2008-12-10 07:10:24 +08:00
|
|
|
/*
|
2009-03-18 02:42:40 +08:00
|
|
|
* Copyright (c) 2008-2009 Cisco Systems, Inc. All rights reserved.
|
|
|
|
|
* Copyright (c) 2007-2008 Intel Corporation. All rights reserved.
|
2008-12-10 07:10:24 +08:00
|
|
|
*
|
|
|
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
|
|
|
* under the terms and conditions of the GNU General Public License,
|
|
|
|
|
* version 2, as published by the Free Software Foundation.
|
|
|
|
|
*
|
|
|
|
|
* This program is distributed in the hope it will be useful, but WITHOUT
|
|
|
|
|
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
|
|
|
|
|
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
|
|
|
|
|
* more details.
|
|
|
|
|
*
|
|
|
|
|
* You should have received a copy of the GNU General Public License along with
|
|
|
|
|
* this program; if not, write to the Free Software Foundation, Inc.,
|
|
|
|
|
* 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
|
|
|
|
|
*
|
|
|
|
|
* Maintained at www.Open-FCoE.org
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#ifndef _LIBFCOE_H
|
|
|
|
|
#define _LIBFCOE_H
|
|
|
|
|
|
2009-03-18 02:42:40 +08:00
|
|
|
#include <linux/etherdevice.h>
|
|
|
|
|
#include <linux/if_ether.h>
|
2008-12-10 07:10:24 +08:00
|
|
|
#include <linux/netdevice.h>
|
|
|
|
|
#include <linux/skbuff.h>
|
2009-03-18 02:42:40 +08:00
|
|
|
#include <linux/workqueue.h>
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
#include <linux/random.h>
|
2008-12-10 07:10:24 +08:00
|
|
|
#include <scsi/fc/fc_fcoe.h>
|
|
|
|
|
#include <scsi/libfc.h>
|
|
|
|
|
|
2010-04-10 05:22:59 +08:00
|
|
|
#define FCOE_MAX_CMD_LEN 16 /* Supported CDB length */
|
|
|
|
|
|
2009-03-18 02:42:40 +08:00
|
|
|
/*
|
|
|
|
|
* FIP tunable parameters.
|
|
|
|
|
*/
|
|
|
|
|
#define FCOE_CTLR_START_DELAY 2000 /* mS after first adv. to choose FCF */
|
|
|
|
|
#define FCOE_CTRL_SOL_TOV 2000 /* min. solicitation interval (mS) */
|
|
|
|
|
#define FCOE_CTLR_FCF_LIMIT 20 /* max. number of FCF entries */
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
#define FCOE_CTLR_VN2VN_LOGIN_LIMIT 3 /* max. VN2VN rport login retries */
|
2009-03-18 02:42:40 +08:00
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* enum fip_state - internal state of FCoE controller.
|
|
|
|
|
* @FIP_ST_DISABLED: controller has been disabled or not yet enabled.
|
|
|
|
|
* @FIP_ST_LINK_WAIT: the physical link is down or unusable.
|
|
|
|
|
* @FIP_ST_AUTO: determining whether to use FIP or non-FIP mode.
|
|
|
|
|
* @FIP_ST_NON_FIP: non-FIP mode selected.
|
|
|
|
|
* @FIP_ST_ENABLED: FIP mode selected.
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
* @FIP_ST_VNMP_START: VN2VN multipath mode start, wait
|
|
|
|
|
* @FIP_ST_VNMP_PROBE1: VN2VN sent first probe, listening
|
|
|
|
|
* @FIP_ST_VNMP_PROBE2: VN2VN sent second probe, listening
|
|
|
|
|
* @FIP_ST_VNMP_CLAIM: VN2VN sent claim, waiting for responses
|
|
|
|
|
* @FIP_ST_VNMP_UP: VN2VN multipath mode operation
|
2009-03-18 02:42:40 +08:00
|
|
|
*/
|
|
|
|
|
enum fip_state {
|
|
|
|
|
FIP_ST_DISABLED,
|
|
|
|
|
FIP_ST_LINK_WAIT,
|
|
|
|
|
FIP_ST_AUTO,
|
|
|
|
|
FIP_ST_NON_FIP,
|
|
|
|
|
FIP_ST_ENABLED,
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
FIP_ST_VNMP_START,
|
|
|
|
|
FIP_ST_VNMP_PROBE1,
|
|
|
|
|
FIP_ST_VNMP_PROBE2,
|
|
|
|
|
FIP_ST_VNMP_CLAIM,
|
|
|
|
|
FIP_ST_VNMP_UP,
|
2009-03-18 02:42:40 +08:00
|
|
|
};
|
|
|
|
|
|
2010-07-21 06:19:58 +08:00
|
|
|
/*
|
|
|
|
|
* Modes:
|
|
|
|
|
* The mode is the state that is to be entered after link up.
|
|
|
|
|
* It must not change after fcoe_ctlr_init() sets it.
|
|
|
|
|
*/
|
|
|
|
|
#define FIP_MODE_AUTO FIP_ST_AUTO
|
|
|
|
|
#define FIP_MODE_NON_FIP FIP_ST_NON_FIP
|
|
|
|
|
#define FIP_MODE_FABRIC FIP_ST_ENABLED
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
#define FIP_MODE_VN2VN FIP_ST_VNMP_START
|
2010-07-21 06:19:58 +08:00
|
|
|
|
2009-03-18 02:42:40 +08:00
|
|
|
/**
|
2009-11-04 03:47:45 +08:00
|
|
|
* struct fcoe_ctlr - FCoE Controller and FIP state
|
|
|
|
|
* @state: internal FIP state for network link and FIP or non-FIP mode.
|
2009-11-04 03:48:11 +08:00
|
|
|
* @mode: LLD-selected mode.
|
2009-11-04 03:47:45 +08:00
|
|
|
* @lp: &fc_lport: libfc local port.
|
|
|
|
|
* @sel_fcf: currently selected FCF, or NULL.
|
|
|
|
|
* @fcfs: list of discovered FCFs.
|
|
|
|
|
* @fcf_count: number of discovered FCF entries.
|
|
|
|
|
* @sol_time: time when a multicast solicitation was last sent.
|
|
|
|
|
* @sel_time: time after which to select an FCF.
|
|
|
|
|
* @port_ka_time: time of next port keep-alive.
|
|
|
|
|
* @ctlr_ka_time: time of next controller keep-alive.
|
|
|
|
|
* @timer: timer struct used for all delayed events.
|
2010-03-13 08:08:23 +08:00
|
|
|
* @timer_work: &work_struct for doing keep-alives and resets.
|
2009-11-04 03:47:45 +08:00
|
|
|
* @recv_work: &work_struct for receiving FIP frames.
|
2009-03-18 02:42:40 +08:00
|
|
|
* @fip_recv_list: list of received FIP frames.
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
* @rnd_state: state for pseudo-random number generator.
|
|
|
|
|
* @port_id: proposed or selected local-port ID.
|
2009-11-04 03:47:45 +08:00
|
|
|
* @user_mfs: configured maximum FC frame size, including FC header.
|
|
|
|
|
* @flogi_oxid: exchange ID of most recent fabric login.
|
|
|
|
|
* @flogi_count: number of FLOGI attempts in AUTO mode.
|
|
|
|
|
* @map_dest: use the FC_MAP mode for destination MAC addresses.
|
|
|
|
|
* @spma: supports SPMA server-provided MACs mode
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
* @probe_tries: number of FC_IDs probed
|
2009-11-04 03:47:45 +08:00
|
|
|
* @dest_addr: MAC address of the selected FC forwarder.
|
|
|
|
|
* @ctl_src_addr: the native MAC address of our local port.
|
|
|
|
|
* @send: LLD-supplied function to handle sending FIP Ethernet frames
|
|
|
|
|
* @update_mac: LLD-supplied function to handle changes to MAC addresses.
|
|
|
|
|
* @get_src_addr: LLD-supplied function to supply a source MAC address.
|
2010-07-21 06:19:47 +08:00
|
|
|
* @ctlr_mutex: lock protecting this structure.
|
2009-03-18 02:42:40 +08:00
|
|
|
*
|
|
|
|
|
* This structure is used by all FCoE drivers. It contains information
|
|
|
|
|
* needed by all FCoE low-level drivers (LLDs) as well as internal state
|
|
|
|
|
* for FIP, and fields shared with the LLDS.
|
|
|
|
|
*/
|
|
|
|
|
struct fcoe_ctlr {
|
|
|
|
|
enum fip_state state;
|
2009-11-04 03:48:11 +08:00
|
|
|
enum fip_state mode;
|
2009-03-18 02:42:40 +08:00
|
|
|
struct fc_lport *lp;
|
|
|
|
|
struct fcoe_fcf *sel_fcf;
|
|
|
|
|
struct list_head fcfs;
|
|
|
|
|
u16 fcf_count;
|
|
|
|
|
unsigned long sol_time;
|
|
|
|
|
unsigned long sel_time;
|
|
|
|
|
unsigned long port_ka_time;
|
|
|
|
|
unsigned long ctlr_ka_time;
|
|
|
|
|
struct timer_list timer;
|
2010-03-13 08:08:23 +08:00
|
|
|
struct work_struct timer_work;
|
2009-03-18 02:42:40 +08:00
|
|
|
struct work_struct recv_work;
|
|
|
|
|
struct sk_buff_head fip_recv_list;
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
|
|
|
|
|
struct rnd_state rnd_state;
|
|
|
|
|
u32 port_id;
|
|
|
|
|
|
2009-03-18 02:42:40 +08:00
|
|
|
u16 user_mfs;
|
|
|
|
|
u16 flogi_oxid;
|
|
|
|
|
u8 flogi_count;
|
|
|
|
|
u8 map_dest;
|
2009-05-17 20:33:28 +08:00
|
|
|
u8 spma;
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
u8 probe_tries;
|
2009-03-18 02:42:40 +08:00
|
|
|
u8 dest_addr[ETH_ALEN];
|
|
|
|
|
u8 ctl_src_addr[ETH_ALEN];
|
|
|
|
|
|
|
|
|
|
void (*send)(struct fcoe_ctlr *, struct sk_buff *);
|
2009-11-04 03:46:29 +08:00
|
|
|
void (*update_mac)(struct fc_lport *, u8 *addr);
|
|
|
|
|
u8 * (*get_src_addr)(struct fc_lport *);
|
2010-07-21 06:19:47 +08:00
|
|
|
struct mutex ctlr_mutex;
|
2009-03-18 02:42:40 +08:00
|
|
|
};
|
|
|
|
|
|
2009-11-04 03:47:45 +08:00
|
|
|
/**
|
|
|
|
|
* struct fcoe_fcf - Fibre-Channel Forwarder
|
|
|
|
|
* @list: list linkage
|
|
|
|
|
* @time: system time (jiffies) when an advertisement was last received
|
|
|
|
|
* @switch_name: WWN of switch from advertisement
|
|
|
|
|
* @fabric_name: WWN of fabric from advertisement
|
|
|
|
|
* @fc_map: FC_MAP value from advertisement
|
|
|
|
|
* @fcf_mac: Ethernet address of the FCF
|
|
|
|
|
* @vfid: virtual fabric ID
|
|
|
|
|
* @pri: selection priority, smaller values are better
|
|
|
|
|
* @flags: flags received from advertisement
|
|
|
|
|
* @fka_period: keep-alive period, in jiffies
|
2009-03-18 02:42:40 +08:00
|
|
|
*
|
|
|
|
|
* A Fibre-Channel Forwarder (FCF) is the entity on the Ethernet that
|
|
|
|
|
* passes FCoE frames on to an FC fabric. This structure represents
|
|
|
|
|
* one FCF from which advertisements have been received.
|
|
|
|
|
*
|
|
|
|
|
* When looking up an FCF, @switch_name, @fabric_name, @fc_map, @vfid, and
|
|
|
|
|
* @fcf_mac together form the lookup key.
|
|
|
|
|
*/
|
|
|
|
|
struct fcoe_fcf {
|
|
|
|
|
struct list_head list;
|
|
|
|
|
unsigned long time;
|
|
|
|
|
|
|
|
|
|
u64 switch_name;
|
|
|
|
|
u64 fabric_name;
|
|
|
|
|
u32 fc_map;
|
|
|
|
|
u16 vfid;
|
|
|
|
|
u8 fcf_mac[ETH_ALEN];
|
|
|
|
|
|
|
|
|
|
u8 pri;
|
|
|
|
|
u16 flags;
|
|
|
|
|
u32 fka_period;
|
2009-11-21 06:54:57 +08:00
|
|
|
u8 fd_flags:1;
|
2009-03-18 02:42:40 +08:00
|
|
|
};
|
|
|
|
|
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
/**
|
|
|
|
|
* struct fcoe_rport - VN2VN remote port
|
|
|
|
|
* @time: time of create or last beacon packet received from node
|
|
|
|
|
* @fcoe_len: max FCoE frame size, not including VLAN or Ethernet headers
|
|
|
|
|
* @flags: flags from probe or claim
|
|
|
|
|
* @login_count: number of unsuccessful rport logins to this port
|
|
|
|
|
* @enode_mac: E_Node control MAC address
|
|
|
|
|
* @vn_mac: VN_Node assigned MAC address for data
|
|
|
|
|
*/
|
|
|
|
|
struct fcoe_rport {
|
|
|
|
|
unsigned long time;
|
|
|
|
|
u16 fcoe_len;
|
|
|
|
|
u16 flags;
|
|
|
|
|
u8 login_count;
|
|
|
|
|
u8 enode_mac[ETH_ALEN];
|
|
|
|
|
u8 vn_mac[ETH_ALEN];
|
|
|
|
|
};
|
|
|
|
|
|
2009-03-18 02:42:40 +08:00
|
|
|
/* FIP API functions */
|
2010-07-21 06:19:58 +08:00
|
|
|
void fcoe_ctlr_init(struct fcoe_ctlr *, enum fip_state);
|
2009-03-18 02:42:40 +08:00
|
|
|
void fcoe_ctlr_destroy(struct fcoe_ctlr *);
|
|
|
|
|
void fcoe_ctlr_link_up(struct fcoe_ctlr *);
|
|
|
|
|
int fcoe_ctlr_link_down(struct fcoe_ctlr *);
|
2009-11-04 03:46:29 +08:00
|
|
|
int fcoe_ctlr_els_send(struct fcoe_ctlr *, struct fc_lport *, struct sk_buff *);
|
2009-03-18 02:42:40 +08:00
|
|
|
void fcoe_ctlr_recv(struct fcoe_ctlr *, struct sk_buff *);
|
2009-11-04 03:47:45 +08:00
|
|
|
int fcoe_ctlr_recv_flogi(struct fcoe_ctlr *, struct fc_lport *,
|
2009-11-04 03:49:16 +08:00
|
|
|
struct fc_frame *);
|
2009-03-18 02:42:40 +08:00
|
|
|
|
2008-12-10 07:10:24 +08:00
|
|
|
/* libfcoe funcs */
|
2009-03-18 02:42:24 +08:00
|
|
|
u64 fcoe_wwn_from_mac(unsigned char mac[], unsigned int, unsigned int);
|
[SCSI] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support
The FC-BB-6 committee is proposing a new FIP usage model called
VN_port to VN_port mode. It allows VN_ports to discover each other
over a loss-free L2 Ethernet without any FCF or Fibre-channel fabric
services. This is point-to-multipoint. There is also a variant
of this called point-to-point which provides for making sure there
is just one pair of ports operating over the Ethernet fabric.
We add these new states: VNMP_START, _PROBE1, _PROBE2, _CLAIM, and _UP.
These usually go quickly in that sequence. After waiting a random
amount of time up to 100 ms in START, we select a pseudo-random
proposed locally-unique port ID and send out probes in states PROBE1
and PROBE2, 100 ms apart. If no probe responses are heard, we
proceed to CLAIM state 400 ms later and send a claim notification.
We wait another 400 ms to receive claim responses, which give us
a list of the other nodes on the network, including their FC-4
capabilities. After another 400 ms we go to VNMP_UP state and
should start interoperating with any of the nodes for whic we
receivec claim responses. More details are in the spec.j
Add the new mode as FIP_MODE_VN2VN. The driver must specify
explicitly that it wants to operate in this mode. There is
no automatic detection between point-to-multipoint and fabric
mode, and the local port initialization is affected, so it isn't
anticipated that there will ever be any such automatic switchover.
It may eventually be possible to have both fabric and VN2VN
modes on the same L2 network, which may be done by two separate
local VN_ports (lports).
When in VN2VN mode, FIP replaces libfc's fabric-oriented discovery
module with its own simple code that adds remote ports as they
are discovered from incoming claim notifications and responses.
These hooks are placed by fcoe_disc_init().
A linear list of discovered vn_ports is maintained under the
fcoe_ctlr struct. It is expected to be short for now, and
accessed infrequently. It is kept under RCU for lock-ordering
reasons. The lport and/or rport mutexes may be held when we
need to lookup a fcoe_vnport during an ELS send.
Change fcoe_ctlr_encaps() to lookup the destination vn_port in
the list of peers for the destination MAC address of the
FIP-encapsulated frame.
Add a new function fcoe_disc_init() to initialize just the
discovery portion of libfcoe for VN2VN mode.
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-21 06:20:30 +08:00
|
|
|
int fcoe_libfc_config(struct fc_lport *, struct fcoe_ctlr *,
|
|
|
|
|
const struct libfc_function_template *, int init_fcp);
|
2009-03-18 02:42:24 +08:00
|
|
|
|
2010-07-21 06:19:32 +08:00
|
|
|
/**
|
|
|
|
|
* is_fip_mode() - returns true if FIP mode selected.
|
|
|
|
|
* @fip: FCoE controller.
|
|
|
|
|
*/
|
|
|
|
|
static inline bool is_fip_mode(struct fcoe_ctlr *fip)
|
|
|
|
|
{
|
|
|
|
|
return fip->state == FIP_ST_ENABLED;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2008-12-10 07:10:24 +08:00
|
|
|
#endif /* _LIBFCOE_H */
|