_named_ioctl.h"
#endif
#define _UVERBS_PASTE(x, y) x ## y
#define _UVERBS_NAME(x, y) _UVERBS_PASTE(x, y)
#define UVERBS_METHOD(id) _UVERBS_NAME(UVERBS_MODULE_NAME, _method_##id)
#define UVERBS_HANDLER(id) _UVERBS_NAME(UVERBS_MODULE_NAME, _handler_##id)
#define UVERBS_OBJECT(id) _UVERBS_NAME(UVERBS_MOUDLE_NAME, _object_##id)
/* These are static so they do not need to be qualified */
#define UVERBS_METHOD_ATTRS(method_id) _method_attrs_##method_id
#define UVERBS_OBJECT_METHODS(object_id) _object_methods_##object_id
#define DECLARE_UVERBS_NAMED_METHOD(_method_id, ...) \
static const struct uverbs_attr_def *const UVERBS_METHOD_ATTRS( \
_method_id)[] = { __VA_ARGS__ }; \
static const struct uverbs_method_def UVERBS_METHOD(_method_id) = { \
.id = _method_id, \
.handler = UVERBS_HANDLER(_method_id), \
.num_attrs = ARRAY_SIZE(UVERBS_METHOD_ATTRS(_method_id)), \
.attrs = &UVERBS_METHOD_ATTRS(_method_id), \
}
/* Create a standard destroy method using the default handler. The handle_attr
* argument must be the attribute specifying the handle to destroy, the
* default handler does not support any other attributes.
*/
#define DECLARE_UVERBS_NAMED_METHOD_DESTROY(_method_id, _handle_attr) \
static const struct uverbs_attr_def *const UVERBS_METHOD_ATTRS( \
_method_id)[] = { _handle_attr }; \
static const struct uverbs_method_def UVERBS_METHOD(_method_id) = { \
.id = _method_id, \
.handler = uverbs_destroy_def_handler, \
.num_attrs = ARRAY_SIZE(UVERBS_METHOD_ATTRS(_method_id)), \
.attrs = &UVERBS_METHOD_ATTRS(_method_id), \
}
#define DECLARE_UVERBS_NAMED_OBJECT(_object_id, _type_attrs, ...) \
static const struct uverbs_method_def *const UVERBS_OBJECT_METHODS( \
_object_id)[] = { __VA_ARGS__ }; \
const struct uverbs_object_def UVERBS_OBJECT(_object_id) = { \
.id = _object_id, \
.type_attrs = &_type_attrs, \
.num_methods = ARRAY_SIZE(UVERBS_OBJECT_METHODS(_object_id)), \
.methods = &UVERBS_OBJECT_METHODS(_object_id) \
}
/*
* Declare global methods. These still have a unique object_id because we
* identify all uapi methods with a (object,method) tuple. However, they have
* no type pointer.
*/
#define DECLARE_UVERBS_GLOBAL_METHODS(_object_id, ...) \
static const struct uverbs_method_def *const UVERBS_OBJECT_METHODS( \
_object_id)[] = { __VA_ARGS__ }; \
const struct uverbs_object_def UVERBS_OBJECT(_object_id) = { \
.id = _object_id, \
.num_methods = ARRAY_SIZE(UVERBS_OBJECT_METHODS(_object_id)), \
.methods = &UVERBS_OBJECT_METHODS(_object_id) \
}
/* Used by drivers to declare a complete parsing tree for new methods
*/
#define ADD_UVERBS_METHODS(_name, _object_id, ...) \
static const struct uverbs_method_def *const UVERBS_OBJECT_METHODS( \
_object_id)[] = { __VA_ARGS__ }; \
static const struct uverbs_object_def _name##_struct = { \
.id = _object_id, \
.num_methods = ARRAY_SIZE(UVERBS_OBJECT_METHODS(_object_id)), \
.methods = &UVERBS_OBJECT_METHODS(_object_id) \
}; \
static const struct uverbs_object_def *const _name##_ptrs[] = { \
&_name##_struct, \
}; \
static const struct uverbs_object_tree_def _name = { \
.num_objects = 1, \
.objects = &_name##_ptrs, \
}
/* Used by drivers to declare a complete parsing tree for a single method that
* differs only in having additional driver specific attributes.
*/
#define ADD_UVERBS_ATTRIBUTES_SIMPLE(_name, _object_id, _method_id, ...) \
static const struct uverbs_attr_def *const UVERBS_METHOD_ATTRS( \
_method_id)[] = { __VA_ARGS__ }; \
static const struct uverbs_method_def UVERBS_METHOD(_method_id) = { \
.id = _method_id, \
.num_attrs = ARRAY_SIZE(UVERBS_METHOD_ATTRS(_method_id)), \
.attrs = &UVERBS_METHOD_ATTRS(_method_id), \
}; \
ADD_UVERBS_METHODS(_name, _object_id, &UVERBS_METHOD(_method_id))
#endif
+144
View File
@@ -0,0 +1,144 @@
/*
* Copyright (c) 2017, Mellanox Technologies inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _UVERBS_STD_TYPES__
#define _UVERBS_STD_TYPES__
#include <rdma/uverbs_types.h>
#include <rdma/uverbs_ioctl.h>
#include <rdma/ib_user_ioctl_verbs.h>
#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
const struct uverbs_object_tree_def *uverbs_default_get_objects(void);
#else
static inline const struct uverbs_object_tree_def *uverbs_default_get_objects(void)
{
return NULL;
}
#endif
/* Returns _id, or causes a compile error if _id is not a u32.
*
* The uobj APIs should only be used with the write based uAPI to access
* object IDs. The write API must use a u32 for the object handle, which is
* checked by this macro.
*/
#define _uobj_check_id(_id) ((_id) * typecheck(u32, _id))
#define uobj_get_type(_ufile, _object) \
uapi_get_object((_ufile)->device->uapi, _object)
#define uobj_get_read(_type, _id, _ufile) \
rdma_lookup_get_uobject(uobj_get_type(_ufile, _type), _ufile, \
_uobj_check_id(_id), UVERBS_LOOKUP_READ)
#define ufd_get_read(_type, _fdnum, _ufile) \
rdma_lookup_get_uobject(uobj_get_type(_ufile, _type), _ufile, \
(_fdnum)*typecheck(s32, _fdnum), \
UVERBS_LOOKUP_READ)
static inline void *_uobj_get_obj_read(struct ib_uobject *uobj)
{
if (IS_ERR(uobj))
return NULL;
return uobj->object;
}
#define uobj_get_obj_read(_object, _type, _id, _ufile) \
((struct ib_##_object *)_uobj_get_obj_read( \
uobj_get_read(_type, _id, _ufile)))
#define uobj_get_write(_type, _id, _ufile) \
rdma_lookup_get_uobject(uobj_get_type(_ufile, _type), _ufile, \
_uobj_check_id(_id), UVERBS_LOOKUP_WRITE)
int __uobj_perform_destroy(const struct uverbs_api_object *obj, u32 id,
struct ib_uverbs_file *ufile, int success_res);
#define uobj_perform_destroy(_type, _id, _ufile, _success_res) \
__uobj_perform_destroy(uobj_get_type(_ufile, _type), \
_uobj_check_id(_id), _ufile, _success_res)
struct ib_uobject *__uobj_get_destroy(const struct uverbs_api_object *obj,
u32 id, struct ib_uverbs_file *ufile);
#define uobj_get_destroy(_type, _id, _ufile) \
__uobj_get_destroy(uobj_get_type(_ufile, _type), _uobj_check_id(_id), \
_ufile)
static inline void uobj_put_destroy(struct ib_uobject *uobj)
{
rdma_lookup_put_uobject(uobj, UVERBS_LOOKUP_DESTROY);
}
static inline void uobj_put_read(struct ib_uobject *uobj)
{
rdma_lookup_put_uobject(uobj, UVERBS_LOOKUP_READ);
}
#define uobj_put_obj_read(_obj) \
uobj_put_read((_obj)->uobject)
static inline void uobj_put_write(struct ib_uobject *uobj)
{
rdma_lookup_put_uobject(uobj, UVERBS_LOOKUP_WRITE);
}
static inline int __must_check uobj_alloc_commit(struct ib_uobject *uobj,
int success_res)
{
int ret = rdma_alloc_commit_uobject(uobj);
if (ret)
return ret;
return success_res;
}
static inline void uobj_alloc_abort(struct ib_uobject *uobj)
{
rdma_alloc_abort_uobject(uobj);
}
static inline struct ib_uobject *
__uobj_alloc(const struct uverbs_api_object *obj, struct ib_uverbs_file *ufile,
struct ib_device **ib_dev)
{
struct ib_uobject *uobj = rdma_alloc_begin_uobject(obj, ufile);
if (!IS_ERR(uobj))
*ib_dev = uobj->context->device;
return uobj;
}
#define uobj_alloc(_type, _ufile, _ib_dev) \
__uobj_alloc(uobj_get_type(_ufile, _type), _ufile, _ib_dev)
#endif
+188
View File
@@ -0,0 +1,188 @@
/*
* Copyright (c) 2017, Mellanox Technologies inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef _UVERBS_TYPES_
#define _UVERBS_TYPES_
#include <linux/kernel.h>
#include <rdma/ib_verbs.h>
struct uverbs_obj_type;
struct uverbs_api_object;
enum rdma_lookup_mode {
UVERBS_LOOKUP_READ,
UVERBS_LOOKUP_WRITE,
/*
* Destroy is like LOOKUP_WRITE, except that the uobject is not
* locked. uobj_destroy is used to convert a LOOKUP_DESTROY lock into
* a LOOKUP_WRITE lock.
*/
UVERBS_LOOKUP_DESTROY,
};
/*
* The following sequences are valid:
* Success flow:
* alloc_begin
* alloc_commit
* [..]
* Access flow:
* lookup_get(exclusive=false) & uverbs_try_lock_object
* lookup_put(exclusive=false) via rdma_lookup_put_uobject
* Destruction flow:
* lookup_get(exclusive=true) & uverbs_try_lock_object
* remove_commit
* remove_handle (optional)
* lookup_put(exclusive=true) via rdma_lookup_put_uobject
*
* Allocate Error flow #1
* alloc_begin
* alloc_abort
* Allocate Error flow #2
* alloc_begin
* remove_commit
* alloc_abort
* Allocate Error flow #3
* alloc_begin
* alloc_commit (fails)
* remove_commit
* alloc_abort
*
* In all cases the caller must hold the ufile kref until alloc_commit or
* alloc_abort returns.
*/
struct uverbs_obj_type_class {
struct ib_uobject *(*alloc_begin)(const struct uverbs_api_object *obj,
struct ib_uverbs_file *ufile);
/* This consumes the kref on uobj */
int (*alloc_commit)(struct ib_uobject *uobj);
/* This does not consume the kref on uobj */
void (*alloc_abort)(struct ib_uobject *uobj);
struct ib_uobject *(*lookup_get)(const struct uverbs_api_object *obj,
struct ib_uverbs_file *ufile, s64 id,
enum rdma_lookup_mode mode);
void (*lookup_put)(struct ib_uobject *uobj, enum rdma_lookup_mode mode);
/* This does not consume the kref on uobj */
int __must_check (*destroy_hw)(struct ib_uobject *uobj,
enum rdma_remove_reason why);
void (*remove_handle)(struct ib_uobject *uobj);
u8 needs_kfree_rcu;
};
struct uverbs_obj_type {
const struct uverbs_obj_type_class * const type_class;
size_t obj_size;
};
/*
* Objects type classes which support a detach state (object is still alive but
* it's not attached to any context need to make sure:
* (a) no call through to a driver after a detach is called
* (b) detach isn't called concurrently with context_cleanup
*/
struct uverbs_obj_idr_type {
/*
* In idr based objects, uverbs_obj_type_class points to a generic
* idr operations. In order to specialize the underlying types (e.g. CQ,
* QPs, etc.), we add destroy_object specific callbacks.
*/
struct uverbs_obj_type type;
/* Free driver resources from the uobject, make the driver uncallable,
* and move the uobject to the detached state. If the object was
* destroyed by the user's request, a failure should leave the uobject
* completely unchanged.
*/
int __must_check (*destroy_object)(struct ib_uobject *uobj,
enum rdma_remove_reason why);
};
struct ib_uobject *rdma_lookup_get_uobject(const struct uverbs_api_object *obj,
struct ib_uverbs_file *ufile, s64 id,
enum rdma_lookup_mode mode);
void rdma_lookup_put_uobject(struct ib_uobject *uobj,
enum rdma_lookup_mode mode);
struct ib_uobject *rdma_alloc_begin_uobject(const struct uverbs_api_object *obj,
struct ib_uverbs_file *ufile);
void rdma_alloc_abort_uobject(struct ib_uobject *uobj);
int __must_check rdma_alloc_commit_uobject(struct ib_uobject *uobj);
struct uverbs_obj_fd_type {
/*
* In fd based objects, uverbs_obj_type_ops points to generic
* fd operations. In order to specialize the underlying types (e.g.
* completion_channel), we use fops, name and flags for fd creation.
* context_closed is called when the context is closed either when
* the driver is removed or the process terminated.
*/
struct uverbs_obj_type type;
int (*context_closed)(struct ib_uobject *uobj,
enum rdma_remove_reason why);
const struct file_operations *fops;
const char *name;
int flags;
};
extern const struct uverbs_obj_type_class uverbs_idr_class;
extern const struct uverbs_obj_type_class uverbs_fd_class;
#define UVERBS_BUILD_BUG_ON(cond) (sizeof(char[1 - 2 * !!(cond)]) - \
sizeof(char))
#define UVERBS_TYPE_ALLOC_FD(_obj_size, _context_closed, _fops, _name, _flags)\
((&((const struct uverbs_obj_fd_type) \
{.type = { \
.type_class = &uverbs_fd_class, \
.obj_size = (_obj_size) + \
UVERBS_BUILD_BUG_ON((_obj_size) < \
sizeof(struct ib_uobject)), \
}, \
.context_closed = _context_closed, \
.fops = _fops, \
.name = _name, \
.flags = _flags}))->type)
#define UVERBS_TYPE_ALLOC_IDR_SZ(_size, _destroy_object) \
((&((const struct uverbs_obj_idr_type) \
{.type = { \
.type_class = &uverbs_idr_class, \
.obj_size = (_size) + \
UVERBS_BUILD_BUG_ON((_size) < \
sizeof(struct ib_uobject)) \
}, \
.destroy_object = _destroy_object,}))->type)
#define UVERBS_TYPE_ALLOC_IDR(_destroy_object) \
UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_uobject), \
_destroy_object)
#endif