2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* 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.
|
2005-08-11 00:03:10 -06:00
|
|
|
* Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
|
2007-03-04 17:15:11 -07:00
|
|
|
* Copyright (c) 2005, 2006, 2007 Cisco Systems. All rights reserved.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
|
|
|
* 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_VERBS_H)
|
|
|
|
#define IB_VERBS_H
|
|
|
|
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/device.h>
|
2006-12-12 15:27:41 -07:00
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/dma-mapping.h>
|
2007-02-04 15:11:55 -07:00
|
|
|
#include <linux/kref.h>
|
2007-07-31 07:49:15 -06:00
|
|
|
#include <linux/list.h>
|
|
|
|
#include <linux/rwsem.h>
|
2007-10-30 03:35:04 -06:00
|
|
|
#include <linux/scatterlist.h>
|
2010-10-19 09:24:36 -06:00
|
|
|
#include <linux/workqueue.h>
|
2015-07-30 08:50:15 -06:00
|
|
|
#include <linux/socket.h>
|
2015-12-11 12:53:03 -07:00
|
|
|
#include <linux/irq_poll.h>
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
#include <uapi/linux/if_ether.h>
|
2015-12-23 05:56:51 -07:00
|
|
|
#include <net/ipv6.h>
|
|
|
|
#include <net/ip.h>
|
2015-12-15 11:30:10 -07:00
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/slab.h>
|
2005-07-07 18:57:10 -06:00
|
|
|
|
2016-03-11 13:58:38 -07:00
|
|
|
#include <linux/if_link.h>
|
2011-07-26 17:09:06 -06:00
|
|
|
#include <linux/atomic.h>
|
2014-12-11 08:04:18 -07:00
|
|
|
#include <linux/mmu_notifier.h>
|
2005-07-07 18:57:10 -06:00
|
|
|
#include <asm/uaccess.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-19 09:24:36 -06:00
|
|
|
extern struct workqueue_struct *ib_wq;
|
2015-12-11 12:53:03 -07:00
|
|
|
extern struct workqueue_struct *ib_comp_wq;
|
2010-10-19 09:24:36 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
union ib_gid {
|
|
|
|
u8 raw[16];
|
|
|
|
struct {
|
2005-08-13 22:05:57 -06:00
|
|
|
__be64 subnet_prefix;
|
|
|
|
__be64 interface_id;
|
2005-04-16 16:20:36 -06:00
|
|
|
} global;
|
|
|
|
};
|
|
|
|
|
2015-07-30 09:33:29 -06:00
|
|
|
extern union ib_gid zgid;
|
|
|
|
|
2015-12-23 05:56:47 -07:00
|
|
|
enum ib_gid_type {
|
|
|
|
/* If link layer is Ethernet, this is RoCE V1 */
|
|
|
|
IB_GID_TYPE_IB = 0,
|
|
|
|
IB_GID_TYPE_ROCE = 0,
|
2015-12-23 05:56:50 -07:00
|
|
|
IB_GID_TYPE_ROCE_UDP_ENCAP = 1,
|
2015-12-23 05:56:47 -07:00
|
|
|
IB_GID_TYPE_SIZE
|
|
|
|
};
|
|
|
|
|
2016-01-14 08:50:38 -07:00
|
|
|
#define ROCE_V2_UDP_DPORT 4791
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
struct ib_gid_attr {
|
2015-12-23 05:56:47 -07:00
|
|
|
enum ib_gid_type gid_type;
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
struct net_device *ndev;
|
|
|
|
};
|
|
|
|
|
2006-08-03 15:02:42 -06:00
|
|
|
enum rdma_node_type {
|
|
|
|
/* IB values map to NodeInfo:NodeType. */
|
|
|
|
RDMA_NODE_IB_CA = 1,
|
|
|
|
RDMA_NODE_IB_SWITCH,
|
|
|
|
RDMA_NODE_IB_ROUTER,
|
2013-09-09 21:36:59 -06:00
|
|
|
RDMA_NODE_RNIC,
|
|
|
|
RDMA_NODE_USNIC,
|
2014-01-15 18:02:36 -07:00
|
|
|
RDMA_NODE_USNIC_UDP,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2016-03-11 13:58:37 -07:00
|
|
|
enum {
|
|
|
|
/* set the local administered indication */
|
|
|
|
IB_SA_WELL_KNOWN_GUID = BIT_ULL(57) | 2,
|
|
|
|
};
|
|
|
|
|
2006-08-03 15:02:42 -06:00
|
|
|
enum rdma_transport_type {
|
|
|
|
RDMA_TRANSPORT_IB,
|
2013-09-09 21:36:59 -06:00
|
|
|
RDMA_TRANSPORT_IWARP,
|
2014-01-09 15:48:19 -07:00
|
|
|
RDMA_TRANSPORT_USNIC,
|
|
|
|
RDMA_TRANSPORT_USNIC_UDP
|
2006-08-03 15:02:42 -06:00
|
|
|
};
|
|
|
|
|
2015-05-05 06:50:18 -06:00
|
|
|
enum rdma_protocol_type {
|
|
|
|
RDMA_PROTOCOL_IB,
|
|
|
|
RDMA_PROTOCOL_IBOE,
|
|
|
|
RDMA_PROTOCOL_IWARP,
|
|
|
|
RDMA_PROTOCOL_USNIC_UDP
|
|
|
|
};
|
|
|
|
|
2014-06-04 11:00:16 -06:00
|
|
|
__attribute_const__ enum rdma_transport_type
|
|
|
|
rdma_node_get_transport(enum rdma_node_type node_type);
|
2006-08-03 15:02:42 -06:00
|
|
|
|
2015-12-23 05:56:51 -07:00
|
|
|
enum rdma_network_type {
|
|
|
|
RDMA_NETWORK_IB,
|
|
|
|
RDMA_NETWORK_ROCE_V1 = RDMA_NETWORK_IB,
|
|
|
|
RDMA_NETWORK_IPV4,
|
|
|
|
RDMA_NETWORK_IPV6
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline enum ib_gid_type ib_network_to_gid_type(enum rdma_network_type network_type)
|
|
|
|
{
|
|
|
|
if (network_type == RDMA_NETWORK_IPV4 ||
|
|
|
|
network_type == RDMA_NETWORK_IPV6)
|
|
|
|
return IB_GID_TYPE_ROCE_UDP_ENCAP;
|
|
|
|
|
|
|
|
/* IB_GID_TYPE_IB same as RDMA_NETWORK_ROCE_V1 */
|
|
|
|
return IB_GID_TYPE_IB;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline enum rdma_network_type ib_gid_to_network_type(enum ib_gid_type gid_type,
|
|
|
|
union ib_gid *gid)
|
|
|
|
{
|
|
|
|
if (gid_type == IB_GID_TYPE_IB)
|
|
|
|
return RDMA_NETWORK_IB;
|
|
|
|
|
|
|
|
if (ipv6_addr_v4mapped((struct in6_addr *)gid))
|
|
|
|
return RDMA_NETWORK_IPV4;
|
|
|
|
else
|
|
|
|
return RDMA_NETWORK_IPV6;
|
|
|
|
}
|
|
|
|
|
2010-09-27 18:51:10 -06:00
|
|
|
enum rdma_link_layer {
|
|
|
|
IB_LINK_LAYER_UNSPECIFIED,
|
|
|
|
IB_LINK_LAYER_INFINIBAND,
|
|
|
|
IB_LINK_LAYER_ETHERNET,
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_device_cap_flags {
|
2015-12-20 03:16:09 -07:00
|
|
|
IB_DEVICE_RESIZE_MAX_WR = (1 << 0),
|
|
|
|
IB_DEVICE_BAD_PKEY_CNTR = (1 << 1),
|
|
|
|
IB_DEVICE_BAD_QKEY_CNTR = (1 << 2),
|
|
|
|
IB_DEVICE_RAW_MULTI = (1 << 3),
|
|
|
|
IB_DEVICE_AUTO_PATH_MIG = (1 << 4),
|
|
|
|
IB_DEVICE_CHANGE_PHY_PORT = (1 << 5),
|
|
|
|
IB_DEVICE_UD_AV_PORT_ENFORCE = (1 << 6),
|
|
|
|
IB_DEVICE_CURR_QP_STATE_MOD = (1 << 7),
|
|
|
|
IB_DEVICE_SHUTDOWN_PORT = (1 << 8),
|
|
|
|
IB_DEVICE_INIT_TYPE = (1 << 9),
|
|
|
|
IB_DEVICE_PORT_ACTIVE_EVENT = (1 << 10),
|
|
|
|
IB_DEVICE_SYS_IMAGE_GUID = (1 << 11),
|
|
|
|
IB_DEVICE_RC_RNR_NAK_GEN = (1 << 12),
|
|
|
|
IB_DEVICE_SRQ_RESIZE = (1 << 13),
|
|
|
|
IB_DEVICE_N_NOTIFY_CQ = (1 << 14),
|
2015-12-23 11:12:45 -07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* This device supports a per-device lkey or stag that can be
|
|
|
|
* used without performing a memory registration for the local
|
|
|
|
* memory. Note that ULPs should never check this flag, but
|
|
|
|
* instead of use the local_dma_lkey flag in the ib_pd structure,
|
|
|
|
* which will always contain a usable lkey.
|
|
|
|
*/
|
2015-12-20 03:16:09 -07:00
|
|
|
IB_DEVICE_LOCAL_DMA_LKEY = (1 << 15),
|
|
|
|
IB_DEVICE_RESERVED /* old SEND_W_INV */ = (1 << 16),
|
|
|
|
IB_DEVICE_MEM_WINDOW = (1 << 17),
|
2008-01-30 09:30:57 -07:00
|
|
|
/*
|
|
|
|
* Devices should set IB_DEVICE_UD_IP_SUM if they support
|
|
|
|
* insertion of UDP and TCP checksum on outgoing UD IPoIB
|
|
|
|
* messages and can verify the validity of checksum for
|
|
|
|
* incoming messages. Setting this flag implies that the
|
|
|
|
* IPoIB driver may set NETIF_F_IP_CSUM for datagram mode.
|
|
|
|
*/
|
2015-12-20 03:16:09 -07:00
|
|
|
IB_DEVICE_UD_IP_CSUM = (1 << 18),
|
|
|
|
IB_DEVICE_UD_TSO = (1 << 19),
|
|
|
|
IB_DEVICE_XRC = (1 << 20),
|
2015-12-23 11:12:45 -07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* This device supports the IB "base memory management extension",
|
|
|
|
* which includes support for fast registrations (IB_WR_REG_MR,
|
|
|
|
* IB_WR_LOCAL_INV and IB_WR_SEND_WITH_INV verbs). This flag should
|
|
|
|
* also be set by any iWarp device which must support FRs to comply
|
|
|
|
* to the iWarp verbs spec. iWarp devices also support the
|
|
|
|
* IB_WR_RDMA_READ_WITH_INV verb for RDMA READs that invalidate the
|
|
|
|
* stag.
|
|
|
|
*/
|
2015-12-20 03:16:09 -07:00
|
|
|
IB_DEVICE_MEM_MGT_EXTENSIONS = (1 << 21),
|
|
|
|
IB_DEVICE_BLOCK_MULTICAST_LOOPBACK = (1 << 22),
|
|
|
|
IB_DEVICE_MEM_WINDOW_TYPE_2A = (1 << 23),
|
|
|
|
IB_DEVICE_MEM_WINDOW_TYPE_2B = (1 << 24),
|
|
|
|
IB_DEVICE_RC_IP_CSUM = (1 << 25),
|
|
|
|
IB_DEVICE_RAW_IP_CSUM = (1 << 26),
|
2015-12-20 03:16:10 -07:00
|
|
|
/*
|
|
|
|
* Devices should set IB_DEVICE_CROSS_CHANNEL if they
|
|
|
|
* support execution of WQEs that involve synchronization
|
|
|
|
* of I/O operations with single completion queue managed
|
|
|
|
* by hardware.
|
|
|
|
*/
|
|
|
|
IB_DEVICE_CROSS_CHANNEL = (1 << 27),
|
2015-12-20 03:16:09 -07:00
|
|
|
IB_DEVICE_MANAGED_FLOW_STEERING = (1 << 29),
|
|
|
|
IB_DEVICE_SIGNATURE_HANDOVER = (1 << 30),
|
|
|
|
IB_DEVICE_ON_DEMAND_PAGING = (1 << 31),
|
2016-02-29 10:07:32 -07:00
|
|
|
IB_DEVICE_SG_GAPS_REG = (1ULL << 32),
|
2016-03-11 13:58:38 -07:00
|
|
|
IB_DEVICE_VIRTUAL_FUNCTION = ((u64)1 << 33),
|
2014-02-23 05:19:05 -07:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_signature_prot_cap {
|
|
|
|
IB_PROT_T10DIF_TYPE_1 = 1,
|
|
|
|
IB_PROT_T10DIF_TYPE_2 = 1 << 1,
|
|
|
|
IB_PROT_T10DIF_TYPE_3 = 1 << 2,
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_signature_guard_cap {
|
|
|
|
IB_GUARD_T10DIF_CRC = 1,
|
|
|
|
IB_GUARD_T10DIF_CSUM = 1 << 1,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_atomic_cap {
|
|
|
|
IB_ATOMIC_NONE,
|
|
|
|
IB_ATOMIC_HCA,
|
|
|
|
IB_ATOMIC_GLOB
|
|
|
|
};
|
|
|
|
|
2014-12-11 08:04:16 -07:00
|
|
|
enum ib_odp_general_cap_bits {
|
|
|
|
IB_ODP_SUPPORT = 1 << 0,
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_odp_transport_cap_bits {
|
|
|
|
IB_ODP_SUPPORT_SEND = 1 << 0,
|
|
|
|
IB_ODP_SUPPORT_RECV = 1 << 1,
|
|
|
|
IB_ODP_SUPPORT_WRITE = 1 << 2,
|
|
|
|
IB_ODP_SUPPORT_READ = 1 << 3,
|
|
|
|
IB_ODP_SUPPORT_ATOMIC = 1 << 4,
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_odp_caps {
|
|
|
|
uint64_t general_caps;
|
|
|
|
struct {
|
|
|
|
uint32_t rc_odp_caps;
|
|
|
|
uint32_t uc_odp_caps;
|
|
|
|
uint32_t ud_odp_caps;
|
|
|
|
} per_transport_caps;
|
|
|
|
};
|
|
|
|
|
2015-06-11 07:35:22 -06:00
|
|
|
enum ib_cq_creation_flags {
|
|
|
|
IB_CQ_FLAGS_TIMESTAMP_COMPLETION = 1 << 0,
|
2015-12-20 03:16:10 -07:00
|
|
|
IB_CQ_FLAGS_IGNORE_OVERRUN = 1 << 1,
|
2015-06-11 07:35:22 -06:00
|
|
|
};
|
|
|
|
|
2015-06-11 07:35:20 -06:00
|
|
|
struct ib_cq_init_attr {
|
|
|
|
unsigned int cqe;
|
|
|
|
int comp_vector;
|
|
|
|
u32 flags;
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_device_attr {
|
|
|
|
u64 fw_ver;
|
2005-08-13 22:05:57 -06:00
|
|
|
__be64 sys_image_guid;
|
2005-04-16 16:20:36 -06:00
|
|
|
u64 max_mr_size;
|
|
|
|
u64 page_size_cap;
|
|
|
|
u32 vendor_id;
|
|
|
|
u32 vendor_part_id;
|
|
|
|
u32 hw_ver;
|
|
|
|
int max_qp;
|
|
|
|
int max_qp_wr;
|
2016-02-23 01:25:25 -07:00
|
|
|
u64 device_cap_flags;
|
2005-04-16 16:20:36 -06:00
|
|
|
int max_sge;
|
|
|
|
int max_sge_rd;
|
|
|
|
int max_cq;
|
|
|
|
int max_cqe;
|
|
|
|
int max_mr;
|
|
|
|
int max_pd;
|
|
|
|
int max_qp_rd_atom;
|
|
|
|
int max_ee_rd_atom;
|
|
|
|
int max_res_rd_atom;
|
|
|
|
int max_qp_init_rd_atom;
|
|
|
|
int max_ee_init_rd_atom;
|
|
|
|
enum ib_atomic_cap atomic_cap;
|
IB/core: Add support for masked atomic operations
- Add new IB_WR_MASKED_ATOMIC_CMP_AND_SWP and IB_WR_MASKED_ATOMIC_FETCH_AND_ADD
send opcodes that can be used to post "masked atomic compare and
swap" and "masked atomic fetch and add" work request respectively.
- Add masked_atomic_cap capability.
- Add mask fields to atomic struct of ib_send_wr
- Add new opcodes to ib_wc_opcode
The new operations are described more precisely below:
* Masked Compare and Swap (MskCmpSwap)
The MskCmpSwap atomic operation is an extension to the CmpSwap
operation defined in the IB spec. MskCmpSwap allows the user to
select a portion of the 64 bit target data for the “compare” check as
well as to restrict the swap to a (possibly different) portion. The
pseudo code below describes the operation:
| atomic_response = *va
| if (!((compare_add ^ *va) & compare_add_mask)) then
| *va = (*va & ~(swap_mask)) | (swap & swap_mask)
|
| return atomic_response
The additional operands are carried in the Extended Transport Header.
Atomic response generation and packet format for MskCmpSwap is as for
standard IB Atomic operations.
* Masked Fetch and Add (MFetchAdd)
The MFetchAdd Atomic operation extends the functionality of the
standard IB FetchAdd by allowing the user to split the target into
multiple fields of selectable length. The atomic add is done
independently on each one of this fields. A bit set in the
field_boundary parameter specifies the field boundaries. The pseudo
code below describes the operation:
| bit_adder(ci, b1, b2, *co)
| {
| value = ci + b1 + b2
| *co = !!(value & 2)
|
| return value & 1
| }
|
| #define MASK_IS_SET(mask, attr) (!!((mask)&(attr)))
| bit_position = 1
| carry = 0
| atomic_response = 0
|
| for i = 0 to 63
| {
| if ( i != 0 )
| bit_position = bit_position << 1
|
| bit_add_res = bit_adder(carry, MASK_IS_SET(*va, bit_position),
| MASK_IS_SET(compare_add, bit_position), &new_carry)
| if (bit_add_res)
| atomic_response |= bit_position
|
| carry = ((new_carry) && (!MASK_IS_SET(compare_add_mask, bit_position)))
| }
|
| return atomic_response
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-14 08:23:01 -06:00
|
|
|
enum ib_atomic_cap masked_atomic_cap;
|
2005-04-16 16:20:36 -06:00
|
|
|
int max_ee;
|
|
|
|
int max_rdd;
|
|
|
|
int max_mw;
|
|
|
|
int max_raw_ipv6_qp;
|
|
|
|
int max_raw_ethy_qp;
|
|
|
|
int max_mcast_grp;
|
|
|
|
int max_mcast_qp_attach;
|
|
|
|
int max_total_mcast_qp_attach;
|
|
|
|
int max_ah;
|
|
|
|
int max_fmr;
|
|
|
|
int max_map_per_fmr;
|
|
|
|
int max_srq;
|
|
|
|
int max_srq_wr;
|
|
|
|
int max_srq_sge;
|
2008-07-15 00:48:45 -06:00
|
|
|
unsigned int max_fast_reg_page_list_len;
|
2005-04-16 16:20:36 -06:00
|
|
|
u16 max_pkeys;
|
|
|
|
u8 local_ca_ack_delay;
|
2014-02-23 05:19:05 -07:00
|
|
|
int sig_prot_cap;
|
|
|
|
int sig_guard_cap;
|
2014-12-11 08:04:16 -07:00
|
|
|
struct ib_odp_caps odp_caps;
|
2015-06-11 07:35:24 -06:00
|
|
|
uint64_t timestamp_mask;
|
|
|
|
uint64_t hca_core_clock; /* in KHZ */
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_mtu {
|
|
|
|
IB_MTU_256 = 1,
|
|
|
|
IB_MTU_512 = 2,
|
|
|
|
IB_MTU_1024 = 3,
|
|
|
|
IB_MTU_2048 = 4,
|
|
|
|
IB_MTU_4096 = 5
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline int ib_mtu_enum_to_int(enum ib_mtu mtu)
|
|
|
|
{
|
|
|
|
switch (mtu) {
|
|
|
|
case IB_MTU_256: return 256;
|
|
|
|
case IB_MTU_512: return 512;
|
|
|
|
case IB_MTU_1024: return 1024;
|
|
|
|
case IB_MTU_2048: return 2048;
|
|
|
|
case IB_MTU_4096: return 4096;
|
|
|
|
default: return -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
enum ib_port_state {
|
|
|
|
IB_PORT_NOP = 0,
|
|
|
|
IB_PORT_DOWN = 1,
|
|
|
|
IB_PORT_INIT = 2,
|
|
|
|
IB_PORT_ARMED = 3,
|
|
|
|
IB_PORT_ACTIVE = 4,
|
|
|
|
IB_PORT_ACTIVE_DEFER = 5
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_port_cap_flags {
|
|
|
|
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,
|
2011-10-05 05:21:47 -06:00
|
|
|
IB_PORT_EXTENDED_SPEEDS_SUP = 1 << 14,
|
2005-04-16 16:20:36 -06:00
|
|
|
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,
|
2014-02-09 02:54:34 -07:00
|
|
|
IB_PORT_CLIENT_REG_SUP = 1 << 25,
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
IB_PORT_IP_BASED_GIDS = 1 << 26,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_port_width {
|
|
|
|
IB_WIDTH_1X = 1,
|
|
|
|
IB_WIDTH_4X = 2,
|
|
|
|
IB_WIDTH_8X = 4,
|
|
|
|
IB_WIDTH_12X = 8
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline int ib_width_enum_to_int(enum ib_port_width width)
|
|
|
|
{
|
|
|
|
switch (width) {
|
|
|
|
case IB_WIDTH_1X: return 1;
|
|
|
|
case IB_WIDTH_4X: return 4;
|
|
|
|
case IB_WIDTH_8X: return 8;
|
|
|
|
case IB_WIDTH_12X: return 12;
|
|
|
|
default: return -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-02-28 09:49:50 -07:00
|
|
|
enum ib_port_speed {
|
|
|
|
IB_SPEED_SDR = 1,
|
|
|
|
IB_SPEED_DDR = 2,
|
|
|
|
IB_SPEED_QDR = 4,
|
|
|
|
IB_SPEED_FDR10 = 8,
|
|
|
|
IB_SPEED_FDR = 16,
|
|
|
|
IB_SPEED_EDR = 32
|
|
|
|
};
|
|
|
|
|
2008-07-15 00:48:48 -06:00
|
|
|
struct ib_protocol_stats {
|
|
|
|
/* TBD... */
|
|
|
|
};
|
|
|
|
|
|
|
|
struct iw_protocol_stats {
|
|
|
|
u64 ipInReceives;
|
|
|
|
u64 ipInHdrErrors;
|
|
|
|
u64 ipInTooBigErrors;
|
|
|
|
u64 ipInNoRoutes;
|
|
|
|
u64 ipInAddrErrors;
|
|
|
|
u64 ipInUnknownProtos;
|
|
|
|
u64 ipInTruncatedPkts;
|
|
|
|
u64 ipInDiscards;
|
|
|
|
u64 ipInDelivers;
|
|
|
|
u64 ipOutForwDatagrams;
|
|
|
|
u64 ipOutRequests;
|
|
|
|
u64 ipOutDiscards;
|
|
|
|
u64 ipOutNoRoutes;
|
|
|
|
u64 ipReasmTimeout;
|
|
|
|
u64 ipReasmReqds;
|
|
|
|
u64 ipReasmOKs;
|
|
|
|
u64 ipReasmFails;
|
|
|
|
u64 ipFragOKs;
|
|
|
|
u64 ipFragFails;
|
|
|
|
u64 ipFragCreates;
|
|
|
|
u64 ipInMcastPkts;
|
|
|
|
u64 ipOutMcastPkts;
|
|
|
|
u64 ipInBcastPkts;
|
|
|
|
u64 ipOutBcastPkts;
|
|
|
|
|
|
|
|
u64 tcpRtoAlgorithm;
|
|
|
|
u64 tcpRtoMin;
|
|
|
|
u64 tcpRtoMax;
|
|
|
|
u64 tcpMaxConn;
|
|
|
|
u64 tcpActiveOpens;
|
|
|
|
u64 tcpPassiveOpens;
|
|
|
|
u64 tcpAttemptFails;
|
|
|
|
u64 tcpEstabResets;
|
|
|
|
u64 tcpCurrEstab;
|
|
|
|
u64 tcpInSegs;
|
|
|
|
u64 tcpOutSegs;
|
|
|
|
u64 tcpRetransSegs;
|
|
|
|
u64 tcpInErrs;
|
|
|
|
u64 tcpOutRsts;
|
|
|
|
};
|
|
|
|
|
|
|
|
union rdma_protocol_stats {
|
|
|
|
struct ib_protocol_stats ib;
|
|
|
|
struct iw_protocol_stats iw;
|
|
|
|
};
|
|
|
|
|
2015-05-13 18:02:59 -06:00
|
|
|
/* Define bits for the various functionality this port needs to be supported by
|
|
|
|
* the core.
|
|
|
|
*/
|
|
|
|
/* Management 0x00000FFF */
|
|
|
|
#define RDMA_CORE_CAP_IB_MAD 0x00000001
|
|
|
|
#define RDMA_CORE_CAP_IB_SMI 0x00000002
|
|
|
|
#define RDMA_CORE_CAP_IB_CM 0x00000004
|
|
|
|
#define RDMA_CORE_CAP_IW_CM 0x00000008
|
|
|
|
#define RDMA_CORE_CAP_IB_SA 0x00000010
|
2015-06-06 12:38:32 -06:00
|
|
|
#define RDMA_CORE_CAP_OPA_MAD 0x00000020
|
2015-05-13 18:02:59 -06:00
|
|
|
|
|
|
|
/* Address format 0x000FF000 */
|
|
|
|
#define RDMA_CORE_CAP_AF_IB 0x00001000
|
|
|
|
#define RDMA_CORE_CAP_ETH_AH 0x00002000
|
|
|
|
|
|
|
|
/* Protocol 0xFFF00000 */
|
|
|
|
#define RDMA_CORE_CAP_PROT_IB 0x00100000
|
|
|
|
#define RDMA_CORE_CAP_PROT_ROCE 0x00200000
|
|
|
|
#define RDMA_CORE_CAP_PROT_IWARP 0x00400000
|
2015-12-23 05:56:50 -07:00
|
|
|
#define RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP 0x00800000
|
2015-05-13 18:02:59 -06:00
|
|
|
|
|
|
|
#define RDMA_CORE_PORT_IBA_IB (RDMA_CORE_CAP_PROT_IB \
|
|
|
|
| RDMA_CORE_CAP_IB_MAD \
|
|
|
|
| RDMA_CORE_CAP_IB_SMI \
|
|
|
|
| RDMA_CORE_CAP_IB_CM \
|
|
|
|
| RDMA_CORE_CAP_IB_SA \
|
|
|
|
| RDMA_CORE_CAP_AF_IB)
|
|
|
|
#define RDMA_CORE_PORT_IBA_ROCE (RDMA_CORE_CAP_PROT_ROCE \
|
|
|
|
| RDMA_CORE_CAP_IB_MAD \
|
|
|
|
| RDMA_CORE_CAP_IB_CM \
|
|
|
|
| RDMA_CORE_CAP_AF_IB \
|
|
|
|
| RDMA_CORE_CAP_ETH_AH)
|
2015-12-23 05:56:50 -07:00
|
|
|
#define RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP \
|
|
|
|
(RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP \
|
|
|
|
| RDMA_CORE_CAP_IB_MAD \
|
|
|
|
| RDMA_CORE_CAP_IB_CM \
|
|
|
|
| RDMA_CORE_CAP_AF_IB \
|
|
|
|
| RDMA_CORE_CAP_ETH_AH)
|
2015-05-13 18:02:59 -06:00
|
|
|
#define RDMA_CORE_PORT_IWARP (RDMA_CORE_CAP_PROT_IWARP \
|
|
|
|
| RDMA_CORE_CAP_IW_CM)
|
2015-06-06 12:38:32 -06:00
|
|
|
#define RDMA_CORE_PORT_INTEL_OPA (RDMA_CORE_PORT_IBA_IB \
|
|
|
|
| RDMA_CORE_CAP_OPA_MAD)
|
2015-05-13 18:02:59 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_port_attr {
|
2016-03-11 13:58:36 -07:00
|
|
|
u64 subnet_prefix;
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_port_state state;
|
|
|
|
enum ib_mtu max_mtu;
|
|
|
|
enum ib_mtu active_mtu;
|
|
|
|
int gid_tbl_len;
|
|
|
|
u32 port_cap_flags;
|
|
|
|
u32 max_msg_sz;
|
|
|
|
u32 bad_pkey_cntr;
|
|
|
|
u32 qkey_viol_cntr;
|
|
|
|
u16 pkey_tbl_len;
|
|
|
|
u16 lid;
|
|
|
|
u16 sm_lid;
|
|
|
|
u8 lmc;
|
|
|
|
u8 max_vl_num;
|
|
|
|
u8 sm_sl;
|
|
|
|
u8 subnet_timeout;
|
|
|
|
u8 init_type_reply;
|
|
|
|
u8 active_width;
|
|
|
|
u8 active_speed;
|
|
|
|
u8 phys_state;
|
2016-03-11 13:58:37 -07:00
|
|
|
bool grh_required;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_device_modify_flags {
|
2006-02-02 10:47:14 -07:00
|
|
|
IB_DEVICE_MODIFY_SYS_IMAGE_GUID = 1 << 0,
|
|
|
|
IB_DEVICE_MODIFY_NODE_DESC = 1 << 1
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_device_modify {
|
|
|
|
u64 sys_image_guid;
|
2006-02-02 10:47:14 -07:00
|
|
|
char node_desc[64];
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_port_modify_flags {
|
|
|
|
IB_PORT_SHUTDOWN = 1,
|
|
|
|
IB_PORT_INIT_TYPE = (1<<2),
|
|
|
|
IB_PORT_RESET_QKEY_CNTR = (1<<3)
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_port_modify {
|
|
|
|
u32 set_port_cap_mask;
|
|
|
|
u32 clr_port_cap_mask;
|
|
|
|
u8 init_type;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_event_type {
|
|
|
|
IB_EVENT_CQ_ERR,
|
|
|
|
IB_EVENT_QP_FATAL,
|
|
|
|
IB_EVENT_QP_REQ_ERR,
|
|
|
|
IB_EVENT_QP_ACCESS_ERR,
|
|
|
|
IB_EVENT_COMM_EST,
|
|
|
|
IB_EVENT_SQ_DRAINED,
|
|
|
|
IB_EVENT_PATH_MIG,
|
|
|
|
IB_EVENT_PATH_MIG_ERR,
|
|
|
|
IB_EVENT_DEVICE_FATAL,
|
|
|
|
IB_EVENT_PORT_ACTIVE,
|
|
|
|
IB_EVENT_PORT_ERR,
|
|
|
|
IB_EVENT_LID_CHANGE,
|
|
|
|
IB_EVENT_PKEY_CHANGE,
|
2005-08-18 13:23:08 -06:00
|
|
|
IB_EVENT_SM_CHANGE,
|
|
|
|
IB_EVENT_SRQ_ERR,
|
|
|
|
IB_EVENT_SRQ_LIMIT_REACHED,
|
2006-06-17 21:37:35 -06:00
|
|
|
IB_EVENT_QP_LAST_WQE_REACHED,
|
2011-06-15 08:39:29 -06:00
|
|
|
IB_EVENT_CLIENT_REREGISTER,
|
|
|
|
IB_EVENT_GID_CHANGE,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2015-08-03 11:01:52 -06:00
|
|
|
const char *__attribute_const__ ib_event_msg(enum ib_event_type event);
|
2015-05-18 04:40:28 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_event {
|
|
|
|
struct ib_device *device;
|
|
|
|
union {
|
|
|
|
struct ib_cq *cq;
|
|
|
|
struct ib_qp *qp;
|
2005-08-18 13:23:08 -06:00
|
|
|
struct ib_srq *srq;
|
2005-04-16 16:20:36 -06:00
|
|
|
u8 port_num;
|
|
|
|
} element;
|
|
|
|
enum ib_event_type event;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_event_handler {
|
|
|
|
struct ib_device *device;
|
|
|
|
void (*handler)(struct ib_event_handler *, struct ib_event *);
|
|
|
|
struct list_head list;
|
|
|
|
};
|
|
|
|
|
|
|
|
#define INIT_IB_EVENT_HANDLER(_ptr, _device, _handler) \
|
|
|
|
do { \
|
|
|
|
(_ptr)->device = _device; \
|
|
|
|
(_ptr)->handler = _handler; \
|
|
|
|
INIT_LIST_HEAD(&(_ptr)->list); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
struct ib_global_route {
|
|
|
|
union ib_gid dgid;
|
|
|
|
u32 flow_label;
|
|
|
|
u8 sgid_index;
|
|
|
|
u8 hop_limit;
|
|
|
|
u8 traffic_class;
|
|
|
|
};
|
|
|
|
|
2005-07-27 12:45:34 -06:00
|
|
|
struct ib_grh {
|
2005-08-13 22:05:57 -06:00
|
|
|
__be32 version_tclass_flow;
|
|
|
|
__be16 paylen;
|
2005-07-27 12:45:34 -06:00
|
|
|
u8 next_hdr;
|
|
|
|
u8 hop_limit;
|
|
|
|
union ib_gid sgid;
|
|
|
|
union ib_gid dgid;
|
|
|
|
};
|
|
|
|
|
2015-12-23 05:56:51 -07:00
|
|
|
union rdma_network_hdr {
|
|
|
|
struct ib_grh ibgrh;
|
|
|
|
struct {
|
|
|
|
/* The IB spec states that if it's IPv4, the header
|
|
|
|
* is located in the last 20 bytes of the header.
|
|
|
|
*/
|
|
|
|
u8 reserved[20];
|
|
|
|
struct iphdr roce4grh;
|
|
|
|
};
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
enum {
|
|
|
|
IB_MULTICAST_QPN = 0xffffff
|
|
|
|
};
|
|
|
|
|
2009-02-14 23:58:35 -07:00
|
|
|
#define IB_LID_PERMISSIVE cpu_to_be16(0xFFFF)
|
2016-01-06 11:04:31 -07:00
|
|
|
#define IB_MULTICAST_LID_BASE cpu_to_be16(0xC000)
|
2005-08-13 22:05:57 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_ah_flags {
|
|
|
|
IB_AH_GRH = 1
|
|
|
|
};
|
|
|
|
|
2006-04-10 10:43:47 -06:00
|
|
|
enum ib_rate {
|
|
|
|
IB_RATE_PORT_CURRENT = 0,
|
|
|
|
IB_RATE_2_5_GBPS = 2,
|
|
|
|
IB_RATE_5_GBPS = 5,
|
|
|
|
IB_RATE_10_GBPS = 3,
|
|
|
|
IB_RATE_20_GBPS = 6,
|
|
|
|
IB_RATE_30_GBPS = 4,
|
|
|
|
IB_RATE_40_GBPS = 7,
|
|
|
|
IB_RATE_60_GBPS = 8,
|
|
|
|
IB_RATE_80_GBPS = 9,
|
2011-10-05 05:21:47 -06:00
|
|
|
IB_RATE_120_GBPS = 10,
|
|
|
|
IB_RATE_14_GBPS = 11,
|
|
|
|
IB_RATE_56_GBPS = 12,
|
|
|
|
IB_RATE_112_GBPS = 13,
|
|
|
|
IB_RATE_168_GBPS = 14,
|
|
|
|
IB_RATE_25_GBPS = 15,
|
|
|
|
IB_RATE_100_GBPS = 16,
|
|
|
|
IB_RATE_200_GBPS = 17,
|
|
|
|
IB_RATE_300_GBPS = 18
|
2006-04-10 10:43:47 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_rate_to_mult - Convert the IB rate enum to a multiple of the
|
|
|
|
* base rate of 2.5 Gbit/sec. For example, IB_RATE_5_GBPS will be
|
|
|
|
* converted to 2, since 5 Gbit/sec is 2 * 2.5 Gbit/sec.
|
|
|
|
* @rate: rate to convert.
|
|
|
|
*/
|
2014-06-04 11:00:16 -06:00
|
|
|
__attribute_const__ int ib_rate_to_mult(enum ib_rate rate);
|
2006-04-10 10:43:47 -06:00
|
|
|
|
2011-10-05 05:21:47 -06:00
|
|
|
/**
|
|
|
|
* ib_rate_to_mbps - Convert the IB rate enum to Mbps.
|
|
|
|
* For example, IB_RATE_2_5_GBPS will be converted to 2500.
|
|
|
|
* @rate: rate to convert.
|
|
|
|
*/
|
2014-06-04 11:00:16 -06:00
|
|
|
__attribute_const__ int ib_rate_to_mbps(enum ib_rate rate);
|
2011-10-05 05:21:47 -06:00
|
|
|
|
2014-02-23 05:19:04 -07:00
|
|
|
|
|
|
|
/**
|
2015-07-30 01:32:35 -06:00
|
|
|
* enum ib_mr_type - memory region type
|
|
|
|
* @IB_MR_TYPE_MEM_REG: memory region that is used for
|
|
|
|
* normal registration
|
|
|
|
* @IB_MR_TYPE_SIGNATURE: memory region that is used for
|
|
|
|
* signature operations (data-integrity
|
|
|
|
* capable regions)
|
2016-02-29 10:07:32 -07:00
|
|
|
* @IB_MR_TYPE_SG_GAPS: memory region that is capable to
|
|
|
|
* register any arbitrary sg lists (without
|
|
|
|
* the normal mr constraints - see
|
|
|
|
* ib_map_mr_sg)
|
2014-02-23 05:19:04 -07:00
|
|
|
*/
|
2015-07-30 01:32:35 -06:00
|
|
|
enum ib_mr_type {
|
|
|
|
IB_MR_TYPE_MEM_REG,
|
|
|
|
IB_MR_TYPE_SIGNATURE,
|
2016-02-29 10:07:32 -07:00
|
|
|
IB_MR_TYPE_SG_GAPS,
|
2014-02-23 05:19:04 -07:00
|
|
|
};
|
|
|
|
|
2014-02-23 05:19:05 -07:00
|
|
|
/**
|
2014-08-13 10:54:35 -06:00
|
|
|
* Signature types
|
|
|
|
* IB_SIG_TYPE_NONE: Unprotected.
|
|
|
|
* IB_SIG_TYPE_T10_DIF: Type T10-DIF
|
2014-02-23 05:19:05 -07:00
|
|
|
*/
|
2014-08-13 10:54:35 -06:00
|
|
|
enum ib_signature_type {
|
|
|
|
IB_SIG_TYPE_NONE,
|
|
|
|
IB_SIG_TYPE_T10_DIF,
|
2014-02-23 05:19:05 -07:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Signature T10-DIF block-guard types
|
|
|
|
* IB_T10DIF_CRC: Corresponds to T10-PI mandated CRC checksum rules.
|
|
|
|
* IB_T10DIF_CSUM: Corresponds to IP checksum rules.
|
|
|
|
*/
|
|
|
|
enum ib_t10_dif_bg_type {
|
|
|
|
IB_T10DIF_CRC,
|
|
|
|
IB_T10DIF_CSUM
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ib_t10_dif_domain - Parameters specific for T10-DIF
|
|
|
|
* domain.
|
|
|
|
* @bg_type: T10-DIF block guard type (CRC|CSUM)
|
|
|
|
* @pi_interval: protection information interval.
|
|
|
|
* @bg: seed of guard computation.
|
|
|
|
* @app_tag: application tag of guard block
|
|
|
|
* @ref_tag: initial guard block reference tag.
|
2014-08-13 10:54:35 -06:00
|
|
|
* @ref_remap: Indicate wethear the reftag increments each block
|
|
|
|
* @app_escape: Indicate to skip block check if apptag=0xffff
|
|
|
|
* @ref_escape: Indicate to skip block check if reftag=0xffffffff
|
|
|
|
* @apptag_check_mask: check bitmask of application tag.
|
2014-02-23 05:19:05 -07:00
|
|
|
*/
|
|
|
|
struct ib_t10_dif_domain {
|
|
|
|
enum ib_t10_dif_bg_type bg_type;
|
|
|
|
u16 pi_interval;
|
|
|
|
u16 bg;
|
|
|
|
u16 app_tag;
|
|
|
|
u32 ref_tag;
|
2014-08-13 10:54:35 -06:00
|
|
|
bool ref_remap;
|
|
|
|
bool app_escape;
|
|
|
|
bool ref_escape;
|
|
|
|
u16 apptag_check_mask;
|
2014-02-23 05:19:05 -07:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ib_sig_domain - Parameters for signature domain
|
|
|
|
* @sig_type: specific signauture type
|
|
|
|
* @sig: union of all signature domain attributes that may
|
|
|
|
* be used to set domain layout.
|
|
|
|
*/
|
|
|
|
struct ib_sig_domain {
|
|
|
|
enum ib_signature_type sig_type;
|
|
|
|
union {
|
|
|
|
struct ib_t10_dif_domain dif;
|
|
|
|
} sig;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ib_sig_attrs - Parameters for signature handover operation
|
|
|
|
* @check_mask: bitmask for signature byte check (8 bytes)
|
|
|
|
* @mem: memory domain layout desciptor.
|
|
|
|
* @wire: wire domain layout desciptor.
|
|
|
|
*/
|
|
|
|
struct ib_sig_attrs {
|
|
|
|
u8 check_mask;
|
|
|
|
struct ib_sig_domain mem;
|
|
|
|
struct ib_sig_domain wire;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_sig_err_type {
|
|
|
|
IB_SIG_BAD_GUARD,
|
|
|
|
IB_SIG_BAD_REFTAG,
|
|
|
|
IB_SIG_BAD_APPTAG,
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ib_sig_err - signature error descriptor
|
|
|
|
*/
|
|
|
|
struct ib_sig_err {
|
|
|
|
enum ib_sig_err_type err_type;
|
|
|
|
u32 expected;
|
|
|
|
u32 actual;
|
|
|
|
u64 sig_err_offset;
|
|
|
|
u32 key;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_mr_status_check {
|
|
|
|
IB_MR_CHECK_SIG_STATUS = 1,
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ib_mr_status - Memory region status container
|
|
|
|
*
|
|
|
|
* @fail_status: Bitmask of MR checks status. For each
|
|
|
|
* failed check a corresponding status bit is set.
|
|
|
|
* @sig_err: Additional info for IB_MR_CEHCK_SIG_STATUS
|
|
|
|
* failure.
|
|
|
|
*/
|
|
|
|
struct ib_mr_status {
|
|
|
|
u32 fail_status;
|
|
|
|
struct ib_sig_err sig_err;
|
|
|
|
};
|
|
|
|
|
2006-04-10 10:43:47 -06:00
|
|
|
/**
|
|
|
|
* mult_to_ib_rate - Convert a multiple of 2.5 Gbit/sec to an IB rate
|
|
|
|
* enum.
|
|
|
|
* @mult: multiple to convert.
|
|
|
|
*/
|
2014-06-04 11:00:16 -06:00
|
|
|
__attribute_const__ enum ib_rate mult_to_ib_rate(int mult);
|
2006-04-10 10:43:47 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_ah_attr {
|
|
|
|
struct ib_global_route grh;
|
|
|
|
u16 dlid;
|
|
|
|
u8 sl;
|
|
|
|
u8 src_path_bits;
|
|
|
|
u8 static_rate;
|
|
|
|
u8 ah_flags;
|
|
|
|
u8 port_num;
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
u8 dmac[ETH_ALEN];
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_wc_status {
|
|
|
|
IB_WC_SUCCESS,
|
|
|
|
IB_WC_LOC_LEN_ERR,
|
|
|
|
IB_WC_LOC_QP_OP_ERR,
|
|
|
|
IB_WC_LOC_EEC_OP_ERR,
|
|
|
|
IB_WC_LOC_PROT_ERR,
|
|
|
|
IB_WC_WR_FLUSH_ERR,
|
|
|
|
IB_WC_MW_BIND_ERR,
|
|
|
|
IB_WC_BAD_RESP_ERR,
|
|
|
|
IB_WC_LOC_ACCESS_ERR,
|
|
|
|
IB_WC_REM_INV_REQ_ERR,
|
|
|
|
IB_WC_REM_ACCESS_ERR,
|
|
|
|
IB_WC_REM_OP_ERR,
|
|
|
|
IB_WC_RETRY_EXC_ERR,
|
|
|
|
IB_WC_RNR_RETRY_EXC_ERR,
|
|
|
|
IB_WC_LOC_RDD_VIOL_ERR,
|
|
|
|
IB_WC_REM_INV_RD_REQ_ERR,
|
|
|
|
IB_WC_REM_ABORT_ERR,
|
|
|
|
IB_WC_INV_EECN_ERR,
|
|
|
|
IB_WC_INV_EEC_STATE_ERR,
|
|
|
|
IB_WC_FATAL_ERR,
|
|
|
|
IB_WC_RESP_TIMEOUT_ERR,
|
|
|
|
IB_WC_GENERAL_ERR
|
|
|
|
};
|
|
|
|
|
2015-08-03 11:01:52 -06:00
|
|
|
const char *__attribute_const__ ib_wc_status_msg(enum ib_wc_status status);
|
2015-05-18 04:40:28 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_wc_opcode {
|
|
|
|
IB_WC_SEND,
|
|
|
|
IB_WC_RDMA_WRITE,
|
|
|
|
IB_WC_RDMA_READ,
|
|
|
|
IB_WC_COMP_SWAP,
|
|
|
|
IB_WC_FETCH_ADD,
|
2008-04-16 22:09:27 -06:00
|
|
|
IB_WC_LSO,
|
2008-07-15 00:48:45 -06:00
|
|
|
IB_WC_LOCAL_INV,
|
2015-10-13 10:11:24 -06:00
|
|
|
IB_WC_REG_MR,
|
IB/core: Add support for masked atomic operations
- Add new IB_WR_MASKED_ATOMIC_CMP_AND_SWP and IB_WR_MASKED_ATOMIC_FETCH_AND_ADD
send opcodes that can be used to post "masked atomic compare and
swap" and "masked atomic fetch and add" work request respectively.
- Add masked_atomic_cap capability.
- Add mask fields to atomic struct of ib_send_wr
- Add new opcodes to ib_wc_opcode
The new operations are described more precisely below:
* Masked Compare and Swap (MskCmpSwap)
The MskCmpSwap atomic operation is an extension to the CmpSwap
operation defined in the IB spec. MskCmpSwap allows the user to
select a portion of the 64 bit target data for the “compare” check as
well as to restrict the swap to a (possibly different) portion. The
pseudo code below describes the operation:
| atomic_response = *va
| if (!((compare_add ^ *va) & compare_add_mask)) then
| *va = (*va & ~(swap_mask)) | (swap & swap_mask)
|
| return atomic_response
The additional operands are carried in the Extended Transport Header.
Atomic response generation and packet format for MskCmpSwap is as for
standard IB Atomic operations.
* Masked Fetch and Add (MFetchAdd)
The MFetchAdd Atomic operation extends the functionality of the
standard IB FetchAdd by allowing the user to split the target into
multiple fields of selectable length. The atomic add is done
independently on each one of this fields. A bit set in the
field_boundary parameter specifies the field boundaries. The pseudo
code below describes the operation:
| bit_adder(ci, b1, b2, *co)
| {
| value = ci + b1 + b2
| *co = !!(value & 2)
|
| return value & 1
| }
|
| #define MASK_IS_SET(mask, attr) (!!((mask)&(attr)))
| bit_position = 1
| carry = 0
| atomic_response = 0
|
| for i = 0 to 63
| {
| if ( i != 0 )
| bit_position = bit_position << 1
|
| bit_add_res = bit_adder(carry, MASK_IS_SET(*va, bit_position),
| MASK_IS_SET(compare_add, bit_position), &new_carry)
| if (bit_add_res)
| atomic_response |= bit_position
|
| carry = ((new_carry) && (!MASK_IS_SET(compare_add_mask, bit_position)))
| }
|
| return atomic_response
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-14 08:23:01 -06:00
|
|
|
IB_WC_MASKED_COMP_SWAP,
|
|
|
|
IB_WC_MASKED_FETCH_ADD,
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Set value of IB_WC_RECV so consumers can test if a completion is a
|
|
|
|
* receive by testing (opcode & IB_WC_RECV).
|
|
|
|
*/
|
|
|
|
IB_WC_RECV = 1 << 7,
|
|
|
|
IB_WC_RECV_RDMA_WITH_IMM
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_wc_flags {
|
|
|
|
IB_WC_GRH = 1,
|
2008-07-15 00:48:45 -06:00
|
|
|
IB_WC_WITH_IMM = (1<<1),
|
|
|
|
IB_WC_WITH_INVALIDATE = (1<<2),
|
2012-01-11 10:03:51 -07:00
|
|
|
IB_WC_IP_CSUM_OK = (1<<3),
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
IB_WC_WITH_SMAC = (1<<4),
|
|
|
|
IB_WC_WITH_VLAN = (1<<5),
|
2015-12-23 05:56:51 -07:00
|
|
|
IB_WC_WITH_NETWORK_HDR_TYPE = (1<<6),
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_wc {
|
2015-12-11 12:53:03 -07:00
|
|
|
union {
|
|
|
|
u64 wr_id;
|
|
|
|
struct ib_cqe *wr_cqe;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_wc_status status;
|
|
|
|
enum ib_wc_opcode opcode;
|
|
|
|
u32 vendor_err;
|
|
|
|
u32 byte_len;
|
2006-12-31 12:09:42 -07:00
|
|
|
struct ib_qp *qp;
|
2008-07-15 00:48:45 -06:00
|
|
|
union {
|
|
|
|
__be32 imm_data;
|
|
|
|
u32 invalidate_rkey;
|
|
|
|
} ex;
|
2005-04-16 16:20:36 -06:00
|
|
|
u32 src_qp;
|
|
|
|
int wc_flags;
|
|
|
|
u16 pkey_index;
|
|
|
|
u16 slid;
|
|
|
|
u8 sl;
|
|
|
|
u8 dlid_path_bits;
|
|
|
|
u8 port_num; /* valid only for DR SMPs on switches */
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
u8 smac[ETH_ALEN];
|
|
|
|
u16 vlan_id;
|
2015-12-23 05:56:51 -07:00
|
|
|
u8 network_hdr_type;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
IB: Return "maybe missed event" hint from ib_req_notify_cq()
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-06 22:02:48 -06:00
|
|
|
enum ib_cq_notify_flags {
|
|
|
|
IB_CQ_SOLICITED = 1 << 0,
|
|
|
|
IB_CQ_NEXT_COMP = 1 << 1,
|
|
|
|
IB_CQ_SOLICITED_MASK = IB_CQ_SOLICITED | IB_CQ_NEXT_COMP,
|
|
|
|
IB_CQ_REPORT_MISSED_EVENTS = 1 << 2,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2011-05-23 17:31:36 -06:00
|
|
|
enum ib_srq_type {
|
2011-05-23 20:42:29 -06:00
|
|
|
IB_SRQT_BASIC,
|
|
|
|
IB_SRQT_XRC
|
2011-05-23 17:31:36 -06:00
|
|
|
};
|
|
|
|
|
2005-08-18 13:23:08 -06:00
|
|
|
enum ib_srq_attr_mask {
|
|
|
|
IB_SRQ_MAX_WR = 1 << 0,
|
|
|
|
IB_SRQ_LIMIT = 1 << 1,
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_srq_attr {
|
|
|
|
u32 max_wr;
|
|
|
|
u32 max_sge;
|
|
|
|
u32 srq_limit;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_srq_init_attr {
|
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
|
|
|
void *srq_context;
|
|
|
|
struct ib_srq_attr attr;
|
2011-05-23 17:31:36 -06:00
|
|
|
enum ib_srq_type srq_type;
|
2011-05-23 20:42:29 -06:00
|
|
|
|
|
|
|
union {
|
|
|
|
struct {
|
|
|
|
struct ib_xrcd *xrcd;
|
|
|
|
struct ib_cq *cq;
|
|
|
|
} xrc;
|
|
|
|
} ext;
|
2005-08-18 13:23:08 -06:00
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_qp_cap {
|
|
|
|
u32 max_send_wr;
|
|
|
|
u32 max_recv_wr;
|
|
|
|
u32 max_send_sge;
|
|
|
|
u32 max_recv_sge;
|
|
|
|
u32 max_inline_data;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_sig_type {
|
|
|
|
IB_SIGNAL_ALL_WR,
|
|
|
|
IB_SIGNAL_REQ_WR
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_qp_type {
|
|
|
|
/*
|
|
|
|
* IB_QPT_SMI and IB_QPT_GSI have to be the first two entries
|
|
|
|
* here (and in that order) since the MAD layer uses them as
|
|
|
|
* indices into a 2-entry table.
|
|
|
|
*/
|
|
|
|
IB_QPT_SMI,
|
|
|
|
IB_QPT_GSI,
|
|
|
|
|
|
|
|
IB_QPT_RC,
|
|
|
|
IB_QPT_UC,
|
|
|
|
IB_QPT_UD,
|
|
|
|
IB_QPT_RAW_IPV6,
|
2011-05-23 20:59:25 -06:00
|
|
|
IB_QPT_RAW_ETHERTYPE,
|
2012-03-01 03:17:51 -07:00
|
|
|
IB_QPT_RAW_PACKET = 8,
|
2011-05-23 20:59:25 -06:00
|
|
|
IB_QPT_XRC_INI = 9,
|
|
|
|
IB_QPT_XRC_TGT,
|
2013-07-07 08:25:52 -06:00
|
|
|
IB_QPT_MAX,
|
|
|
|
/* Reserve a range for qp types internal to the low level driver.
|
|
|
|
* These qp types will not be visible at the IB core layer, so the
|
|
|
|
* IB_QPT_MAX usages should not be affected in the core layer
|
|
|
|
*/
|
|
|
|
IB_QPT_RESERVED1 = 0x1000,
|
|
|
|
IB_QPT_RESERVED2,
|
|
|
|
IB_QPT_RESERVED3,
|
|
|
|
IB_QPT_RESERVED4,
|
|
|
|
IB_QPT_RESERVED5,
|
|
|
|
IB_QPT_RESERVED6,
|
|
|
|
IB_QPT_RESERVED7,
|
|
|
|
IB_QPT_RESERVED8,
|
|
|
|
IB_QPT_RESERVED9,
|
|
|
|
IB_QPT_RESERVED10,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2008-04-16 22:09:27 -06:00
|
|
|
enum ib_qp_create_flags {
|
2008-07-15 00:48:48 -06:00
|
|
|
IB_QP_CREATE_IPOIB_UD_LSO = 1 << 0,
|
|
|
|
IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK = 1 << 1,
|
2015-12-20 03:16:10 -07:00
|
|
|
IB_QP_CREATE_CROSS_CHANNEL = 1 << 2,
|
|
|
|
IB_QP_CREATE_MANAGED_SEND = 1 << 3,
|
|
|
|
IB_QP_CREATE_MANAGED_RECV = 1 << 4,
|
2013-11-07 06:25:12 -07:00
|
|
|
IB_QP_CREATE_NETIF_QP = 1 << 5,
|
2014-02-23 05:19:05 -07:00
|
|
|
IB_QP_CREATE_SIGNATURE_EN = 1 << 6,
|
2014-05-11 06:15:11 -06:00
|
|
|
IB_QP_CREATE_USE_GFP_NOIO = 1 << 7,
|
2012-08-03 02:40:37 -06:00
|
|
|
/* reserve bits 26-31 for low level drivers' internal use */
|
|
|
|
IB_QP_CREATE_RESERVED_START = 1 << 26,
|
|
|
|
IB_QP_CREATE_RESERVED_END = 1 << 31,
|
2008-04-16 22:09:27 -06:00
|
|
|
};
|
|
|
|
|
2013-08-01 09:49:53 -06:00
|
|
|
/*
|
|
|
|
* Note: users may not call ib_close_qp or ib_destroy_qp from the event_handler
|
|
|
|
* callback to destroy the passed in QP.
|
|
|
|
*/
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_qp_init_attr {
|
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
|
|
|
void *qp_context;
|
|
|
|
struct ib_cq *send_cq;
|
|
|
|
struct ib_cq *recv_cq;
|
|
|
|
struct ib_srq *srq;
|
2011-05-23 20:59:25 -06:00
|
|
|
struct ib_xrcd *xrcd; /* XRC TGT QPs only */
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_qp_cap cap;
|
|
|
|
enum ib_sig_type sq_sig_type;
|
|
|
|
enum ib_qp_type qp_type;
|
2008-04-16 22:09:27 -06:00
|
|
|
enum ib_qp_create_flags create_flags;
|
2005-04-16 16:20:36 -06:00
|
|
|
u8 port_num; /* special QP types only */
|
|
|
|
};
|
|
|
|
|
2011-08-08 16:31:51 -06:00
|
|
|
struct ib_qp_open_attr {
|
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
|
|
|
void *qp_context;
|
|
|
|
u32 qp_num;
|
|
|
|
enum ib_qp_type qp_type;
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_rnr_timeout {
|
|
|
|
IB_RNR_TIMER_655_36 = 0,
|
|
|
|
IB_RNR_TIMER_000_01 = 1,
|
|
|
|
IB_RNR_TIMER_000_02 = 2,
|
|
|
|
IB_RNR_TIMER_000_03 = 3,
|
|
|
|
IB_RNR_TIMER_000_04 = 4,
|
|
|
|
IB_RNR_TIMER_000_06 = 5,
|
|
|
|
IB_RNR_TIMER_000_08 = 6,
|
|
|
|
IB_RNR_TIMER_000_12 = 7,
|
|
|
|
IB_RNR_TIMER_000_16 = 8,
|
|
|
|
IB_RNR_TIMER_000_24 = 9,
|
|
|
|
IB_RNR_TIMER_000_32 = 10,
|
|
|
|
IB_RNR_TIMER_000_48 = 11,
|
|
|
|
IB_RNR_TIMER_000_64 = 12,
|
|
|
|
IB_RNR_TIMER_000_96 = 13,
|
|
|
|
IB_RNR_TIMER_001_28 = 14,
|
|
|
|
IB_RNR_TIMER_001_92 = 15,
|
|
|
|
IB_RNR_TIMER_002_56 = 16,
|
|
|
|
IB_RNR_TIMER_003_84 = 17,
|
|
|
|
IB_RNR_TIMER_005_12 = 18,
|
|
|
|
IB_RNR_TIMER_007_68 = 19,
|
|
|
|
IB_RNR_TIMER_010_24 = 20,
|
|
|
|
IB_RNR_TIMER_015_36 = 21,
|
|
|
|
IB_RNR_TIMER_020_48 = 22,
|
|
|
|
IB_RNR_TIMER_030_72 = 23,
|
|
|
|
IB_RNR_TIMER_040_96 = 24,
|
|
|
|
IB_RNR_TIMER_061_44 = 25,
|
|
|
|
IB_RNR_TIMER_081_92 = 26,
|
|
|
|
IB_RNR_TIMER_122_88 = 27,
|
|
|
|
IB_RNR_TIMER_163_84 = 28,
|
|
|
|
IB_RNR_TIMER_245_76 = 29,
|
|
|
|
IB_RNR_TIMER_327_68 = 30,
|
|
|
|
IB_RNR_TIMER_491_52 = 31
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_qp_attr_mask {
|
|
|
|
IB_QP_STATE = 1,
|
|
|
|
IB_QP_CUR_STATE = (1<<1),
|
|
|
|
IB_QP_EN_SQD_ASYNC_NOTIFY = (1<<2),
|
|
|
|
IB_QP_ACCESS_FLAGS = (1<<3),
|
|
|
|
IB_QP_PKEY_INDEX = (1<<4),
|
|
|
|
IB_QP_PORT = (1<<5),
|
|
|
|
IB_QP_QKEY = (1<<6),
|
|
|
|
IB_QP_AV = (1<<7),
|
|
|
|
IB_QP_PATH_MTU = (1<<8),
|
|
|
|
IB_QP_TIMEOUT = (1<<9),
|
|
|
|
IB_QP_RETRY_CNT = (1<<10),
|
|
|
|
IB_QP_RNR_RETRY = (1<<11),
|
|
|
|
IB_QP_RQ_PSN = (1<<12),
|
|
|
|
IB_QP_MAX_QP_RD_ATOMIC = (1<<13),
|
|
|
|
IB_QP_ALT_PATH = (1<<14),
|
|
|
|
IB_QP_MIN_RNR_TIMER = (1<<15),
|
|
|
|
IB_QP_SQ_PSN = (1<<16),
|
|
|
|
IB_QP_MAX_DEST_RD_ATOMIC = (1<<17),
|
|
|
|
IB_QP_PATH_MIG_STATE = (1<<18),
|
|
|
|
IB_QP_CAP = (1<<19),
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
IB_QP_DEST_QPN = (1<<20),
|
2015-10-15 09:38:53 -06:00
|
|
|
IB_QP_RESERVED1 = (1<<21),
|
|
|
|
IB_QP_RESERVED2 = (1<<22),
|
|
|
|
IB_QP_RESERVED3 = (1<<23),
|
|
|
|
IB_QP_RESERVED4 = (1<<24),
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_qp_state {
|
|
|
|
IB_QPS_RESET,
|
|
|
|
IB_QPS_INIT,
|
|
|
|
IB_QPS_RTR,
|
|
|
|
IB_QPS_RTS,
|
|
|
|
IB_QPS_SQD,
|
|
|
|
IB_QPS_SQE,
|
|
|
|
IB_QPS_ERR
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_mig_state {
|
|
|
|
IB_MIG_MIGRATED,
|
|
|
|
IB_MIG_REARM,
|
|
|
|
IB_MIG_ARMED
|
|
|
|
};
|
|
|
|
|
2013-02-06 09:19:12 -07:00
|
|
|
enum ib_mw_type {
|
|
|
|
IB_MW_TYPE_1 = 1,
|
|
|
|
IB_MW_TYPE_2 = 2
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_qp_attr {
|
|
|
|
enum ib_qp_state qp_state;
|
|
|
|
enum ib_qp_state cur_qp_state;
|
|
|
|
enum ib_mtu path_mtu;
|
|
|
|
enum ib_mig_state path_mig_state;
|
|
|
|
u32 qkey;
|
|
|
|
u32 rq_psn;
|
|
|
|
u32 sq_psn;
|
|
|
|
u32 dest_qp_num;
|
|
|
|
int qp_access_flags;
|
|
|
|
struct ib_qp_cap cap;
|
|
|
|
struct ib_ah_attr ah_attr;
|
|
|
|
struct ib_ah_attr alt_ah_attr;
|
|
|
|
u16 pkey_index;
|
|
|
|
u16 alt_pkey_index;
|
|
|
|
u8 en_sqd_async_notify;
|
|
|
|
u8 sq_draining;
|
|
|
|
u8 max_rd_atomic;
|
|
|
|
u8 max_dest_rd_atomic;
|
|
|
|
u8 min_rnr_timer;
|
|
|
|
u8 port_num;
|
|
|
|
u8 timeout;
|
|
|
|
u8 retry_cnt;
|
|
|
|
u8 rnr_retry;
|
|
|
|
u8 alt_port_num;
|
|
|
|
u8 alt_timeout;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_wr_opcode {
|
|
|
|
IB_WR_RDMA_WRITE,
|
|
|
|
IB_WR_RDMA_WRITE_WITH_IMM,
|
|
|
|
IB_WR_SEND,
|
|
|
|
IB_WR_SEND_WITH_IMM,
|
|
|
|
IB_WR_RDMA_READ,
|
|
|
|
IB_WR_ATOMIC_CMP_AND_SWP,
|
2008-04-16 22:09:27 -06:00
|
|
|
IB_WR_ATOMIC_FETCH_AND_ADD,
|
2008-04-16 22:09:32 -06:00
|
|
|
IB_WR_LSO,
|
|
|
|
IB_WR_SEND_WITH_INV,
|
2008-07-15 00:48:45 -06:00
|
|
|
IB_WR_RDMA_READ_WITH_INV,
|
|
|
|
IB_WR_LOCAL_INV,
|
2015-10-13 10:11:24 -06:00
|
|
|
IB_WR_REG_MR,
|
IB/core: Add support for masked atomic operations
- Add new IB_WR_MASKED_ATOMIC_CMP_AND_SWP and IB_WR_MASKED_ATOMIC_FETCH_AND_ADD
send opcodes that can be used to post "masked atomic compare and
swap" and "masked atomic fetch and add" work request respectively.
- Add masked_atomic_cap capability.
- Add mask fields to atomic struct of ib_send_wr
- Add new opcodes to ib_wc_opcode
The new operations are described more precisely below:
* Masked Compare and Swap (MskCmpSwap)
The MskCmpSwap atomic operation is an extension to the CmpSwap
operation defined in the IB spec. MskCmpSwap allows the user to
select a portion of the 64 bit target data for the “compare” check as
well as to restrict the swap to a (possibly different) portion. The
pseudo code below describes the operation:
| atomic_response = *va
| if (!((compare_add ^ *va) & compare_add_mask)) then
| *va = (*va & ~(swap_mask)) | (swap & swap_mask)
|
| return atomic_response
The additional operands are carried in the Extended Transport Header.
Atomic response generation and packet format for MskCmpSwap is as for
standard IB Atomic operations.
* Masked Fetch and Add (MFetchAdd)
The MFetchAdd Atomic operation extends the functionality of the
standard IB FetchAdd by allowing the user to split the target into
multiple fields of selectable length. The atomic add is done
independently on each one of this fields. A bit set in the
field_boundary parameter specifies the field boundaries. The pseudo
code below describes the operation:
| bit_adder(ci, b1, b2, *co)
| {
| value = ci + b1 + b2
| *co = !!(value & 2)
|
| return value & 1
| }
|
| #define MASK_IS_SET(mask, attr) (!!((mask)&(attr)))
| bit_position = 1
| carry = 0
| atomic_response = 0
|
| for i = 0 to 63
| {
| if ( i != 0 )
| bit_position = bit_position << 1
|
| bit_add_res = bit_adder(carry, MASK_IS_SET(*va, bit_position),
| MASK_IS_SET(compare_add, bit_position), &new_carry)
| if (bit_add_res)
| atomic_response |= bit_position
|
| carry = ((new_carry) && (!MASK_IS_SET(compare_add_mask, bit_position)))
| }
|
| return atomic_response
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-14 08:23:01 -06:00
|
|
|
IB_WR_MASKED_ATOMIC_CMP_AND_SWP,
|
|
|
|
IB_WR_MASKED_ATOMIC_FETCH_AND_ADD,
|
2014-02-23 05:19:05 -07:00
|
|
|
IB_WR_REG_SIG_MR,
|
2013-07-07 08:25:52 -06:00
|
|
|
/* reserve values for low level drivers' internal use.
|
|
|
|
* These values will not be used at all in the ib core layer.
|
|
|
|
*/
|
|
|
|
IB_WR_RESERVED1 = 0xf0,
|
|
|
|
IB_WR_RESERVED2,
|
|
|
|
IB_WR_RESERVED3,
|
|
|
|
IB_WR_RESERVED4,
|
|
|
|
IB_WR_RESERVED5,
|
|
|
|
IB_WR_RESERVED6,
|
|
|
|
IB_WR_RESERVED7,
|
|
|
|
IB_WR_RESERVED8,
|
|
|
|
IB_WR_RESERVED9,
|
|
|
|
IB_WR_RESERVED10,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_send_flags {
|
|
|
|
IB_SEND_FENCE = 1,
|
|
|
|
IB_SEND_SIGNALED = (1<<1),
|
|
|
|
IB_SEND_SOLICITED = (1<<2),
|
2008-01-30 09:30:57 -07:00
|
|
|
IB_SEND_INLINE = (1<<3),
|
2013-07-07 08:25:52 -06:00
|
|
|
IB_SEND_IP_CSUM = (1<<4),
|
|
|
|
|
|
|
|
/* reserve bits 26-31 for low level drivers' internal use */
|
|
|
|
IB_SEND_RESERVED_START = (1 << 26),
|
|
|
|
IB_SEND_RESERVED_END = (1 << 31),
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_sge {
|
|
|
|
u64 addr;
|
|
|
|
u32 length;
|
|
|
|
u32 lkey;
|
|
|
|
};
|
|
|
|
|
2015-12-11 12:53:03 -07:00
|
|
|
struct ib_cqe {
|
|
|
|
void (*done)(struct ib_cq *cq, struct ib_wc *wc);
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_send_wr {
|
|
|
|
struct ib_send_wr *next;
|
2015-12-11 12:53:03 -07:00
|
|
|
union {
|
|
|
|
u64 wr_id;
|
|
|
|
struct ib_cqe *wr_cqe;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_sge *sg_list;
|
|
|
|
int num_sge;
|
|
|
|
enum ib_wr_opcode opcode;
|
|
|
|
int send_flags;
|
2008-04-16 22:09:32 -06:00
|
|
|
union {
|
|
|
|
__be32 imm_data;
|
|
|
|
u32 invalidate_rkey;
|
|
|
|
} ex;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2015-10-08 02:16:33 -06:00
|
|
|
struct ib_rdma_wr {
|
|
|
|
struct ib_send_wr wr;
|
|
|
|
u64 remote_addr;
|
|
|
|
u32 rkey;
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct ib_rdma_wr *rdma_wr(struct ib_send_wr *wr)
|
|
|
|
{
|
|
|
|
return container_of(wr, struct ib_rdma_wr, wr);
|
|
|
|
}
|
|
|
|
|
|
|
|
struct ib_atomic_wr {
|
|
|
|
struct ib_send_wr wr;
|
|
|
|
u64 remote_addr;
|
|
|
|
u64 compare_add;
|
|
|
|
u64 swap;
|
|
|
|
u64 compare_add_mask;
|
|
|
|
u64 swap_mask;
|
|
|
|
u32 rkey;
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct ib_atomic_wr *atomic_wr(struct ib_send_wr *wr)
|
|
|
|
{
|
|
|
|
return container_of(wr, struct ib_atomic_wr, wr);
|
|
|
|
}
|
|
|
|
|
|
|
|
struct ib_ud_wr {
|
|
|
|
struct ib_send_wr wr;
|
|
|
|
struct ib_ah *ah;
|
|
|
|
void *header;
|
|
|
|
int hlen;
|
|
|
|
int mss;
|
|
|
|
u32 remote_qpn;
|
|
|
|
u32 remote_qkey;
|
|
|
|
u16 pkey_index; /* valid for GSI only */
|
|
|
|
u8 port_num; /* valid for DR SMPs on switch only */
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct ib_ud_wr *ud_wr(struct ib_send_wr *wr)
|
|
|
|
{
|
|
|
|
return container_of(wr, struct ib_ud_wr, wr);
|
|
|
|
}
|
|
|
|
|
2015-10-13 10:11:24 -06:00
|
|
|
struct ib_reg_wr {
|
|
|
|
struct ib_send_wr wr;
|
|
|
|
struct ib_mr *mr;
|
|
|
|
u32 key;
|
|
|
|
int access;
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct ib_reg_wr *reg_wr(struct ib_send_wr *wr)
|
|
|
|
{
|
|
|
|
return container_of(wr, struct ib_reg_wr, wr);
|
|
|
|
}
|
|
|
|
|
2015-10-08 02:16:33 -06:00
|
|
|
struct ib_sig_handover_wr {
|
|
|
|
struct ib_send_wr wr;
|
|
|
|
struct ib_sig_attrs *sig_attrs;
|
|
|
|
struct ib_mr *sig_mr;
|
|
|
|
int access_flags;
|
|
|
|
struct ib_sge *prot;
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct ib_sig_handover_wr *sig_handover_wr(struct ib_send_wr *wr)
|
|
|
|
{
|
|
|
|
return container_of(wr, struct ib_sig_handover_wr, wr);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_recv_wr {
|
|
|
|
struct ib_recv_wr *next;
|
2015-12-11 12:53:03 -07:00
|
|
|
union {
|
|
|
|
u64 wr_id;
|
|
|
|
struct ib_cqe *wr_cqe;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_sge *sg_list;
|
|
|
|
int num_sge;
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_access_flags {
|
|
|
|
IB_ACCESS_LOCAL_WRITE = 1,
|
|
|
|
IB_ACCESS_REMOTE_WRITE = (1<<1),
|
|
|
|
IB_ACCESS_REMOTE_READ = (1<<2),
|
|
|
|
IB_ACCESS_REMOTE_ATOMIC = (1<<3),
|
2013-02-06 09:19:12 -07:00
|
|
|
IB_ACCESS_MW_BIND = (1<<4),
|
2014-12-11 08:04:16 -07:00
|
|
|
IB_ZERO_BASED = (1<<5),
|
|
|
|
IB_ACCESS_ON_DEMAND = (1<<6),
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2015-12-23 11:12:47 -07:00
|
|
|
/*
|
|
|
|
* XXX: these are apparently used for ->rereg_user_mr, no idea why they
|
|
|
|
* are hidden here instead of a uapi header!
|
|
|
|
*/
|
2005-04-16 16:20:36 -06:00
|
|
|
enum ib_mr_rereg_flags {
|
|
|
|
IB_MR_REREG_TRANS = 1,
|
|
|
|
IB_MR_REREG_PD = (1<<1),
|
2014-07-31 02:01:28 -06:00
|
|
|
IB_MR_REREG_ACCESS = (1<<2),
|
|
|
|
IB_MR_REREG_SUPPORTED = ((IB_MR_REREG_ACCESS << 1) - 1)
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_fmr_attr {
|
|
|
|
int max_pages;
|
|
|
|
int max_maps;
|
2006-02-02 11:43:45 -07:00
|
|
|
u8 page_shift;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2014-12-11 08:04:18 -07:00
|
|
|
struct ib_umem;
|
|
|
|
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_ucontext {
|
|
|
|
struct ib_device *device;
|
|
|
|
struct list_head pd_list;
|
|
|
|
struct list_head mr_list;
|
|
|
|
struct list_head mw_list;
|
|
|
|
struct list_head cq_list;
|
|
|
|
struct list_head qp_list;
|
|
|
|
struct list_head srq_list;
|
|
|
|
struct list_head ah_list;
|
2011-05-24 09:33:46 -06:00
|
|
|
struct list_head xrcd_list;
|
2013-08-14 04:58:30 -06:00
|
|
|
struct list_head rule_list;
|
2007-03-04 17:15:11 -07:00
|
|
|
int closing;
|
2014-12-11 08:04:17 -07:00
|
|
|
|
|
|
|
struct pid *tgid;
|
2014-12-11 08:04:18 -07:00
|
|
|
#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
|
|
|
|
struct rb_root umem_tree;
|
|
|
|
/*
|
|
|
|
* Protects .umem_rbroot and tree, as well as odp_mrs_count and
|
|
|
|
* mmu notifiers registration.
|
|
|
|
*/
|
|
|
|
struct rw_semaphore umem_rwsem;
|
|
|
|
void (*invalidate_range)(struct ib_umem *umem,
|
|
|
|
unsigned long start, unsigned long end);
|
|
|
|
|
|
|
|
struct mmu_notifier mn;
|
|
|
|
atomic_t notifier_count;
|
|
|
|
/* A list of umems that don't have private mmu notifier counters yet. */
|
|
|
|
struct list_head no_private_counters;
|
|
|
|
int odp_mrs_count;
|
|
|
|
#endif
|
2005-07-07 18:57:10 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_uobject {
|
|
|
|
u64 user_handle; /* handle given to us by userspace */
|
|
|
|
struct ib_ucontext *context; /* associated user context */
|
IB/uverbs: Don't serialize with ib_uverbs_idr_mutex
Currently, all userspace verbs operations that call into the kernel
are serialized by ib_uverbs_idr_mutex. This can be a scalability
issue for some workloads, especially for devices driven by the ipath
driver, which needs to call into the kernel even for datapath
operations.
Fix this by adding reference counts to the userspace objects, and then
converting ib_uverbs_idr_mutex into a spinlock that only protects the
idrs long enough to take a reference on the object being looked up.
Because remove operations may fail, we have to do a slightly funky
two-step deletion, which is described in the comments at the top of
uverbs_cmd.c.
This also still leaves ib_uverbs_idr_lock as a single lock that is
possibly subject to contention. However, the lock hold time will only
be a single idr operation, so multiple threads should still be able to
make progress, even if ib_uverbs_idr_lock is being ping-ponged.
Surprisingly, these changes even shrink the object code:
add/remove: 23/5 grow/shrink: 4/21 up/down: 633/-693 (-60)
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-06-17 21:44:49 -06:00
|
|
|
void *object; /* containing object */
|
2005-07-07 18:57:10 -06:00
|
|
|
struct list_head list; /* link to context's list */
|
2008-04-16 22:01:06 -06:00
|
|
|
int id; /* index into kernel idr */
|
IB/uverbs: Don't serialize with ib_uverbs_idr_mutex
Currently, all userspace verbs operations that call into the kernel
are serialized by ib_uverbs_idr_mutex. This can be a scalability
issue for some workloads, especially for devices driven by the ipath
driver, which needs to call into the kernel even for datapath
operations.
Fix this by adding reference counts to the userspace objects, and then
converting ib_uverbs_idr_mutex into a spinlock that only protects the
idrs long enough to take a reference on the object being looked up.
Because remove operations may fail, we have to do a slightly funky
two-step deletion, which is described in the comments at the top of
uverbs_cmd.c.
This also still leaves ib_uverbs_idr_lock as a single lock that is
possibly subject to contention. However, the lock hold time will only
be a single idr operation, so multiple threads should still be able to
make progress, even if ib_uverbs_idr_lock is being ping-ponged.
Surprisingly, these changes even shrink the object code:
add/remove: 23/5 grow/shrink: 4/21 up/down: 633/-693 (-60)
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-06-17 21:44:49 -06:00
|
|
|
struct kref ref;
|
|
|
|
struct rw_semaphore mutex; /* protects .live */
|
2015-11-02 10:13:25 -07:00
|
|
|
struct rcu_head rcu; /* kfree_rcu() overhead */
|
IB/uverbs: Don't serialize with ib_uverbs_idr_mutex
Currently, all userspace verbs operations that call into the kernel
are serialized by ib_uverbs_idr_mutex. This can be a scalability
issue for some workloads, especially for devices driven by the ipath
driver, which needs to call into the kernel even for datapath
operations.
Fix this by adding reference counts to the userspace objects, and then
converting ib_uverbs_idr_mutex into a spinlock that only protects the
idrs long enough to take a reference on the object being looked up.
Because remove operations may fail, we have to do a slightly funky
two-step deletion, which is described in the comments at the top of
uverbs_cmd.c.
This also still leaves ib_uverbs_idr_lock as a single lock that is
possibly subject to contention. However, the lock hold time will only
be a single idr operation, so multiple threads should still be able to
make progress, even if ib_uverbs_idr_lock is being ping-ponged.
Surprisingly, these changes even shrink the object code:
add/remove: 23/5 grow/shrink: 4/21 up/down: 633/-693 (-60)
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2006-06-17 21:44:49 -06:00
|
|
|
int live;
|
2005-07-07 18:57:10 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_udata {
|
2013-12-11 15:01:44 -07:00
|
|
|
const void __user *inbuf;
|
2005-07-07 18:57:10 -06:00
|
|
|
void __user *outbuf;
|
|
|
|
size_t inlen;
|
|
|
|
size_t outlen;
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_pd {
|
2015-08-05 14:14:45 -06:00
|
|
|
u32 local_dma_lkey;
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_uobject *uobject;
|
|
|
|
atomic_t usecnt; /* count all resources */
|
2015-08-05 14:14:45 -06:00
|
|
|
struct ib_mr *local_mr;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2011-05-23 18:52:46 -06:00
|
|
|
struct ib_xrcd {
|
|
|
|
struct ib_device *device;
|
2011-05-27 00:06:44 -06:00
|
|
|
atomic_t usecnt; /* count all exposed resources */
|
2011-05-24 09:33:46 -06:00
|
|
|
struct inode *inode;
|
2011-05-27 00:06:44 -06:00
|
|
|
|
|
|
|
struct mutex tgt_qp_mutex;
|
|
|
|
struct list_head tgt_qp_list;
|
2011-05-23 18:52:46 -06:00
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_ah {
|
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_uobject *uobject;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
typedef void (*ib_comp_handler)(struct ib_cq *cq, void *cq_context);
|
|
|
|
|
2015-12-11 12:53:03 -07:00
|
|
|
enum ib_poll_context {
|
|
|
|
IB_POLL_DIRECT, /* caller context, no hw completions */
|
|
|
|
IB_POLL_SOFTIRQ, /* poll from softirq context */
|
|
|
|
IB_POLL_WORKQUEUE, /* poll from workqueue */
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_cq {
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_uobject *uobject;
|
|
|
|
ib_comp_handler comp_handler;
|
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
2008-07-15 00:48:44 -06:00
|
|
|
void *cq_context;
|
2005-07-07 18:57:10 -06:00
|
|
|
int cqe;
|
|
|
|
atomic_t usecnt; /* count number of work queues */
|
2015-12-11 12:53:03 -07:00
|
|
|
enum ib_poll_context poll_ctx;
|
|
|
|
struct ib_wc *wc;
|
|
|
|
union {
|
|
|
|
struct irq_poll iop;
|
|
|
|
struct work_struct work;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_srq {
|
2005-08-18 13:23:08 -06:00
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
|
|
|
struct ib_uobject *uobject;
|
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
|
|
|
void *srq_context;
|
2011-05-23 17:31:36 -06:00
|
|
|
enum ib_srq_type srq_type;
|
2005-04-16 16:20:36 -06:00
|
|
|
atomic_t usecnt;
|
2011-05-23 20:42:29 -06:00
|
|
|
|
|
|
|
union {
|
|
|
|
struct {
|
|
|
|
struct ib_xrcd *xrcd;
|
|
|
|
struct ib_cq *cq;
|
|
|
|
u32 srq_num;
|
|
|
|
} xrc;
|
|
|
|
} ext;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_qp {
|
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
|
|
|
struct ib_cq *send_cq;
|
|
|
|
struct ib_cq *recv_cq;
|
|
|
|
struct ib_srq *srq;
|
2011-05-23 20:59:25 -06:00
|
|
|
struct ib_xrcd *xrcd; /* XRC TGT QPs only */
|
2011-05-27 00:06:44 -06:00
|
|
|
struct list_head xrcd_list;
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
/* count times opened, mcast attaches, flow attaches */
|
|
|
|
atomic_t usecnt;
|
2011-08-08 16:31:51 -06:00
|
|
|
struct list_head open_list;
|
|
|
|
struct ib_qp *real_qp;
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_uobject *uobject;
|
2005-04-16 16:20:36 -06:00
|
|
|
void (*event_handler)(struct ib_event *, void *);
|
|
|
|
void *qp_context;
|
|
|
|
u32 qp_num;
|
|
|
|
enum ib_qp_type qp_type;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_mr {
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
|
|
|
struct ib_uobject *uobject;
|
|
|
|
u32 lkey;
|
|
|
|
u32 rkey;
|
2015-10-13 10:11:24 -06:00
|
|
|
u64 iova;
|
|
|
|
u32 length;
|
|
|
|
unsigned int page_size;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_mw {
|
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_uobject *uobject;
|
2005-04-16 16:20:36 -06:00
|
|
|
u32 rkey;
|
2013-02-06 09:19:12 -07:00
|
|
|
enum ib_mw_type type;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_fmr {
|
|
|
|
struct ib_device *device;
|
|
|
|
struct ib_pd *pd;
|
|
|
|
struct list_head list;
|
|
|
|
u32 lkey;
|
|
|
|
u32 rkey;
|
|
|
|
};
|
|
|
|
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
/* Supported steering options */
|
|
|
|
enum ib_flow_attr_type {
|
|
|
|
/* steering according to rule specifications */
|
|
|
|
IB_FLOW_ATTR_NORMAL = 0x0,
|
|
|
|
/* default unicast and multicast rule -
|
|
|
|
* receive all Eth traffic which isn't steered to any QP
|
|
|
|
*/
|
|
|
|
IB_FLOW_ATTR_ALL_DEFAULT = 0x1,
|
|
|
|
/* default multicast rule -
|
|
|
|
* receive all Eth multicast traffic which isn't steered to any QP
|
|
|
|
*/
|
|
|
|
IB_FLOW_ATTR_MC_DEFAULT = 0x2,
|
|
|
|
/* sniffer rule - receive all port traffic */
|
|
|
|
IB_FLOW_ATTR_SNIFFER = 0x3
|
|
|
|
};
|
|
|
|
|
|
|
|
/* Supported steering header types */
|
|
|
|
enum ib_flow_spec_type {
|
|
|
|
/* L2 headers*/
|
|
|
|
IB_FLOW_SPEC_ETH = 0x20,
|
2013-11-07 06:25:13 -07:00
|
|
|
IB_FLOW_SPEC_IB = 0x22,
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
/* L3 header*/
|
|
|
|
IB_FLOW_SPEC_IPV4 = 0x30,
|
|
|
|
/* L4 headers*/
|
|
|
|
IB_FLOW_SPEC_TCP = 0x40,
|
|
|
|
IB_FLOW_SPEC_UDP = 0x41
|
|
|
|
};
|
2013-11-07 06:25:13 -07:00
|
|
|
#define IB_FLOW_SPEC_LAYER_MASK 0xF0
|
2013-09-01 09:39:52 -06:00
|
|
|
#define IB_FLOW_SPEC_SUPPORT_LAYERS 4
|
|
|
|
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
/* Flow steering rule priority is set according to it's domain.
|
|
|
|
* Lower domain value means higher priority.
|
|
|
|
*/
|
|
|
|
enum ib_flow_domain {
|
|
|
|
IB_FLOW_DOMAIN_USER,
|
|
|
|
IB_FLOW_DOMAIN_ETHTOOL,
|
|
|
|
IB_FLOW_DOMAIN_RFS,
|
|
|
|
IB_FLOW_DOMAIN_NIC,
|
|
|
|
IB_FLOW_DOMAIN_NUM /* Must be last */
|
|
|
|
};
|
|
|
|
|
2016-02-18 09:31:05 -07:00
|
|
|
enum ib_flow_flags {
|
|
|
|
IB_FLOW_ATTR_FLAGS_DONT_TRAP = 1UL << 1, /* Continue match, no steal */
|
|
|
|
IB_FLOW_ATTR_FLAGS_RESERVED = 1UL << 2 /* Must be last */
|
|
|
|
};
|
|
|
|
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
struct ib_flow_eth_filter {
|
|
|
|
u8 dst_mac[6];
|
|
|
|
u8 src_mac[6];
|
|
|
|
__be16 ether_type;
|
|
|
|
__be16 vlan_tag;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_spec_eth {
|
|
|
|
enum ib_flow_spec_type type;
|
|
|
|
u16 size;
|
|
|
|
struct ib_flow_eth_filter val;
|
|
|
|
struct ib_flow_eth_filter mask;
|
|
|
|
};
|
|
|
|
|
2013-11-07 06:25:13 -07:00
|
|
|
struct ib_flow_ib_filter {
|
|
|
|
__be16 dlid;
|
|
|
|
__u8 sl;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_spec_ib {
|
|
|
|
enum ib_flow_spec_type type;
|
|
|
|
u16 size;
|
|
|
|
struct ib_flow_ib_filter val;
|
|
|
|
struct ib_flow_ib_filter mask;
|
|
|
|
};
|
|
|
|
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
struct ib_flow_ipv4_filter {
|
|
|
|
__be32 src_ip;
|
|
|
|
__be32 dst_ip;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_spec_ipv4 {
|
|
|
|
enum ib_flow_spec_type type;
|
|
|
|
u16 size;
|
|
|
|
struct ib_flow_ipv4_filter val;
|
|
|
|
struct ib_flow_ipv4_filter mask;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_tcp_udp_filter {
|
|
|
|
__be16 dst_port;
|
|
|
|
__be16 src_port;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_spec_tcp_udp {
|
|
|
|
enum ib_flow_spec_type type;
|
|
|
|
u16 size;
|
|
|
|
struct ib_flow_tcp_udp_filter val;
|
|
|
|
struct ib_flow_tcp_udp_filter mask;
|
|
|
|
};
|
|
|
|
|
|
|
|
union ib_flow_spec {
|
|
|
|
struct {
|
|
|
|
enum ib_flow_spec_type type;
|
|
|
|
u16 size;
|
|
|
|
};
|
|
|
|
struct ib_flow_spec_eth eth;
|
2013-11-07 06:25:13 -07:00
|
|
|
struct ib_flow_spec_ib ib;
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
struct ib_flow_spec_ipv4 ipv4;
|
|
|
|
struct ib_flow_spec_tcp_udp tcp_udp;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow_attr {
|
|
|
|
enum ib_flow_attr_type type;
|
|
|
|
u16 size;
|
|
|
|
u16 priority;
|
|
|
|
u32 flags;
|
|
|
|
u8 num_of_specs;
|
|
|
|
u8 port;
|
|
|
|
/* Following are the optional layers according to user request
|
|
|
|
* struct ib_flow_spec_xxx
|
|
|
|
* struct ib_flow_spec_yyy
|
|
|
|
*/
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_flow {
|
|
|
|
struct ib_qp *qp;
|
|
|
|
struct ib_uobject *uobject;
|
|
|
|
};
|
|
|
|
|
2015-06-06 12:38:31 -06:00
|
|
|
struct ib_mad_hdr;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_grh;
|
|
|
|
|
|
|
|
enum ib_process_mad_flags {
|
|
|
|
IB_MAD_IGNORE_MKEY = 1,
|
|
|
|
IB_MAD_IGNORE_BKEY = 2,
|
|
|
|
IB_MAD_IGNORE_ALL = IB_MAD_IGNORE_MKEY | IB_MAD_IGNORE_BKEY
|
|
|
|
};
|
|
|
|
|
|
|
|
enum ib_mad_result {
|
|
|
|
IB_MAD_RESULT_FAILURE = 0, /* (!SUCCESS is the important flag) */
|
|
|
|
IB_MAD_RESULT_SUCCESS = 1 << 0, /* MAD was successfully processed */
|
|
|
|
IB_MAD_RESULT_REPLY = 1 << 1, /* Reply packet needs to be sent */
|
|
|
|
IB_MAD_RESULT_CONSUMED = 1 << 2 /* Packet consumed: stop processing */
|
|
|
|
};
|
|
|
|
|
|
|
|
#define IB_DEVICE_NAME_MAX 64
|
|
|
|
|
|
|
|
struct ib_cache {
|
|
|
|
rwlock_t lock;
|
|
|
|
struct ib_event_handler event_handler;
|
|
|
|
struct ib_pkey_cache **pkey_cache;
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
struct ib_gid_table **gid_cache;
|
2006-06-17 21:37:34 -06:00
|
|
|
u8 *lmc_cache;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2006-12-12 15:27:41 -07:00
|
|
|
struct ib_dma_mapping_ops {
|
|
|
|
int (*mapping_error)(struct ib_device *dev,
|
|
|
|
u64 dma_addr);
|
|
|
|
u64 (*map_single)(struct ib_device *dev,
|
|
|
|
void *ptr, size_t size,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
void (*unmap_single)(struct ib_device *dev,
|
|
|
|
u64 addr, size_t size,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
u64 (*map_page)(struct ib_device *dev,
|
|
|
|
struct page *page, unsigned long offset,
|
|
|
|
size_t size,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
void (*unmap_page)(struct ib_device *dev,
|
|
|
|
u64 addr, size_t size,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
int (*map_sg)(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
void (*unmap_sg)(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction);
|
|
|
|
void (*sync_single_for_cpu)(struct ib_device *dev,
|
|
|
|
u64 dma_handle,
|
|
|
|
size_t size,
|
2008-07-15 00:48:44 -06:00
|
|
|
enum dma_data_direction dir);
|
2006-12-12 15:27:41 -07:00
|
|
|
void (*sync_single_for_device)(struct ib_device *dev,
|
|
|
|
u64 dma_handle,
|
|
|
|
size_t size,
|
|
|
|
enum dma_data_direction dir);
|
|
|
|
void *(*alloc_coherent)(struct ib_device *dev,
|
|
|
|
size_t size,
|
|
|
|
u64 *dma_handle,
|
|
|
|
gfp_t flag);
|
|
|
|
void (*free_coherent)(struct ib_device *dev,
|
|
|
|
size_t size, void *cpu_addr,
|
|
|
|
u64 dma_handle);
|
|
|
|
};
|
|
|
|
|
2006-08-03 15:02:42 -06:00
|
|
|
struct iw_cm_verbs;
|
|
|
|
|
2015-05-13 18:02:58 -06:00
|
|
|
struct ib_port_immutable {
|
|
|
|
int pkey_tbl_len;
|
|
|
|
int gid_tbl_len;
|
2015-05-13 18:02:59 -06:00
|
|
|
u32 core_cap_flags;
|
2015-06-06 12:38:29 -06:00
|
|
|
u32 max_mad_size;
|
2015-05-13 18:02:58 -06:00
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_device {
|
|
|
|
struct device *dma_device;
|
|
|
|
|
|
|
|
char name[IB_DEVICE_NAME_MAX];
|
|
|
|
|
|
|
|
struct list_head event_handler_list;
|
|
|
|
spinlock_t event_handler_lock;
|
|
|
|
|
2010-02-02 12:09:16 -07:00
|
|
|
spinlock_t client_data_lock;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct list_head core_list;
|
2015-07-30 08:50:14 -06:00
|
|
|
/* Access to the client_data_list is protected by the client_data_lock
|
|
|
|
* spinlock and the lists_rwsem read-write semaphore */
|
2005-04-16 16:20:36 -06:00
|
|
|
struct list_head client_data_list;
|
|
|
|
|
|
|
|
struct ib_cache cache;
|
2015-05-13 18:02:58 -06:00
|
|
|
/**
|
|
|
|
* port_immutable is indexed by port number
|
|
|
|
*/
|
|
|
|
struct ib_port_immutable *port_immutable;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2007-05-03 04:48:47 -06:00
|
|
|
int num_comp_vectors;
|
|
|
|
|
2006-08-03 15:02:42 -06:00
|
|
|
struct iw_cm_verbs *iwcm;
|
|
|
|
|
2008-07-15 00:48:48 -06:00
|
|
|
int (*get_protocol_stats)(struct ib_device *device,
|
|
|
|
union rdma_protocol_stats *stats);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*query_device)(struct ib_device *device,
|
2015-06-11 07:35:25 -06:00
|
|
|
struct ib_device_attr *device_attr,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*query_port)(struct ib_device *device,
|
|
|
|
u8 port_num,
|
|
|
|
struct ib_port_attr *port_attr);
|
2010-09-27 18:51:10 -06:00
|
|
|
enum rdma_link_layer (*get_link_layer)(struct ib_device *device,
|
|
|
|
u8 port_num);
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
/* When calling get_netdev, the HW vendor's driver should return the
|
|
|
|
* net device of device @device at port @port_num or NULL if such
|
|
|
|
* a net device doesn't exist. The vendor driver should call dev_hold
|
|
|
|
* on this net device. The HW vendor's device driver must guarantee
|
|
|
|
* that this function returns NULL before the net device reaches
|
|
|
|
* NETDEV_UNREGISTER_FINAL state.
|
|
|
|
*/
|
|
|
|
struct net_device *(*get_netdev)(struct ib_device *device,
|
|
|
|
u8 port_num);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*query_gid)(struct ib_device *device,
|
|
|
|
u8 port_num, int index,
|
|
|
|
union ib_gid *gid);
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
/* When calling add_gid, the HW vendor's driver should
|
|
|
|
* add the gid of device @device at gid index @index of
|
|
|
|
* port @port_num to be @gid. Meta-info of that gid (for example,
|
|
|
|
* the network device related to this gid is available
|
|
|
|
* at @attr. @context allows the HW vendor driver to store extra
|
|
|
|
* information together with a GID entry. The HW vendor may allocate
|
|
|
|
* memory to contain this information and store it in @context when a
|
|
|
|
* new GID entry is written to. Params are consistent until the next
|
|
|
|
* call of add_gid or delete_gid. The function should return 0 on
|
|
|
|
* success or error otherwise. The function could be called
|
|
|
|
* concurrently for different ports. This function is only called
|
|
|
|
* when roce_gid_table is used.
|
|
|
|
*/
|
|
|
|
int (*add_gid)(struct ib_device *device,
|
|
|
|
u8 port_num,
|
|
|
|
unsigned int index,
|
|
|
|
const union ib_gid *gid,
|
|
|
|
const struct ib_gid_attr *attr,
|
|
|
|
void **context);
|
|
|
|
/* When calling del_gid, the HW vendor's driver should delete the
|
|
|
|
* gid of device @device at gid index @index of port @port_num.
|
|
|
|
* Upon the deletion of a GID entry, the HW vendor must free any
|
|
|
|
* allocated memory. The caller will clear @context afterwards.
|
|
|
|
* This function is only called when roce_gid_table is used.
|
|
|
|
*/
|
|
|
|
int (*del_gid)(struct ib_device *device,
|
|
|
|
u8 port_num,
|
|
|
|
unsigned int index,
|
|
|
|
void **context);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*query_pkey)(struct ib_device *device,
|
|
|
|
u8 port_num, u16 index, u16 *pkey);
|
|
|
|
int (*modify_device)(struct ib_device *device,
|
|
|
|
int device_modify_mask,
|
|
|
|
struct ib_device_modify *device_modify);
|
|
|
|
int (*modify_port)(struct ib_device *device,
|
|
|
|
u8 port_num, int port_modify_mask,
|
|
|
|
struct ib_port_modify *port_modify);
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_ucontext * (*alloc_ucontext)(struct ib_device *device,
|
|
|
|
struct ib_udata *udata);
|
|
|
|
int (*dealloc_ucontext)(struct ib_ucontext *context);
|
|
|
|
int (*mmap)(struct ib_ucontext *context,
|
|
|
|
struct vm_area_struct *vma);
|
|
|
|
struct ib_pd * (*alloc_pd)(struct ib_device *device,
|
|
|
|
struct ib_ucontext *context,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*dealloc_pd)(struct ib_pd *pd);
|
|
|
|
struct ib_ah * (*create_ah)(struct ib_pd *pd,
|
|
|
|
struct ib_ah_attr *ah_attr);
|
|
|
|
int (*modify_ah)(struct ib_ah *ah,
|
|
|
|
struct ib_ah_attr *ah_attr);
|
|
|
|
int (*query_ah)(struct ib_ah *ah,
|
|
|
|
struct ib_ah_attr *ah_attr);
|
|
|
|
int (*destroy_ah)(struct ib_ah *ah);
|
2005-08-18 13:23:08 -06:00
|
|
|
struct ib_srq * (*create_srq)(struct ib_pd *pd,
|
|
|
|
struct ib_srq_init_attr *srq_init_attr,
|
|
|
|
struct ib_udata *udata);
|
|
|
|
int (*modify_srq)(struct ib_srq *srq,
|
|
|
|
struct ib_srq_attr *srq_attr,
|
2006-08-11 15:58:09 -06:00
|
|
|
enum ib_srq_attr_mask srq_attr_mask,
|
|
|
|
struct ib_udata *udata);
|
2005-08-18 13:23:08 -06:00
|
|
|
int (*query_srq)(struct ib_srq *srq,
|
|
|
|
struct ib_srq_attr *srq_attr);
|
|
|
|
int (*destroy_srq)(struct ib_srq *srq);
|
|
|
|
int (*post_srq_recv)(struct ib_srq *srq,
|
|
|
|
struct ib_recv_wr *recv_wr,
|
|
|
|
struct ib_recv_wr **bad_recv_wr);
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_qp * (*create_qp)(struct ib_pd *pd,
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_qp_init_attr *qp_init_attr,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*modify_qp)(struct ib_qp *qp,
|
|
|
|
struct ib_qp_attr *qp_attr,
|
2006-08-11 15:58:09 -06:00
|
|
|
int qp_attr_mask,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*query_qp)(struct ib_qp *qp,
|
|
|
|
struct ib_qp_attr *qp_attr,
|
|
|
|
int qp_attr_mask,
|
|
|
|
struct ib_qp_init_attr *qp_init_attr);
|
|
|
|
int (*destroy_qp)(struct ib_qp *qp);
|
|
|
|
int (*post_send)(struct ib_qp *qp,
|
|
|
|
struct ib_send_wr *send_wr,
|
|
|
|
struct ib_send_wr **bad_send_wr);
|
|
|
|
int (*post_recv)(struct ib_qp *qp,
|
|
|
|
struct ib_recv_wr *recv_wr,
|
|
|
|
struct ib_recv_wr **bad_recv_wr);
|
2015-06-11 07:35:20 -06:00
|
|
|
struct ib_cq * (*create_cq)(struct ib_device *device,
|
|
|
|
const struct ib_cq_init_attr *attr,
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_ucontext *context,
|
|
|
|
struct ib_udata *udata);
|
2008-04-16 22:09:33 -06:00
|
|
|
int (*modify_cq)(struct ib_cq *cq, u16 cq_count,
|
|
|
|
u16 cq_period);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*destroy_cq)(struct ib_cq *cq);
|
2006-01-30 15:29:21 -07:00
|
|
|
int (*resize_cq)(struct ib_cq *cq, int cqe,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*poll_cq)(struct ib_cq *cq, int num_entries,
|
|
|
|
struct ib_wc *wc);
|
|
|
|
int (*peek_cq)(struct ib_cq *cq, int wc_cnt);
|
|
|
|
int (*req_notify_cq)(struct ib_cq *cq,
|
IB: Return "maybe missed event" hint from ib_req_notify_cq()
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-06 22:02:48 -06:00
|
|
|
enum ib_cq_notify_flags flags);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*req_ncomp_notif)(struct ib_cq *cq,
|
|
|
|
int wc_cnt);
|
|
|
|
struct ib_mr * (*get_dma_mr)(struct ib_pd *pd,
|
|
|
|
int mr_access_flags);
|
2005-07-07 18:57:10 -06:00
|
|
|
struct ib_mr * (*reg_user_mr)(struct ib_pd *pd,
|
2007-03-04 17:15:11 -07:00
|
|
|
u64 start, u64 length,
|
|
|
|
u64 virt_addr,
|
2005-07-07 18:57:10 -06:00
|
|
|
int mr_access_flags,
|
|
|
|
struct ib_udata *udata);
|
2014-07-31 02:01:28 -06:00
|
|
|
int (*rereg_user_mr)(struct ib_mr *mr,
|
|
|
|
int flags,
|
|
|
|
u64 start, u64 length,
|
|
|
|
u64 virt_addr,
|
|
|
|
int mr_access_flags,
|
|
|
|
struct ib_pd *pd,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*dereg_mr)(struct ib_mr *mr);
|
2015-07-30 01:32:35 -06:00
|
|
|
struct ib_mr * (*alloc_mr)(struct ib_pd *pd,
|
|
|
|
enum ib_mr_type mr_type,
|
|
|
|
u32 max_num_sg);
|
2015-10-13 10:11:24 -06:00
|
|
|
int (*map_mr_sg)(struct ib_mr *mr,
|
|
|
|
struct scatterlist *sg,
|
|
|
|
int sg_nents);
|
2013-02-06 09:19:12 -07:00
|
|
|
struct ib_mw * (*alloc_mw)(struct ib_pd *pd,
|
2016-02-29 09:05:29 -07:00
|
|
|
enum ib_mw_type type,
|
|
|
|
struct ib_udata *udata);
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*dealloc_mw)(struct ib_mw *mw);
|
|
|
|
struct ib_fmr * (*alloc_fmr)(struct ib_pd *pd,
|
|
|
|
int mr_access_flags,
|
|
|
|
struct ib_fmr_attr *fmr_attr);
|
|
|
|
int (*map_phys_fmr)(struct ib_fmr *fmr,
|
|
|
|
u64 *page_list, int list_len,
|
|
|
|
u64 iova);
|
|
|
|
int (*unmap_fmr)(struct list_head *fmr_list);
|
|
|
|
int (*dealloc_fmr)(struct ib_fmr *fmr);
|
|
|
|
int (*attach_mcast)(struct ib_qp *qp,
|
|
|
|
union ib_gid *gid,
|
|
|
|
u16 lid);
|
|
|
|
int (*detach_mcast)(struct ib_qp *qp,
|
|
|
|
union ib_gid *gid,
|
|
|
|
u16 lid);
|
|
|
|
int (*process_mad)(struct ib_device *device,
|
|
|
|
int process_mad_flags,
|
|
|
|
u8 port_num,
|
2015-05-31 15:15:30 -06:00
|
|
|
const struct ib_wc *in_wc,
|
|
|
|
const struct ib_grh *in_grh,
|
2015-06-06 12:38:31 -06:00
|
|
|
const struct ib_mad_hdr *in_mad,
|
|
|
|
size_t in_mad_size,
|
|
|
|
struct ib_mad_hdr *out_mad,
|
|
|
|
size_t *out_mad_size,
|
|
|
|
u16 *out_mad_pkey_index);
|
2011-05-23 18:52:46 -06:00
|
|
|
struct ib_xrcd * (*alloc_xrcd)(struct ib_device *device,
|
|
|
|
struct ib_ucontext *ucontext,
|
|
|
|
struct ib_udata *udata);
|
|
|
|
int (*dealloc_xrcd)(struct ib_xrcd *xrcd);
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
struct ib_flow * (*create_flow)(struct ib_qp *qp,
|
|
|
|
struct ib_flow_attr
|
|
|
|
*flow_attr,
|
|
|
|
int domain);
|
|
|
|
int (*destroy_flow)(struct ib_flow *flow_id);
|
2014-02-23 05:19:05 -07:00
|
|
|
int (*check_mr_status)(struct ib_mr *mr, u32 check_mask,
|
|
|
|
struct ib_mr_status *mr_status);
|
2015-08-13 09:32:05 -06:00
|
|
|
void (*disassociate_ucontext)(struct ib_ucontext *ibcontext);
|
2016-02-17 09:15:41 -07:00
|
|
|
void (*drain_rq)(struct ib_qp *qp);
|
|
|
|
void (*drain_sq)(struct ib_qp *qp);
|
2016-03-11 13:58:38 -07:00
|
|
|
int (*set_vf_link_state)(struct ib_device *device, int vf, u8 port,
|
|
|
|
int state);
|
|
|
|
int (*get_vf_config)(struct ib_device *device, int vf, u8 port,
|
|
|
|
struct ifla_vf_info *ivf);
|
|
|
|
int (*get_vf_stats)(struct ib_device *device, int vf, u8 port,
|
|
|
|
struct ifla_vf_stats *stats);
|
|
|
|
int (*set_vf_guid)(struct ib_device *device, int vf, u8 port, u64 guid,
|
|
|
|
int type);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-12-12 15:27:41 -07:00
|
|
|
struct ib_dma_mapping_ops *dma_ops;
|
|
|
|
|
2005-07-07 18:57:10 -06:00
|
|
|
struct module *owner;
|
2008-02-21 16:13:36 -07:00
|
|
|
struct device dev;
|
2007-12-17 12:54:39 -07:00
|
|
|
struct kobject *ports_parent;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct list_head port_list;
|
|
|
|
|
|
|
|
enum {
|
|
|
|
IB_DEV_UNINITIALIZED,
|
|
|
|
IB_DEV_REGISTERED,
|
|
|
|
IB_DEV_UNREGISTERED
|
|
|
|
} reg_state;
|
|
|
|
|
2005-09-29 15:17:48 -06:00
|
|
|
int uverbs_abi_ver;
|
2010-02-02 12:09:16 -07:00
|
|
|
u64 uverbs_cmd_mask;
|
IB/core: extended command: an improved infrastructure for uverbs commands
Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs
commands") added an infrastructure for extensible uverbs commands
while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow
through uverbs") exported ib_create_flow()/ib_destroy_flow() functions
using this new infrastructure.
According to the commit 400dbc96583f, the purpose of this
infrastructure is to support passing around provider (eg. hardware)
specific buffers when userspace issue commands to the kernel, so that
it would be possible to extend uverbs (eg. core) buffers independently
from the provider buffers.
But the new kernel command function prototypes were not modified to
take advantage of this extension. This issue was exposed by Roland
Dreier in a previous review[1].
So the following patch is an attempt to a revised extensible command
infrastructure.
This improved extensible command infrastructure distinguish between
core (eg. legacy)'s command/response buffers from provider
(eg. hardware)'s command/response buffers: each extended command
implementing function is given a struct ib_udata to hold core
(eg. uverbs) input and output buffers, and another struct ib_udata to
hold the hw (eg. provider) input and output buffers.
Having those buffers identified separately make it easier to increase
one buffer to support extension without having to add some code to
guess the exact size of each command/response parts: This should make
the extended functions more reliable.
Additionally, instead of relying on command identifier being greater
than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on
unused bits in command field: on the 32 bits provided by command
field, only 6 bits are really needed to encode the identifier of
commands currently supported by the kernel. (Even using only 6 bits
leaves room for about 23 new commands).
So this patch makes use of some high order bits in command field to
store flags, leaving enough room for more command identifiers than one
will ever need (eg. 256).
The new flags are used to specify if the command should be processed
as an extended one or a legacy one. While designing the new command
format, care was taken to make usage of flags itself extensible.
Using high order bits of the commands field ensure that newer
libibverbs on older kernel will properly fail when trying to call
extended commands. On the other hand, older libibverbs on newer kernel
will never be able to issue calls to extended commands.
The extended command header includes the optional response pointer so
that output buffer length and output buffer pointer are located
together in the command, allowing proper parameters checking. This
should make implementing functions easier and safer.
Additionally the extended header ensure 64bits alignment, while making
all sizes multiple of 8 bytes, extending the maximum buffer size:
legacy extended
Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes)
Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes)
For the purpose of doing proper buffer size accounting, the headers
size are no more taken in account in "in_words".
One of the odds of the current extensible infrastructure, reading
twice the "legacy" command header, is fixed by removing the "legacy"
command header from the extended command header: they are processed as
two different parts of the command: memory is read once and
information are not duplicated: it's making clear that's an extended
command scheme and not a different command scheme.
The proposed scheme will format input (command) and output (response)
buffers this way:
- command:
legacy header +
extended header +
command data (core + hw):
+----------------------------------------+
| flags | 00 00 | command |
| in_words | out_words |
+----------------------------------------+
| response |
| response |
| provider_in_words | provider_out_words |
| padding |
+----------------------------------------+
| |
. <uverbs input> .
. (in_words * 8) .
| |
+----------------------------------------+
| |
. <provider input> .
. (provider_in_words * 8) .
| |
+----------------------------------------+
- response, if present:
+----------------------------------------+
| |
. <uverbs output space> .
. (out_words * 8) .
| |
+----------------------------------------+
| |
. <provider output space> .
. (provider_out_words * 8) .
| |
+----------------------------------------+
The overall design is to ensure that the extensible infrastructure is
itself extensible while begin more reliable with more input and bound
checking.
Note:
The unused field in the extended header would be perfect candidate to
hold the command "comp_mask" (eg. bit field used to handle
compatibility). This was suggested by Roland Dreier in a previous
review[2]. But "comp_mask" field is likely to be present in the uverb
input and/or provider input, likewise for the response, as noted by
Matan Barak[3], so it doesn't make sense to put "comp_mask" in the
header.
[1]:
http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com
[2]:
http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com
[3]:
http://marc.info/?i=525C1149.6000701@mellanox.com
Signed-off-by: Yann Droneaud <ydroneaud@opteya.com>
Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com
[ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ]
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 15:21:49 -07:00
|
|
|
u64 uverbs_ex_cmd_mask;
|
2005-09-29 15:17:48 -06:00
|
|
|
|
2006-02-02 10:47:14 -07:00
|
|
|
char node_desc[64];
|
2006-01-10 08:39:34 -07:00
|
|
|
__be64 node_guid;
|
2008-07-15 00:48:53 -06:00
|
|
|
u32 local_dma_lkey;
|
2015-06-29 07:57:00 -06:00
|
|
|
u16 is_switch:1;
|
2005-04-16 16:20:36 -06:00
|
|
|
u8 node_type;
|
|
|
|
u8 phys_port_cnt;
|
2015-12-18 01:59:44 -07:00
|
|
|
struct ib_device_attr attrs;
|
2015-05-13 18:02:58 -06:00
|
|
|
|
|
|
|
/**
|
|
|
|
* The following mandatory functions are used only at device
|
|
|
|
* registration. Keep functions such as these at the end of this
|
|
|
|
* structure to avoid cache line misses when accessing struct ib_device
|
|
|
|
* in fast paths.
|
|
|
|
*/
|
|
|
|
int (*get_port_immutable)(struct ib_device *, u8, struct ib_port_immutable *);
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_client {
|
|
|
|
char *name;
|
|
|
|
void (*add) (struct ib_device *);
|
2015-07-30 08:50:14 -06:00
|
|
|
void (*remove)(struct ib_device *, void *client_data);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2015-07-30 08:50:15 -06:00
|
|
|
/* Returns the net_dev belonging to this ib_client and matching the
|
|
|
|
* given parameters.
|
|
|
|
* @dev: An RDMA device that the net_dev use for communication.
|
|
|
|
* @port: A physical port number on the RDMA device.
|
|
|
|
* @pkey: P_Key that the net_dev uses if applicable.
|
|
|
|
* @gid: A GID that the net_dev uses to communicate.
|
|
|
|
* @addr: An IP address the net_dev is configured with.
|
|
|
|
* @client_data: The device's client data set by ib_set_client_data().
|
|
|
|
*
|
|
|
|
* An ib_client that implements a net_dev on top of RDMA devices
|
|
|
|
* (such as IP over IB) should implement this callback, allowing the
|
|
|
|
* rdma_cm module to find the right net_dev for a given request.
|
|
|
|
*
|
|
|
|
* The caller is responsible for calling dev_put on the returned
|
|
|
|
* netdev. */
|
|
|
|
struct net_device *(*get_net_dev_by_params)(
|
|
|
|
struct ib_device *dev,
|
|
|
|
u8 port,
|
|
|
|
u16 pkey,
|
|
|
|
const union ib_gid *gid,
|
|
|
|
const struct sockaddr *addr,
|
|
|
|
void *client_data);
|
2005-04-16 16:20:36 -06:00
|
|
|
struct list_head list;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct ib_device *ib_alloc_device(size_t size);
|
|
|
|
void ib_dealloc_device(struct ib_device *device);
|
|
|
|
|
2010-05-06 18:03:25 -06:00
|
|
|
int ib_register_device(struct ib_device *device,
|
|
|
|
int (*port_callback)(struct ib_device *,
|
|
|
|
u8, struct kobject *));
|
2005-04-16 16:20:36 -06:00
|
|
|
void ib_unregister_device(struct ib_device *device);
|
|
|
|
|
|
|
|
int ib_register_client (struct ib_client *client);
|
|
|
|
void ib_unregister_client(struct ib_client *client);
|
|
|
|
|
|
|
|
void *ib_get_client_data(struct ib_device *device, struct ib_client *client);
|
|
|
|
void ib_set_client_data(struct ib_device *device, struct ib_client *client,
|
|
|
|
void *data);
|
|
|
|
|
2005-07-07 18:57:10 -06:00
|
|
|
static inline int ib_copy_from_udata(void *dest, struct ib_udata *udata, size_t len)
|
|
|
|
{
|
|
|
|
return copy_from_user(dest, udata->inbuf, len) ? -EFAULT : 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int ib_copy_to_udata(struct ib_udata *udata, void *src, size_t len)
|
|
|
|
{
|
2015-02-05 14:10:18 -07:00
|
|
|
return copy_to_user(udata->outbuf, src, len) ? -EFAULT : 0;
|
2005-07-07 18:57:10 -06:00
|
|
|
}
|
|
|
|
|
2015-12-15 11:30:10 -07:00
|
|
|
static inline bool ib_is_udata_cleared(struct ib_udata *udata,
|
|
|
|
size_t offset,
|
|
|
|
size_t len)
|
|
|
|
{
|
|
|
|
const void __user *p = udata->inbuf + offset;
|
|
|
|
bool ret = false;
|
|
|
|
u8 *buf;
|
|
|
|
|
|
|
|
if (len > USHRT_MAX)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
buf = kmalloc(len, GFP_KERNEL);
|
|
|
|
if (!buf)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
if (copy_from_user(buf, p, len))
|
|
|
|
goto free;
|
|
|
|
|
|
|
|
ret = !memchr_inv(buf, 0, len);
|
|
|
|
|
|
|
|
free:
|
|
|
|
kfree(buf);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2006-02-13 13:48:12 -07:00
|
|
|
/**
|
|
|
|
* ib_modify_qp_is_ok - Check that the supplied attribute mask
|
|
|
|
* contains all required attributes and no attributes not allowed for
|
|
|
|
* the given QP state transition.
|
|
|
|
* @cur_state: Current QP state
|
|
|
|
* @next_state: Next QP state
|
|
|
|
* @type: QP type
|
|
|
|
* @mask: Mask of supplied QP attributes
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
* @ll : link layer of port
|
2006-02-13 13:48:12 -07:00
|
|
|
*
|
|
|
|
* This function is a helper function that a low-level driver's
|
|
|
|
* modify_qp method can use to validate the consumer's input. It
|
|
|
|
* checks that cur_state and next_state are valid QP states, that a
|
|
|
|
* transition from cur_state to next_state is allowed by the IB spec,
|
|
|
|
* and that the attribute mask supplied is allowed for the transition.
|
|
|
|
*/
|
|
|
|
int ib_modify_qp_is_ok(enum ib_qp_state cur_state, enum ib_qp_state next_state,
|
IB/core: Ethernet L2 attributes in verbs/cm structures
This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.
When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.
Thus, those attributes were added to the following structures:
* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id
For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.
On the active side, the CM fills. its internal structures from the
path provided by the ULP. We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).
On the passive side, the CM fills its internal structures from the WC
associated with the REQ message. We add there taking the ETH L2
attributes from the WC.
When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.
ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB. Vendor drivers are modified to support the new
function signature.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-12-12 09:03:11 -07:00
|
|
|
enum ib_qp_type type, enum ib_qp_attr_mask mask,
|
|
|
|
enum rdma_link_layer ll);
|
2006-02-13 13:48:12 -07:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
int ib_register_event_handler (struct ib_event_handler *event_handler);
|
|
|
|
int ib_unregister_event_handler(struct ib_event_handler *event_handler);
|
|
|
|
void ib_dispatch_event(struct ib_event *event);
|
|
|
|
|
|
|
|
int ib_query_port(struct ib_device *device,
|
|
|
|
u8 port_num, struct ib_port_attr *port_attr);
|
|
|
|
|
2010-09-27 18:51:10 -06:00
|
|
|
enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device,
|
|
|
|
u8 port_num);
|
|
|
|
|
2015-06-29 07:57:00 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_ib_switch - Check if the device is IB switch
|
|
|
|
* @device: Device to check
|
|
|
|
*
|
|
|
|
* Device driver is responsible for setting is_switch bit on
|
|
|
|
* in ib_device structure at init time.
|
|
|
|
*
|
|
|
|
* Return: true if the device is IB switch.
|
|
|
|
*/
|
|
|
|
static inline bool rdma_cap_ib_switch(const struct ib_device *device)
|
|
|
|
{
|
|
|
|
return device->is_switch;
|
|
|
|
}
|
|
|
|
|
2015-05-13 18:02:55 -06:00
|
|
|
/**
|
|
|
|
* rdma_start_port - Return the first valid port number for the device
|
|
|
|
* specified
|
|
|
|
*
|
|
|
|
* @device: Device to be checked
|
|
|
|
*
|
|
|
|
* Return start port number
|
|
|
|
*/
|
|
|
|
static inline u8 rdma_start_port(const struct ib_device *device)
|
|
|
|
{
|
2015-06-29 07:57:00 -06:00
|
|
|
return rdma_cap_ib_switch(device) ? 0 : 1;
|
2015-05-13 18:02:55 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* rdma_end_port - Return the last valid port number for the device
|
|
|
|
* specified
|
|
|
|
*
|
|
|
|
* @device: Device to be checked
|
|
|
|
*
|
|
|
|
* Return last port number
|
|
|
|
*/
|
|
|
|
static inline u8 rdma_end_port(const struct ib_device *device)
|
|
|
|
{
|
2015-06-29 07:57:00 -06:00
|
|
|
return rdma_cap_ib_switch(device) ? 0 : device->phys_port_cnt;
|
2015-05-13 18:02:55 -06:00
|
|
|
}
|
|
|
|
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_protocol_ib(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:19 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_IB;
|
2015-05-05 06:50:19 -06:00
|
|
|
}
|
|
|
|
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_protocol_roce(const struct ib_device *device, u8 port_num)
|
2015-12-23 05:56:50 -07:00
|
|
|
{
|
|
|
|
return device->port_immutable[port_num].core_cap_flags &
|
|
|
|
(RDMA_CORE_CAP_PROT_ROCE | RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool rdma_protocol_roce_udp_encap(const struct ib_device *device, u8 port_num)
|
|
|
|
{
|
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool rdma_protocol_roce_eth_encap(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:19 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_ROCE;
|
2015-05-05 06:50:19 -06:00
|
|
|
}
|
|
|
|
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_protocol_iwarp(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:19 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_IWARP;
|
2015-05-05 06:50:19 -06:00
|
|
|
}
|
|
|
|
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_ib_or_roce(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:19 -06:00
|
|
|
{
|
2015-12-23 05:56:50 -07:00
|
|
|
return rdma_protocol_ib(device, port_num) ||
|
|
|
|
rdma_protocol_roce(device, port_num);
|
2015-05-05 06:50:19 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:32 -06:00
|
|
|
/**
|
2015-05-18 02:41:45 -06:00
|
|
|
* rdma_cap_ib_mad - Check if the port of a device supports Infiniband
|
2015-05-05 06:50:32 -06:00
|
|
|
* Management Datagrams.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:32 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Management Datagrams (MAD) are a required part of the InfiniBand
|
|
|
|
* specification and are supported on all InfiniBand devices. A slightly
|
|
|
|
* extended version are also supported on OPA interfaces.
|
2015-05-05 06:50:32 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port supports sending/receiving of MAD packets.
|
2015-05-05 06:50:32 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_ib_mad(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:32 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_IB_MAD;
|
2015-05-05 06:50:32 -06:00
|
|
|
}
|
|
|
|
|
2015-06-06 12:38:32 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_opa_mad - Check if the port of device provides support for OPA
|
|
|
|
* Management Datagrams.
|
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
|
|
|
*
|
|
|
|
* Intel OmniPath devices extend and/or replace the InfiniBand Management
|
|
|
|
* datagrams with their own versions. These OPA MADs share many but not all of
|
|
|
|
* the characteristics of InfiniBand MADs.
|
|
|
|
*
|
|
|
|
* OPA MADs differ in the following ways:
|
|
|
|
*
|
|
|
|
* 1) MADs are variable size up to 2K
|
|
|
|
* IBTA defined MADs remain fixed at 256 bytes
|
|
|
|
* 2) OPA SMPs must carry valid PKeys
|
|
|
|
* 3) OPA SMP packets are a different format
|
|
|
|
*
|
|
|
|
* Return: true if the port supports OPA MAD packet formats.
|
|
|
|
*/
|
|
|
|
static inline bool rdma_cap_opa_mad(struct ib_device *device, u8 port_num)
|
|
|
|
{
|
|
|
|
return (device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_OPA_MAD)
|
|
|
|
== RDMA_CORE_CAP_OPA_MAD;
|
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:33 -06:00
|
|
|
/**
|
2015-05-18 02:41:45 -06:00
|
|
|
* rdma_cap_ib_smi - Check if the port of a device provides an Infiniband
|
|
|
|
* Subnet Management Agent (SMA) on the Subnet Management Interface (SMI).
|
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:33 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Each InfiniBand node is required to provide a Subnet Management Agent
|
|
|
|
* that the subnet manager can access. Prior to the fabric being fully
|
|
|
|
* configured by the subnet manager, the SMA is accessed via a well known
|
|
|
|
* interface called the Subnet Management Interface (SMI). This interface
|
|
|
|
* uses directed route packets to communicate with the SM to get around the
|
|
|
|
* chicken and egg problem of the SM needing to know what's on the fabric
|
|
|
|
* in order to configure the fabric, and needing to configure the fabric in
|
|
|
|
* order to send packets to the devices on the fabric. These directed
|
|
|
|
* route packets do not need the fabric fully configured in order to reach
|
|
|
|
* their destination. The SMI is the only method allowed to send
|
|
|
|
* directed route packets on an InfiniBand fabric.
|
2015-05-05 06:50:33 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port provides an SMI.
|
2015-05-05 06:50:33 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_ib_smi(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:33 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_IB_SMI;
|
2015-05-05 06:50:33 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:34 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_ib_cm - Check if the port of device has the capability Infiniband
|
|
|
|
* Communication Manager.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:34 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* The InfiniBand Communication Manager is one of many pre-defined General
|
|
|
|
* Service Agents (GSA) that are accessed via the General Service
|
|
|
|
* Interface (GSI). It's role is to facilitate establishment of connections
|
|
|
|
* between nodes as well as other management related tasks for established
|
|
|
|
* connections.
|
2015-05-05 06:50:34 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port supports an IB CM (this does not guarantee that
|
|
|
|
* a CM is actually running however).
|
2015-05-05 06:50:34 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_ib_cm(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:34 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_IB_CM;
|
2015-05-05 06:50:34 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:35 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_iw_cm - Check if the port of device has the capability IWARP
|
|
|
|
* Communication Manager.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:35 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Similar to above, but specific to iWARP connections which have a different
|
|
|
|
* managment protocol than InfiniBand.
|
2015-05-05 06:50:35 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port supports an iWARP CM (this does not guarantee that
|
|
|
|
* a CM is actually running however).
|
2015-05-05 06:50:35 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_iw_cm(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:35 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_IW_CM;
|
2015-05-05 06:50:35 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:36 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_ib_sa - Check if the port of device has the capability Infiniband
|
|
|
|
* Subnet Administration.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:36 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* An InfiniBand Subnet Administration (SA) service is a pre-defined General
|
|
|
|
* Service Agent (GSA) provided by the Subnet Manager (SM). On InfiniBand
|
|
|
|
* fabrics, devices should resolve routes to other hosts by contacting the
|
|
|
|
* SA to query the proper route.
|
2015-05-05 06:50:36 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port should act as a client to the fabric Subnet
|
|
|
|
* Administration interface. This does not imply that the SA service is
|
|
|
|
* running locally.
|
2015-05-05 06:50:36 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_ib_sa(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:36 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_IB_SA;
|
2015-05-05 06:50:36 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:37 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_ib_mcast - Check if the port of device has the capability Infiniband
|
|
|
|
* Multicast.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:37 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* InfiniBand multicast registration is more complex than normal IPv4 or
|
|
|
|
* IPv6 multicast registration. Each Host Channel Adapter must register
|
|
|
|
* with the Subnet Manager when it wishes to join a multicast group. It
|
|
|
|
* should do so only once regardless of how many queue pairs it subscribes
|
|
|
|
* to this group. And it should leave the group only after all queue pairs
|
|
|
|
* attached to the group have been detached.
|
2015-05-05 06:50:37 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port must undertake the additional adminstrative
|
|
|
|
* overhead of registering/unregistering with the SM and tracking of the
|
|
|
|
* total number of queue pairs attached to the multicast group.
|
2015-05-05 06:50:37 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_ib_mcast(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:37 -06:00
|
|
|
{
|
|
|
|
return rdma_cap_ib_sa(device, port_num);
|
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:39 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_af_ib - Check if the port of device has the capability
|
|
|
|
* Native Infiniband Address.
|
2015-05-18 02:41:45 -06:00
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:39 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* InfiniBand addressing uses a port's GUID + Subnet Prefix to make a default
|
|
|
|
* GID. RoCE uses a different mechanism, but still generates a GID via
|
|
|
|
* a prescribed mechanism and port specific data.
|
2015-05-05 06:50:39 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if the port uses a GID address to identify devices on the
|
|
|
|
* network.
|
2015-05-05 06:50:39 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_af_ib(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:39 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_AF_IB;
|
2015-05-05 06:50:39 -06:00
|
|
|
}
|
|
|
|
|
2015-05-05 06:50:40 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_eth_ah - Check if the port of device has the capability
|
2015-05-18 02:41:45 -06:00
|
|
|
* Ethernet Address Handle.
|
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
2015-05-05 06:50:40 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* RoCE is InfiniBand over Ethernet, and it uses a well defined technique
|
|
|
|
* to fabricate GIDs over Ethernet/IP specific addresses native to the
|
|
|
|
* port. Normally, packet headers are generated by the sending host
|
|
|
|
* adapter, but when sending connectionless datagrams, we must manually
|
|
|
|
* inject the proper headers for the fabric we are communicating over.
|
2015-05-05 06:50:40 -06:00
|
|
|
*
|
2015-05-18 02:41:45 -06:00
|
|
|
* Return: true if we are running as a RoCE port and must force the
|
|
|
|
* addition of a Global Route Header built from our Ethernet Address
|
|
|
|
* Handle into our header list for connectionless packets.
|
2015-05-05 06:50:40 -06:00
|
|
|
*/
|
2015-05-31 15:15:29 -06:00
|
|
|
static inline bool rdma_cap_eth_ah(const struct ib_device *device, u8 port_num)
|
2015-05-05 06:50:40 -06:00
|
|
|
{
|
2015-05-13 18:02:59 -06:00
|
|
|
return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_ETH_AH;
|
2015-05-05 06:50:40 -06:00
|
|
|
}
|
|
|
|
|
2015-06-06 12:38:29 -06:00
|
|
|
/**
|
|
|
|
* rdma_max_mad_size - Return the max MAD size required by this RDMA Port.
|
|
|
|
*
|
|
|
|
* @device: Device
|
|
|
|
* @port_num: Port number
|
|
|
|
*
|
|
|
|
* This MAD size includes the MAD headers and MAD payload. No other headers
|
|
|
|
* are included.
|
|
|
|
*
|
|
|
|
* Return the max MAD size required by the Port. Will return 0 if the port
|
|
|
|
* does not support MADs
|
|
|
|
*/
|
|
|
|
static inline size_t rdma_max_mad_size(const struct ib_device *device, u8 port_num)
|
|
|
|
{
|
|
|
|
return device->port_immutable[port_num].max_mad_size;
|
|
|
|
}
|
|
|
|
|
IB/core: Add RoCE GID table management
RoCE GIDs are based on IP addresses configured on Ethernet net-devices
which relate to the RDMA (RoCE) device port.
Currently, each of the low-level drivers that support RoCE (ocrdma,
mlx4) manages its own RoCE port GID table. As there's nothing which is
essentially vendor specific, we generalize that, and enhance the RDMA
core GID cache to do this job.
In order to populate the GID table, we listen for events:
(a) netdev up/down/change_addr events - if a netdev is built onto
our RoCE device, we need to add/delete its IPs. This involves
adding all GIDs related to this ndev, add default GIDs, etc.
(b) inet events - add new GIDs (according to the IP addresses)
to the table.
For programming the port RoCE GID table, providers must implement
the add_gid and del_gid callbacks.
RoCE GID management requires us to state the associated net_device
alongside the GID. This information is necessary in order to manage
the GID table. For example, when a net_device is removed, its
associated GIDs need to be removed as well.
RoCE mandates generating a default GID for each port, based on the
related net-device's IPv6 link local. In contrast to the GID based on
the regular IPv6 link-local (as we generate GID per IP address),
the default GID is also available when the net device is down (in
order to support loopback).
Locking is done as follows:
The patch modify the GID table code both for new RoCE drivers
implementing the add_gid/del_gid callbacks and for current RoCE and
IB drivers that do not. The flows for updating the table are
different, so the locking requirements are too.
While updating RoCE GID table, protection against multiple writers is
achieved via mutex_lock(&table->lock). Since writing to a table
requires us to find an entry (possible a free entry) in the table and
then modify it, this mutex protects both the find_gid and write_gid
ensuring the atomicity of the action.
Each entry in the GID cache is protected by rwlock. In RoCE, writing
(usually results from netdev notifier) involves invoking the vendor's
add_gid and del_gid callbacks, which could sleep.
Therefore, an invalid flag is added for each entry. Updates for RoCE are
done via a workqueue, thus sleeping is permitted.
In IB, updates are done in write_lock_irq(&device->cache.lock), thus
write_gid isn't allowed to sleep and add_gid/del_gid are not called.
When passing net-device into/out-of the GID cache, the device
is always passed held (dev_hold).
The code uses a single work item for updating all RDMA devices,
following a netdev or inet notifier.
The patch moves the cache from being a client (which was incorrect,
as the cache is part of the IB infrastructure) to being explicitly
initialized/freed when a device is registered/removed.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-07-30 09:33:26 -06:00
|
|
|
/**
|
|
|
|
* rdma_cap_roce_gid_table - Check if the port of device uses roce_gid_table
|
|
|
|
* @device: Device to check
|
|
|
|
* @port_num: Port number to check
|
|
|
|
*
|
|
|
|
* RoCE GID table mechanism manages the various GIDs for a device.
|
|
|
|
*
|
|
|
|
* NOTE: if allocating the port's GID table has failed, this call will still
|
|
|
|
* return true, but any RoCE GID table API will fail.
|
|
|
|
*
|
|
|
|
* Return: true if the port uses RoCE GID table mechanism in order to manage
|
|
|
|
* its GIDs.
|
|
|
|
*/
|
|
|
|
static inline bool rdma_cap_roce_gid_table(const struct ib_device *device,
|
|
|
|
u8 port_num)
|
|
|
|
{
|
|
|
|
return rdma_protocol_roce(device, port_num) &&
|
|
|
|
device->add_gid && device->del_gid;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
int ib_query_gid(struct ib_device *device,
|
2015-10-15 09:38:45 -06:00
|
|
|
u8 port_num, int index, union ib_gid *gid,
|
|
|
|
struct ib_gid_attr *attr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2016-03-11 13:58:38 -07:00
|
|
|
int ib_set_vf_link_state(struct ib_device *device, int vf, u8 port,
|
|
|
|
int state);
|
|
|
|
int ib_get_vf_config(struct ib_device *device, int vf, u8 port,
|
|
|
|
struct ifla_vf_info *info);
|
|
|
|
int ib_get_vf_stats(struct ib_device *device, int vf, u8 port,
|
|
|
|
struct ifla_vf_stats *stats);
|
|
|
|
int ib_set_vf_guid(struct ib_device *device, int vf, u8 port, u64 guid,
|
|
|
|
int type);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
int ib_query_pkey(struct ib_device *device,
|
|
|
|
u8 port_num, u16 index, u16 *pkey);
|
|
|
|
|
|
|
|
int ib_modify_device(struct ib_device *device,
|
|
|
|
int device_modify_mask,
|
|
|
|
struct ib_device_modify *device_modify);
|
|
|
|
|
|
|
|
int ib_modify_port(struct ib_device *device,
|
|
|
|
u8 port_num, int port_modify_mask,
|
|
|
|
struct ib_port_modify *port_modify);
|
|
|
|
|
2007-05-13 22:26:51 -06:00
|
|
|
int ib_find_gid(struct ib_device *device, union ib_gid *gid,
|
2015-12-23 05:56:47 -07:00
|
|
|
enum ib_gid_type gid_type, struct net_device *ndev,
|
|
|
|
u8 *port_num, u16 *index);
|
2007-05-13 22:26:51 -06:00
|
|
|
|
|
|
|
int ib_find_pkey(struct ib_device *device,
|
|
|
|
u8 port_num, u16 pkey, u16 *index);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ib_pd *ib_alloc_pd(struct ib_device *device);
|
|
|
|
|
2015-08-05 14:34:31 -06:00
|
|
|
void ib_dealloc_pd(struct ib_pd *pd);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_create_ah - Creates an address handle for the given address vector.
|
|
|
|
* @pd: The protection domain associated with the address handle.
|
|
|
|
* @ah_attr: The attributes of the address vector.
|
|
|
|
*
|
|
|
|
* The address handle is used to reference a local or global destination
|
|
|
|
* in all UD QP post sends.
|
|
|
|
*/
|
|
|
|
struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr);
|
|
|
|
|
2006-06-17 21:37:39 -06:00
|
|
|
/**
|
|
|
|
* ib_init_ah_from_wc - Initializes address handle attributes from a
|
|
|
|
* work completion.
|
|
|
|
* @device: Device on which the received message arrived.
|
|
|
|
* @port_num: Port on which the received message arrived.
|
|
|
|
* @wc: Work completion associated with the received message.
|
|
|
|
* @grh: References the received global route header. This parameter is
|
|
|
|
* ignored unless the work completion indicates that the GRH is valid.
|
|
|
|
* @ah_attr: Returned attributes that can be used when creating an address
|
|
|
|
* handle for replying to the message.
|
|
|
|
*/
|
2015-05-31 15:15:31 -06:00
|
|
|
int ib_init_ah_from_wc(struct ib_device *device, u8 port_num,
|
|
|
|
const struct ib_wc *wc, const struct ib_grh *grh,
|
|
|
|
struct ib_ah_attr *ah_attr);
|
2006-06-17 21:37:39 -06:00
|
|
|
|
2005-07-27 12:45:34 -06:00
|
|
|
/**
|
|
|
|
* ib_create_ah_from_wc - Creates an address handle associated with the
|
|
|
|
* sender of the specified work completion.
|
|
|
|
* @pd: The protection domain associated with the address handle.
|
|
|
|
* @wc: Work completion information associated with a received message.
|
|
|
|
* @grh: References the received global route header. This parameter is
|
|
|
|
* ignored unless the work completion indicates that the GRH is valid.
|
|
|
|
* @port_num: The outbound port number to associate with the address.
|
|
|
|
*
|
|
|
|
* The address handle is used to reference a local or global destination
|
|
|
|
* in all UD QP post sends.
|
|
|
|
*/
|
2015-05-31 15:15:31 -06:00
|
|
|
struct ib_ah *ib_create_ah_from_wc(struct ib_pd *pd, const struct ib_wc *wc,
|
|
|
|
const struct ib_grh *grh, u8 port_num);
|
2005-07-27 12:45:34 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_modify_ah - Modifies the address vector associated with an address
|
|
|
|
* handle.
|
|
|
|
* @ah: The address handle to modify.
|
|
|
|
* @ah_attr: The new address vector attributes to associate with the
|
|
|
|
* address handle.
|
|
|
|
*/
|
|
|
|
int ib_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_query_ah - Queries the address vector associated with an address
|
|
|
|
* handle.
|
|
|
|
* @ah: The address handle to query.
|
|
|
|
* @ah_attr: The address vector attributes associated with the address
|
|
|
|
* handle.
|
|
|
|
*/
|
|
|
|
int ib_query_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_destroy_ah - Destroys an address handle.
|
|
|
|
* @ah: The address handle to destroy.
|
|
|
|
*/
|
|
|
|
int ib_destroy_ah(struct ib_ah *ah);
|
|
|
|
|
2005-08-18 13:23:08 -06:00
|
|
|
/**
|
|
|
|
* ib_create_srq - Creates a SRQ associated with the specified protection
|
|
|
|
* domain.
|
|
|
|
* @pd: The protection domain associated with the SRQ.
|
2006-02-23 13:13:51 -07:00
|
|
|
* @srq_init_attr: A list of initial attributes required to create the
|
|
|
|
* SRQ. If SRQ creation succeeds, then the attributes are updated to
|
|
|
|
* the actual capabilities of the created SRQ.
|
2005-08-18 13:23:08 -06:00
|
|
|
*
|
|
|
|
* srq_attr->max_wr and srq_attr->max_sge are read the determine the
|
|
|
|
* requested size of the SRQ, and set to the actual values allocated
|
|
|
|
* on return. If ib_create_srq() succeeds, then max_wr and max_sge
|
|
|
|
* will always be at least as large as the requested values.
|
|
|
|
*/
|
|
|
|
struct ib_srq *ib_create_srq(struct ib_pd *pd,
|
|
|
|
struct ib_srq_init_attr *srq_init_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_modify_srq - Modifies the attributes for the specified SRQ.
|
|
|
|
* @srq: The SRQ to modify.
|
|
|
|
* @srq_attr: On input, specifies the SRQ attributes to modify. On output,
|
|
|
|
* the current values of selected SRQ attributes are returned.
|
|
|
|
* @srq_attr_mask: A bit-mask used to specify which attributes of the SRQ
|
|
|
|
* are being modified.
|
|
|
|
*
|
|
|
|
* The mask may contain IB_SRQ_MAX_WR to resize the SRQ and/or
|
|
|
|
* IB_SRQ_LIMIT to set the SRQ's limit and request notification when
|
|
|
|
* the number of receives queued drops below the limit.
|
|
|
|
*/
|
|
|
|
int ib_modify_srq(struct ib_srq *srq,
|
|
|
|
struct ib_srq_attr *srq_attr,
|
|
|
|
enum ib_srq_attr_mask srq_attr_mask);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_query_srq - Returns the attribute list and current values for the
|
|
|
|
* specified SRQ.
|
|
|
|
* @srq: The SRQ to query.
|
|
|
|
* @srq_attr: The attributes of the specified SRQ.
|
|
|
|
*/
|
|
|
|
int ib_query_srq(struct ib_srq *srq,
|
|
|
|
struct ib_srq_attr *srq_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_destroy_srq - Destroys the specified SRQ.
|
|
|
|
* @srq: The SRQ to destroy.
|
|
|
|
*/
|
|
|
|
int ib_destroy_srq(struct ib_srq *srq);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_post_srq_recv - Posts a list of work requests to the specified SRQ.
|
|
|
|
* @srq: The SRQ to post the work request on.
|
|
|
|
* @recv_wr: A list of work requests to post on the receive queue.
|
|
|
|
* @bad_recv_wr: On an immediate failure, this parameter will reference
|
|
|
|
* the work request that failed to be posted on the QP.
|
|
|
|
*/
|
|
|
|
static inline int ib_post_srq_recv(struct ib_srq *srq,
|
|
|
|
struct ib_recv_wr *recv_wr,
|
|
|
|
struct ib_recv_wr **bad_recv_wr)
|
|
|
|
{
|
|
|
|
return srq->device->post_srq_recv(srq, recv_wr, bad_recv_wr);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_create_qp - Creates a QP associated with the specified protection
|
|
|
|
* domain.
|
|
|
|
* @pd: The protection domain associated with the QP.
|
2006-02-23 13:13:51 -07:00
|
|
|
* @qp_init_attr: A list of initial attributes required to create the
|
|
|
|
* QP. If QP creation succeeds, then the attributes are updated to
|
|
|
|
* the actual capabilities of the created QP.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
struct ib_qp *ib_create_qp(struct ib_pd *pd,
|
|
|
|
struct ib_qp_init_attr *qp_init_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_modify_qp - Modifies the attributes for the specified QP and then
|
|
|
|
* transitions the QP to the given state.
|
|
|
|
* @qp: The QP to modify.
|
|
|
|
* @qp_attr: On input, specifies the QP attributes to modify. On output,
|
|
|
|
* the current values of selected QP attributes are returned.
|
|
|
|
* @qp_attr_mask: A bit-mask used to specify which attributes of the QP
|
|
|
|
* are being modified.
|
|
|
|
*/
|
|
|
|
int ib_modify_qp(struct ib_qp *qp,
|
|
|
|
struct ib_qp_attr *qp_attr,
|
|
|
|
int qp_attr_mask);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_query_qp - Returns the attribute list and current values for the
|
|
|
|
* specified QP.
|
|
|
|
* @qp: The QP to query.
|
|
|
|
* @qp_attr: The attributes of the specified QP.
|
|
|
|
* @qp_attr_mask: A bit-mask used to select specific attributes to query.
|
|
|
|
* @qp_init_attr: Additional attributes of the selected QP.
|
|
|
|
*
|
|
|
|
* The qp_attr_mask may be used to limit the query to gathering only the
|
|
|
|
* selected attributes.
|
|
|
|
*/
|
|
|
|
int ib_query_qp(struct ib_qp *qp,
|
|
|
|
struct ib_qp_attr *qp_attr,
|
|
|
|
int qp_attr_mask,
|
|
|
|
struct ib_qp_init_attr *qp_init_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_destroy_qp - Destroys the specified QP.
|
|
|
|
* @qp: The QP to destroy.
|
|
|
|
*/
|
|
|
|
int ib_destroy_qp(struct ib_qp *qp);
|
|
|
|
|
2011-05-27 00:06:44 -06:00
|
|
|
/**
|
2011-08-08 16:31:51 -06:00
|
|
|
* ib_open_qp - Obtain a reference to an existing sharable QP.
|
|
|
|
* @xrcd - XRC domain
|
|
|
|
* @qp_open_attr: Attributes identifying the QP to open.
|
|
|
|
*
|
|
|
|
* Returns a reference to a sharable QP.
|
|
|
|
*/
|
|
|
|
struct ib_qp *ib_open_qp(struct ib_xrcd *xrcd,
|
|
|
|
struct ib_qp_open_attr *qp_open_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_close_qp - Release an external reference to a QP.
|
2011-05-27 00:06:44 -06:00
|
|
|
* @qp: The QP handle to release
|
|
|
|
*
|
2011-08-08 16:31:51 -06:00
|
|
|
* The opened QP handle is released by the caller. The underlying
|
|
|
|
* shared QP is not destroyed until all internal references are released.
|
2011-05-27 00:06:44 -06:00
|
|
|
*/
|
2011-08-08 16:31:51 -06:00
|
|
|
int ib_close_qp(struct ib_qp *qp);
|
2011-05-27 00:06:44 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_post_send - Posts a list of work requests to the send queue of
|
|
|
|
* the specified QP.
|
|
|
|
* @qp: The QP to post the work request on.
|
|
|
|
* @send_wr: A list of work requests to post on the send queue.
|
|
|
|
* @bad_send_wr: On an immediate failure, this parameter will reference
|
|
|
|
* the work request that failed to be posted on the QP.
|
2009-12-09 15:20:04 -07:00
|
|
|
*
|
|
|
|
* While IBA Vol. 1 section 11.4.1.1 specifies that if an immediate
|
|
|
|
* error is returned, the QP state shall not be affected,
|
|
|
|
* ib_post_send() will return an immediate error after queueing any
|
|
|
|
* earlier work requests in the list.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
static inline int ib_post_send(struct ib_qp *qp,
|
|
|
|
struct ib_send_wr *send_wr,
|
|
|
|
struct ib_send_wr **bad_send_wr)
|
|
|
|
{
|
|
|
|
return qp->device->post_send(qp, send_wr, bad_send_wr);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_post_recv - Posts a list of work requests to the receive queue of
|
|
|
|
* the specified QP.
|
|
|
|
* @qp: The QP to post the work request on.
|
|
|
|
* @recv_wr: A list of work requests to post on the receive queue.
|
|
|
|
* @bad_recv_wr: On an immediate failure, this parameter will reference
|
|
|
|
* the work request that failed to be posted on the QP.
|
|
|
|
*/
|
|
|
|
static inline int ib_post_recv(struct ib_qp *qp,
|
|
|
|
struct ib_recv_wr *recv_wr,
|
|
|
|
struct ib_recv_wr **bad_recv_wr)
|
|
|
|
{
|
|
|
|
return qp->device->post_recv(qp, recv_wr, bad_recv_wr);
|
|
|
|
}
|
|
|
|
|
2015-12-11 12:53:03 -07:00
|
|
|
struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private,
|
|
|
|
int nr_cqe, int comp_vector, enum ib_poll_context poll_ctx);
|
|
|
|
void ib_free_cq(struct ib_cq *cq);
|
|
|
|
int ib_process_cq_direct(struct ib_cq *cq, int budget);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_create_cq - Creates a CQ on the specified device.
|
|
|
|
* @device: The device on which to create the CQ.
|
|
|
|
* @comp_handler: A user-specified callback that is invoked when a
|
|
|
|
* completion event occurs on the CQ.
|
|
|
|
* @event_handler: A user-specified callback that is invoked when an
|
|
|
|
* asynchronous event not associated with a completion occurs on the CQ.
|
|
|
|
* @cq_context: Context associated with the CQ returned to the user via
|
|
|
|
* the associated completion and event handlers.
|
2015-06-11 07:35:21 -06:00
|
|
|
* @cq_attr: The attributes the CQ should be created upon.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
|
|
|
* Users can examine the cq structure to determine the actual CQ size.
|
|
|
|
*/
|
|
|
|
struct ib_cq *ib_create_cq(struct ib_device *device,
|
|
|
|
ib_comp_handler comp_handler,
|
|
|
|
void (*event_handler)(struct ib_event *, void *),
|
2015-06-11 07:35:21 -06:00
|
|
|
void *cq_context,
|
|
|
|
const struct ib_cq_init_attr *cq_attr);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_resize_cq - Modifies the capacity of the CQ.
|
|
|
|
* @cq: The CQ to resize.
|
|
|
|
* @cqe: The minimum size of the CQ.
|
|
|
|
*
|
|
|
|
* Users can examine the cq structure to determine the actual CQ size.
|
|
|
|
*/
|
|
|
|
int ib_resize_cq(struct ib_cq *cq, int cqe);
|
|
|
|
|
2008-04-16 22:09:33 -06:00
|
|
|
/**
|
|
|
|
* ib_modify_cq - Modifies moderation params of the CQ
|
|
|
|
* @cq: The CQ to modify.
|
|
|
|
* @cq_count: number of CQEs that will trigger an event
|
|
|
|
* @cq_period: max period of time in usec before triggering an event
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
int ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_destroy_cq - Destroys the specified CQ.
|
|
|
|
* @cq: The CQ to destroy.
|
|
|
|
*/
|
|
|
|
int ib_destroy_cq(struct ib_cq *cq);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_poll_cq - poll a CQ for completion(s)
|
|
|
|
* @cq:the CQ being polled
|
|
|
|
* @num_entries:maximum number of completions to return
|
|
|
|
* @wc:array of at least @num_entries &struct ib_wc where completions
|
|
|
|
* will be returned
|
|
|
|
*
|
|
|
|
* Poll a CQ for (possibly multiple) completions. If the return value
|
|
|
|
* is < 0, an error occurred. If the return value is >= 0, it is the
|
|
|
|
* number of completions returned. If the return value is
|
|
|
|
* non-negative and < num_entries, then the CQ was emptied.
|
|
|
|
*/
|
|
|
|
static inline int ib_poll_cq(struct ib_cq *cq, int num_entries,
|
|
|
|
struct ib_wc *wc)
|
|
|
|
{
|
|
|
|
return cq->device->poll_cq(cq, num_entries, wc);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_peek_cq - Returns the number of unreaped completions currently
|
|
|
|
* on the specified CQ.
|
|
|
|
* @cq: The CQ to peek.
|
|
|
|
* @wc_cnt: A minimum number of unreaped completions to check for.
|
|
|
|
*
|
|
|
|
* If the number of unreaped completions is greater than or equal to wc_cnt,
|
|
|
|
* this function returns wc_cnt, otherwise, it returns the actual number of
|
|
|
|
* unreaped completions.
|
|
|
|
*/
|
|
|
|
int ib_peek_cq(struct ib_cq *cq, int wc_cnt);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_req_notify_cq - Request completion notification on a CQ.
|
|
|
|
* @cq: The CQ to generate an event for.
|
IB: Return "maybe missed event" hint from ib_req_notify_cq()
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-06 22:02:48 -06:00
|
|
|
* @flags:
|
|
|
|
* Must contain exactly one of %IB_CQ_SOLICITED or %IB_CQ_NEXT_COMP
|
|
|
|
* to request an event on the next solicited event or next work
|
|
|
|
* completion at any type, respectively. %IB_CQ_REPORT_MISSED_EVENTS
|
|
|
|
* may also be |ed in to request a hint about missed events, as
|
|
|
|
* described below.
|
|
|
|
*
|
|
|
|
* Return Value:
|
|
|
|
* < 0 means an error occurred while requesting notification
|
|
|
|
* == 0 means notification was requested successfully, and if
|
|
|
|
* IB_CQ_REPORT_MISSED_EVENTS was passed in, then no events
|
|
|
|
* were missed and it is safe to wait for another event. In
|
|
|
|
* this case is it guaranteed that any work completions added
|
|
|
|
* to the CQ since the last CQ poll will trigger a completion
|
|
|
|
* notification event.
|
|
|
|
* > 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was passed
|
|
|
|
* in. It means that the consumer must poll the CQ again to
|
|
|
|
* make sure it is empty to avoid missing an event because of a
|
|
|
|
* race between requesting notification and an entry being
|
|
|
|
* added to the CQ. This return value means it is possible
|
|
|
|
* (but not guaranteed) that a work completion has been added
|
|
|
|
* to the CQ since the last poll without triggering a
|
|
|
|
* completion notification event.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
static inline int ib_req_notify_cq(struct ib_cq *cq,
|
IB: Return "maybe missed event" hint from ib_req_notify_cq()
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-06 22:02:48 -06:00
|
|
|
enum ib_cq_notify_flags flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
IB: Return "maybe missed event" hint from ib_req_notify_cq()
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-06 22:02:48 -06:00
|
|
|
return cq->device->req_notify_cq(cq, flags);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_req_ncomp_notif - Request completion notification when there are
|
|
|
|
* at least the specified number of unreaped completions on the CQ.
|
|
|
|
* @cq: The CQ to generate an event for.
|
|
|
|
* @wc_cnt: The number of unreaped completions that should be on the
|
|
|
|
* CQ before an event is generated.
|
|
|
|
*/
|
|
|
|
static inline int ib_req_ncomp_notif(struct ib_cq *cq, int wc_cnt)
|
|
|
|
{
|
|
|
|
return cq->device->req_ncomp_notif ?
|
|
|
|
cq->device->req_ncomp_notif(cq, wc_cnt) :
|
|
|
|
-ENOSYS;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_get_dma_mr - Returns a memory region for system memory that is
|
|
|
|
* usable for DMA.
|
|
|
|
* @pd: The protection domain associated with the memory region.
|
|
|
|
* @mr_access_flags: Specifies the memory access rights.
|
2006-12-12 15:27:41 -07:00
|
|
|
*
|
|
|
|
* Note that the ib_dma_*() functions defined below must be used
|
|
|
|
* to create/destroy addresses used with the Lkey or Rkey returned
|
|
|
|
* by ib_get_dma_mr().
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
struct ib_mr *ib_get_dma_mr(struct ib_pd *pd, int mr_access_flags);
|
|
|
|
|
2006-12-12 15:27:41 -07:00
|
|
|
/**
|
|
|
|
* ib_dma_mapping_error - check a DMA addr for error
|
|
|
|
* @dev: The device for which the dma_addr was created
|
|
|
|
* @dma_addr: The DMA address to check
|
|
|
|
*/
|
|
|
|
static inline int ib_dma_mapping_error(struct ib_device *dev, u64 dma_addr)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
return dev->dma_ops->mapping_error(dev, dma_addr);
|
2008-07-25 20:44:49 -06:00
|
|
|
return dma_mapping_error(dev->dma_device, dma_addr);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_map_single - Map a kernel virtual address to DMA address
|
|
|
|
* @dev: The device for which the dma_addr is to be created
|
|
|
|
* @cpu_addr: The kernel virtual address
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline u64 ib_dma_map_single(struct ib_device *dev,
|
|
|
|
void *cpu_addr, size_t size,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
return dev->dma_ops->map_single(dev, cpu_addr, size, direction);
|
|
|
|
return dma_map_single(dev->dma_device, cpu_addr, size, direction);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_unmap_single - Destroy a mapping created by ib_dma_map_single()
|
|
|
|
* @dev: The device for which the DMA address was created
|
|
|
|
* @addr: The DMA address
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_unmap_single(struct ib_device *dev,
|
|
|
|
u64 addr, size_t size,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->unmap_single(dev, addr, size, direction);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_unmap_single(dev->dma_device, addr, size, direction);
|
|
|
|
}
|
|
|
|
|
2008-04-29 02:00:34 -06:00
|
|
|
static inline u64 ib_dma_map_single_attrs(struct ib_device *dev,
|
|
|
|
void *cpu_addr, size_t size,
|
|
|
|
enum dma_data_direction direction,
|
|
|
|
struct dma_attrs *attrs)
|
|
|
|
{
|
|
|
|
return dma_map_single_attrs(dev->dma_device, cpu_addr, size,
|
|
|
|
direction, attrs);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ib_dma_unmap_single_attrs(struct ib_device *dev,
|
|
|
|
u64 addr, size_t size,
|
|
|
|
enum dma_data_direction direction,
|
|
|
|
struct dma_attrs *attrs)
|
|
|
|
{
|
|
|
|
return dma_unmap_single_attrs(dev->dma_device, addr, size,
|
|
|
|
direction, attrs);
|
|
|
|
}
|
|
|
|
|
2006-12-12 15:27:41 -07:00
|
|
|
/**
|
|
|
|
* ib_dma_map_page - Map a physical page to DMA address
|
|
|
|
* @dev: The device for which the dma_addr is to be created
|
|
|
|
* @page: The page to be mapped
|
|
|
|
* @offset: The offset within the page
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline u64 ib_dma_map_page(struct ib_device *dev,
|
|
|
|
struct page *page,
|
|
|
|
unsigned long offset,
|
|
|
|
size_t size,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
return dev->dma_ops->map_page(dev, page, offset, size, direction);
|
|
|
|
return dma_map_page(dev->dma_device, page, offset, size, direction);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_unmap_page - Destroy a mapping created by ib_dma_map_page()
|
|
|
|
* @dev: The device for which the DMA address was created
|
|
|
|
* @addr: The DMA address
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_unmap_page(struct ib_device *dev,
|
|
|
|
u64 addr, size_t size,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->unmap_page(dev, addr, size, direction);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_unmap_page(dev->dma_device, addr, size, direction);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_map_sg - Map a scatter/gather list to DMA addresses
|
|
|
|
* @dev: The device for which the DMA addresses are to be created
|
|
|
|
* @sg: The array of scatter/gather entries
|
|
|
|
* @nents: The number of scatter/gather entries
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline int ib_dma_map_sg(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
return dev->dma_ops->map_sg(dev, sg, nents, direction);
|
|
|
|
return dma_map_sg(dev->dma_device, sg, nents, direction);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_unmap_sg - Unmap a scatter/gather list of DMA addresses
|
|
|
|
* @dev: The device for which the DMA addresses were created
|
|
|
|
* @sg: The array of scatter/gather entries
|
|
|
|
* @nents: The number of scatter/gather entries
|
|
|
|
* @direction: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_unmap_sg(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->unmap_sg(dev, sg, nents, direction);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_unmap_sg(dev->dma_device, sg, nents, direction);
|
|
|
|
}
|
|
|
|
|
2008-04-29 02:00:34 -06:00
|
|
|
static inline int ib_dma_map_sg_attrs(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction,
|
|
|
|
struct dma_attrs *attrs)
|
|
|
|
{
|
|
|
|
return dma_map_sg_attrs(dev->dma_device, sg, nents, direction, attrs);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ib_dma_unmap_sg_attrs(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction direction,
|
|
|
|
struct dma_attrs *attrs)
|
|
|
|
{
|
|
|
|
dma_unmap_sg_attrs(dev->dma_device, sg, nents, direction, attrs);
|
|
|
|
}
|
2006-12-12 15:27:41 -07:00
|
|
|
/**
|
|
|
|
* ib_sg_dma_address - Return the DMA address from a scatter/gather entry
|
|
|
|
* @dev: The device for which the DMA addresses were created
|
|
|
|
* @sg: The scatter/gather entry
|
2014-03-28 11:26:59 -06:00
|
|
|
*
|
|
|
|
* Note: this function is obsolete. To do: change all occurrences of
|
|
|
|
* ib_sg_dma_address() into sg_dma_address().
|
2006-12-12 15:27:41 -07:00
|
|
|
*/
|
|
|
|
static inline u64 ib_sg_dma_address(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
return sg_dma_address(sg);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_sg_dma_len - Return the DMA length from a scatter/gather entry
|
|
|
|
* @dev: The device for which the DMA addresses were created
|
|
|
|
* @sg: The scatter/gather entry
|
2014-03-28 11:26:59 -06:00
|
|
|
*
|
|
|
|
* Note: this function is obsolete. To do: change all occurrences of
|
|
|
|
* ib_sg_dma_len() into sg_dma_len().
|
2006-12-12 15:27:41 -07:00
|
|
|
*/
|
|
|
|
static inline unsigned int ib_sg_dma_len(struct ib_device *dev,
|
|
|
|
struct scatterlist *sg)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
return sg_dma_len(sg);
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_sync_single_for_cpu - Prepare DMA region to be accessed by CPU
|
|
|
|
* @dev: The device for which the DMA address was created
|
|
|
|
* @addr: The DMA address
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @dir: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_sync_single_for_cpu(struct ib_device *dev,
|
|
|
|
u64 addr,
|
|
|
|
size_t size,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->sync_single_for_cpu(dev, addr, size, dir);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_sync_single_for_cpu(dev->dma_device, addr, size, dir);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_sync_single_for_device - Prepare DMA region to be accessed by device
|
|
|
|
* @dev: The device for which the DMA address was created
|
|
|
|
* @addr: The DMA address
|
|
|
|
* @size: The size of the region in bytes
|
|
|
|
* @dir: The direction of the DMA
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_sync_single_for_device(struct ib_device *dev,
|
|
|
|
u64 addr,
|
|
|
|
size_t size,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->sync_single_for_device(dev, addr, size, dir);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_sync_single_for_device(dev->dma_device, addr, size, dir);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_alloc_coherent - Allocate memory and map it for DMA
|
|
|
|
* @dev: The device for which the DMA address is requested
|
|
|
|
* @size: The size of the region to allocate in bytes
|
|
|
|
* @dma_handle: A pointer for returning the DMA address of the region
|
|
|
|
* @flag: memory allocator flags
|
|
|
|
*/
|
|
|
|
static inline void *ib_dma_alloc_coherent(struct ib_device *dev,
|
|
|
|
size_t size,
|
|
|
|
u64 *dma_handle,
|
|
|
|
gfp_t flag)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
return dev->dma_ops->alloc_coherent(dev, size, dma_handle, flag);
|
2006-12-15 14:57:26 -07:00
|
|
|
else {
|
|
|
|
dma_addr_t handle;
|
|
|
|
void *ret;
|
|
|
|
|
|
|
|
ret = dma_alloc_coherent(dev->dma_device, size, &handle, flag);
|
|
|
|
*dma_handle = handle;
|
|
|
|
return ret;
|
|
|
|
}
|
2006-12-12 15:27:41 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dma_free_coherent - Free memory allocated by ib_dma_alloc_coherent()
|
|
|
|
* @dev: The device for which the DMA addresses were allocated
|
|
|
|
* @size: The size of the region
|
|
|
|
* @cpu_addr: the address returned by ib_dma_alloc_coherent()
|
|
|
|
* @dma_handle: the DMA address returned by ib_dma_alloc_coherent()
|
|
|
|
*/
|
|
|
|
static inline void ib_dma_free_coherent(struct ib_device *dev,
|
|
|
|
size_t size, void *cpu_addr,
|
|
|
|
u64 dma_handle)
|
|
|
|
{
|
2006-12-13 20:10:05 -07:00
|
|
|
if (dev->dma_ops)
|
|
|
|
dev->dma_ops->free_coherent(dev, size, cpu_addr, dma_handle);
|
|
|
|
else
|
2006-12-12 15:27:41 -07:00
|
|
|
dma_free_coherent(dev->dma_device, size, cpu_addr, dma_handle);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_dereg_mr - Deregisters a memory region and removes it from the
|
|
|
|
* HCA translation table.
|
|
|
|
* @mr: The memory region to deregister.
|
2013-02-06 09:19:12 -07:00
|
|
|
*
|
|
|
|
* This function can fail, if the memory region has memory windows bound to it.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
int ib_dereg_mr(struct ib_mr *mr);
|
|
|
|
|
2015-07-30 01:32:35 -06:00
|
|
|
struct ib_mr *ib_alloc_mr(struct ib_pd *pd,
|
|
|
|
enum ib_mr_type mr_type,
|
|
|
|
u32 max_num_sg);
|
2008-07-15 00:48:45 -06:00
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_update_fast_reg_key - updates the key portion of the fast_reg MR
|
|
|
|
* R_Key and L_Key.
|
|
|
|
* @mr - struct ib_mr pointer to be updated.
|
|
|
|
* @newkey - new key to be used.
|
|
|
|
*/
|
|
|
|
static inline void ib_update_fast_reg_key(struct ib_mr *mr, u8 newkey)
|
|
|
|
{
|
|
|
|
mr->lkey = (mr->lkey & 0xffffff00) | newkey;
|
|
|
|
mr->rkey = (mr->rkey & 0xffffff00) | newkey;
|
|
|
|
}
|
|
|
|
|
2013-02-06 09:19:12 -07:00
|
|
|
/**
|
|
|
|
* ib_inc_rkey - increments the key portion of the given rkey. Can be used
|
|
|
|
* for calculating a new rkey for type 2 memory windows.
|
|
|
|
* @rkey - the rkey to increment.
|
|
|
|
*/
|
|
|
|
static inline u32 ib_inc_rkey(u32 rkey)
|
|
|
|
{
|
|
|
|
const u32 mask = 0x000000ff;
|
|
|
|
return ((rkey + 1) & mask) | (rkey & ~mask);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/**
|
|
|
|
* ib_alloc_fmr - Allocates a unmapped fast memory region.
|
|
|
|
* @pd: The protection domain associated with the unmapped region.
|
|
|
|
* @mr_access_flags: Specifies the memory access rights.
|
|
|
|
* @fmr_attr: Attributes of the unmapped region.
|
|
|
|
*
|
|
|
|
* A fast memory region must be mapped before it can be used as part of
|
|
|
|
* a work request.
|
|
|
|
*/
|
|
|
|
struct ib_fmr *ib_alloc_fmr(struct ib_pd *pd,
|
|
|
|
int mr_access_flags,
|
|
|
|
struct ib_fmr_attr *fmr_attr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_map_phys_fmr - Maps a list of physical pages to a fast memory region.
|
|
|
|
* @fmr: The fast memory region to associate with the pages.
|
|
|
|
* @page_list: An array of physical pages to map to the fast memory region.
|
|
|
|
* @list_len: The number of pages in page_list.
|
|
|
|
* @iova: The I/O virtual address to use with the mapped region.
|
|
|
|
*/
|
|
|
|
static inline int ib_map_phys_fmr(struct ib_fmr *fmr,
|
|
|
|
u64 *page_list, int list_len,
|
|
|
|
u64 iova)
|
|
|
|
{
|
|
|
|
return fmr->device->map_phys_fmr(fmr, page_list, list_len, iova);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_unmap_fmr - Removes the mapping from a list of fast memory regions.
|
|
|
|
* @fmr_list: A linked list of fast memory regions to unmap.
|
|
|
|
*/
|
|
|
|
int ib_unmap_fmr(struct list_head *fmr_list);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dealloc_fmr - Deallocates a fast memory region.
|
|
|
|
* @fmr: The fast memory region to deallocate.
|
|
|
|
*/
|
|
|
|
int ib_dealloc_fmr(struct ib_fmr *fmr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_attach_mcast - Attaches the specified QP to a multicast group.
|
|
|
|
* @qp: QP to attach to the multicast group. The QP must be type
|
|
|
|
* IB_QPT_UD.
|
|
|
|
* @gid: Multicast group GID.
|
|
|
|
* @lid: Multicast group LID in host byte order.
|
|
|
|
*
|
|
|
|
* In order to send and receive multicast packets, subnet
|
|
|
|
* administration must have created the multicast group and configured
|
|
|
|
* the fabric appropriately. The port associated with the specified
|
|
|
|
* QP must also be a member of the multicast group.
|
|
|
|
*/
|
|
|
|
int ib_attach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_detach_mcast - Detaches the specified QP from a multicast group.
|
|
|
|
* @qp: QP to detach from the multicast group.
|
|
|
|
* @gid: Multicast group GID.
|
|
|
|
* @lid: Multicast group LID in host byte order.
|
|
|
|
*/
|
|
|
|
int ib_detach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid);
|
|
|
|
|
2011-05-23 18:52:46 -06:00
|
|
|
/**
|
|
|
|
* ib_alloc_xrcd - Allocates an XRC domain.
|
|
|
|
* @device: The device on which to allocate the XRC domain.
|
|
|
|
*/
|
|
|
|
struct ib_xrcd *ib_alloc_xrcd(struct ib_device *device);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ib_dealloc_xrcd - Deallocates an XRC domain.
|
|
|
|
* @xrcd: The XRC domain to deallocate.
|
|
|
|
*/
|
|
|
|
int ib_dealloc_xrcd(struct ib_xrcd *xrcd);
|
|
|
|
|
IB/core: Add receive flow steering support
The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side. Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.
This patch adds ib_create_flow(), which allow providing a flow
specification for a QP. When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.
Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers. Currently specs for Ethernet, IPv4,
TCP and UDP are defined. Flow specs are made of values and masks.
The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.
struct ib_flow_attr {
enum ib_flow_attr_type type;
u16 size;
u16 priority;
u32 flags;
u8 num_of_specs;
u8 port;
/* Following are the optional layers according to user request
* struct ib_flow_spec_yyy
* struct ib_flow_spec_zzz
*/
};
As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.
The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.
For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.
Steering rules processing order is according to the domain over which
the rule is set and the rule priority. All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation. Lower numerical value for the priority field means
higher priority.
The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().
Applications that offload TCP/IP traffic can also be written over IB
UD QPs. The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too. A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.
The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:
IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification
IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
all Ethernet traffic which isn't steered to any QP
IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast
IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic
ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-07 05:01:59 -06:00
|
|
|
struct ib_flow *ib_create_flow(struct ib_qp *qp,
|
|
|
|
struct ib_flow_attr *flow_attr, int domain);
|
|
|
|
int ib_destroy_flow(struct ib_flow *flow_id);
|
|
|
|
|
2013-10-31 07:26:32 -06:00
|
|
|
static inline int ib_check_mr_access(int flags)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Local write permission is required if remote write or
|
|
|
|
* remote atomic permission is also requested.
|
|
|
|
*/
|
|
|
|
if (flags & (IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_REMOTE_WRITE) &&
|
|
|
|
!(flags & IB_ACCESS_LOCAL_WRITE))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-02-23 05:19:05 -07:00
|
|
|
/**
|
|
|
|
* ib_check_mr_status: lightweight check of MR status.
|
|
|
|
* This routine may provide status checks on a selected
|
|
|
|
* ib_mr. first use is for signature status check.
|
|
|
|
*
|
|
|
|
* @mr: A memory region.
|
|
|
|
* @check_mask: Bitmask of which checks to perform from
|
|
|
|
* ib_mr_status_check enumeration.
|
|
|
|
* @mr_status: The container of relevant status checks.
|
|
|
|
* failed checks will be indicated in the status bitmask
|
|
|
|
* and the relevant info shall be in the error item.
|
|
|
|
*/
|
|
|
|
int ib_check_mr_status(struct ib_mr *mr, u32 check_mask,
|
|
|
|
struct ib_mr_status *mr_status);
|
|
|
|
|
2015-07-30 08:50:15 -06:00
|
|
|
struct net_device *ib_get_net_dev_by_params(struct ib_device *dev, u8 port,
|
|
|
|
u16 pkey, const union ib_gid *gid,
|
|
|
|
const struct sockaddr *addr);
|
|
|
|
|
2015-10-13 10:11:24 -06:00
|
|
|
int ib_map_mr_sg(struct ib_mr *mr,
|
|
|
|
struct scatterlist *sg,
|
|
|
|
int sg_nents,
|
|
|
|
unsigned int page_size);
|
|
|
|
|
|
|
|
static inline int
|
|
|
|
ib_map_mr_sg_zbva(struct ib_mr *mr,
|
|
|
|
struct scatterlist *sg,
|
|
|
|
int sg_nents,
|
|
|
|
unsigned int page_size)
|
|
|
|
{
|
|
|
|
int n;
|
|
|
|
|
|
|
|
n = ib_map_mr_sg(mr, sg, sg_nents, page_size);
|
|
|
|
mr->iova = 0;
|
|
|
|
|
|
|
|
return n;
|
|
|
|
}
|
|
|
|
|
|
|
|
int ib_sg_to_pages(struct ib_mr *mr,
|
|
|
|
struct scatterlist *sgl,
|
|
|
|
int sg_nents,
|
|
|
|
int (*set_page)(struct ib_mr *, u64));
|
|
|
|
|
2016-02-17 09:15:41 -07:00
|
|
|
void ib_drain_rq(struct ib_qp *qp);
|
|
|
|
void ib_drain_sq(struct ib_qp *qp);
|
|
|
|
void ib_drain_qp(struct ib_qp *qp);
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif /* IB_VERBS_H */
|