Currently, IPoIB is an LLTX driver that uses its own IRQ-disabling
tx_lock. Not only do we want to get rid of LLTX, this actually causes
problems because of the skb_orphan() done with this tx_lock held: some
skb destructors expect to be run with interrupts enabled.
The simplest fix for this is to get rid of the driver-private tx_lock
and stop using LLTX. We kill off priv->tx_lock and use
netif_tx_lock[_bh]() instead; the patch to do this is a tiny bit
tricky because we need to update places that take priv->lock inside
the tx_lock to disable IRQs, rather than relying on tx_lock having
already disabled IRQs.
Also, there are a couple of places where we need to disable BHs to
make sure we have a consistent context to call netif_tx_lock() (since
we no longer can use _irqsave() variants), and we also have to change
ipoib_send_comp_handler() to call drain_tx_cq() through a timer rather
than directly, because ipoib_send_comp_handler() runs in interrupt
context and drain_tx_cq() must run in BH context so it can call
netif_tx_lock().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Taking rtnl_lock in ipoib_mcast_join_complete() causes a deadlock with
ipoib_stop(). We avoid it by scheduling the piece of code that takes
the lock on ipoib_workqueue instead of executing it directly. This
works because we only flush the ipoib_workqueue with the RTNL not held.
The deadlock happens because ipoib_stop() calls ipoib_ib_dev_down()
which calls ipoib_mcast_dev_flush(), which calls ipoib_mcast_free(),
which calls ipoib_mcast_leave(). The latter calls
ib_sa_free_multicast(), and this waits until the multicast completion
handler finishes. This handler is ipoib_mcast_join_complete(), which
waits for the rtnl_lock(), which was already taken by ipoib_stop().
This bug was introduced in commit a77a57a1 ("IPoIB: Fix deadlock on
RTNL in ipoib_stop()").
Signed-off-by: Yossi Etigin <yosefe@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit c8c2afe3 ("IPoIB: Use rtnl lock/unlock when changing device
flags") added a call to rtnl_lock() in ipoib_mcast_join_task(), which
is run from the ipoib_workqueue. However, ipoib_stop() (which is run
inside rtnl_lock()) flushes this workqueue, which leads to a deadlock
if the join task is pending.
Fix this by simply not flushing the workqueue from ipoib_stop(). It
turns out that we really don't care about workqueue tasks running
during or after ipoib_stop(), as long as we make sure to flush the
workqueue before unregistering a netdev.
This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=1114>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Now that we have a specific lock to protect the network
device unicast and multicast lists, remove extraneous
grabs of the TX lock in cases where the code only needs
address list protection.
Signed-off-by: David S. Miller <davem@davemloft.net>
Add netif_addr_{lock,unlock}{,_bh}() helpers.
Use them to protect operations that operate on or read
the network device unicast and multicast address lists.
Also use them in cases where the code simply wants to
block calls into the driver's ->set_rx_mode() and
->set_multicast_list() methods.
Signed-off-by: David S. Miller <davem@davemloft.net>
When the driver sets the MTU of the net device outside of its
change_mtu method, it should make use of dev_set_mtu() instead of
directly setting the mtu field of struct netdevice. Otherwise
functions registered to be called upon MTU change will not get called
(this is done through call_netdevice_notifiers() in dev_set_mtu()).
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use of this lock is required to synchronize changes to the netdvice's
data structs. Also move the call to ipoib_flush_paths() after the
modification of the netdevice flags in set_mode().
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ipoib_mcast_detach() does nothing except call ib_detach_mcast(), so just
use the core API in the one place that does a multicast group detach.
add/remove: 0/1 grow/shrink: 0/1 up/down: 0/-105 (-105)
function old new delta
ipoib_mcast_leave 357 319 -38
ipoib_mcast_detach 67 - -67
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The current code will set the Q_Key for any join of a non-sendonly
multicast group. The operation involves a modify QP operation, which
is fairly heavyweight, and is only really required after the join of
the broadcast group. Fix this by adding a parameter to ipoib_mcast_attach()
to control when the Q_Key is set.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The IPOIB_MCAST_STARTED flag is not used at all since commit b3e2749b
("IPoIB: Don't drop multicast sends when they can be queued"), so
remove it.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
We saw a kernel oops in our regression testing when a multicast "join
finish" occurred just after the interface was -- this is
<https://bugs.openfabrics.org/show_bug.cgi?id=1040>. The test
randomly causes the HCA physical port to go down then up.
The cause of this is that ipoib_mcast_join_finish() processing happen
just after ipoib_mcast_dev_flush() was invoked (in which case the
broadcast pointer is NULL). This patch tests for and handles the case
where priv->broadcast is NULL.
Cc: <stable@kernel.org>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch enables IPoIB to use 4K UD messages (when the underlying
device and fabrics support a 4K MTU) by using two scatter buffers when
PAGE_SIZE is less than or equal to thhe HCA IB MTU size. The first
buffer is for IPoIB header + GRH header, and the second buffer is the
IPoIB payload, which is 4K-4.
Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When set_multicast_list() is called the multicast task is restarted
and the IPOIB_MCAST_STARTED bit is cleared. As a result for some
window of time, multicast packets are not transmitted nor queued but
rather dropped by ipoib_mcast_send(). These dropped packets are
painful in two cases:
- bonding fail-over which both calls set_multicast_list() on the new
active slave and sends Gratuitous ARP through that slave.
- IP_DROP_MEMBERSHIP code which both calls set_multicast_list() on the
device and issues IGMP leave.
In both these cases, depending on the scheduling of the IPoIB
multicast task, the packets would be dropped. As a result, in the
bonding case, the failover would not be detected by the peers until
their neighbour is renewed the neighbour (which takes a few tens of
seconds). In the IGMP case, the IP router doesn't get an IGMP leave
and would only learn on that from further probes on the group (also a
delay of at least a few tens of seconds).
Fix this by allowing transmission (or queuing) depending on the
IPOIB_FLAG_OPER_UP flag instead of the IPOIB_MCAST_STARTED flag.
Signed-off-by: Olga Shern <olgas@voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
An IPoIB subnet on an IB fabric that spans multiple IB subnets can't
use link-local scope in multicast GIDs. The existing routines that
map IP/IPv6 multicast addresses into IB link-level addresses hard-code
the scope to link-local, and they also leave the partition key field
uninitialised. This patch adds a parameter (the link-level broadcast
address) to the mapping routines, allowing them to initialise both the
scope and the P_Key appropriately, and fixes up the call sites.
The next step will be to add a way to configure the scope for an IPoIB
interface.
Signed-off-by: Rolf Manderscheid <rvm@obsidianresearch.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
IPoIB uses a two layer neighboring scheme, such that for each struct neighbour
whose device is an ipoib one, there is a struct ipoib_neigh buddy which is
created on demand at the tx flow by an ipoib_neigh_alloc(skb->dst->neighbour)
call.
When using the bonding driver, neighbours are created by the net stack on behalf
of the bonding (master) device. On the tx flow the bonding code gets an skb such
that skb->dev points to the master device, it changes this skb to point on the
slave device and calls the slave hard_start_xmit function.
Under this scheme, ipoib_neigh_destructor assumption that for each struct
neighbour it gets, n->dev is an ipoib device and hence netdev_priv(n->dev)
can be casted to struct ipoib_dev_priv is buggy.
To fix it, this patch adds a dev field to struct ipoib_neigh which is used
instead of the struct neighbour dev one, when n->dev->flags has the
IFF_MASTER bit set.
Signed-off-by: Moni Shoua <monis at voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz at voltaire.com>
Acked-by: Roland Dreier <rdreier@cisco.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Use the stats member of struct netdevice in IPoIB, so we can save
memory by deleting the stats member of struct ipoib_dev_priv, and save
code by deleting ipoib_get_stats().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kernel IB stack allows (through the RDMA CM) userspace
applications to join and use multicast groups from the IPoIB MGID
range. This allows multicast traffic to be handled directly from
userspace QPs, without going through the kernel stack, which gives
better performance for some applications.
However, to fully interoperate with IP multicast, such userspace
applications need to participate in IGMP reports and queries, or else
routers may not forward the multicast traffic to the system where the
application is running. The simplest way to do this is to share the
kernel IGMP implementation by using the IP_ADD_MEMBERSHIP option to
join multicast groups that are being handled directly in userspace.
However, in such cases, the actual multicast traffic should not also
be handled by the IPoIB interface, because that would burn resources
handling multicast packets that will just be discarded in the kernel.
To handle this, this patch adds lookup on the database used for IB
multicast group reference counting when IPoIB is joining multicast
groups, and if a multicast group is already handled by user space,
then the IPoIB kernel driver ignores the group. This is controlled by
a per-interface policy flag. When the flag is set, IPoIB will not
join and attach its QP to a multicast group which already has an entry
in the database; when the flag is cleared, IPoIB will behave as before
this change.
For each IPoIB interface, the /sys/class/net/$intf/umcast attribute
controls the policy flag. The default value is off/0.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
There's a race between ipoib_mcast_leave() and ipoib_mcast_join_finish()
where we can try to detach from a multicast group before we've
attached to it. Fix this by reordering the code in ipoib_mcast_leave
to free the multicast group first, which waits for the multicast
callback thread (which calls ipoib_mcast_join_finish()) to complete
before detaching from the group.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Do netif_carrier_on() right after the IPv4 broadcast multicast group
is joined, rather than waiting for all of the initial set of multicast
group joins to finish. This allows at least IPv4 traffic to limp
along on broken fabrics where not all multicast groups can be joined.
Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Now that low-level drivers handle the conversion from an absolute rate
to a relative rate, there's no need for the IPoIB driver to keep track
of the local port's data rate.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The IB SA tracks multicast join/leave requests on a per port basis and
does not do any reference counting: if two users of the same port join
the same group, and one leaves that group, then the SA will remove the
port from the group even though there is one user who wants to stay a
member left. Therefore, in order to support multiple users of the
same multicast group from the same port, we need to perform reference
counting locally.
To do this, add an multicast submodule to ib_sa to perform reference
counting of multicast join/leave operations. Modify ib_ipoib (the
only in-kernel user of multicast) to use the new interface.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The following patch adds experimental support for IPoIB connected
mode, as defined by the draft from the IETF ipoib working group. The
idea is to increase performance by increasing the MTU from the maximum
of 2K (theoretically 4K) supported by IPoIB on top of UD. With this
code, I'm able to get 800MByte/sec or more with netperf without
options on a Mellanox 4x back-to-back DDR system.
Some notes on code:
1. SRQ is used for scalability to large cluster sizes
2. Only RC connections are used (UC does not support SRQ now)
3. Retry count is set to 0 since spec draft warns against retries
4. Each connection is used for data transfers in only 1 direction, so
each connection is either active(TX) or passive (RX). 2 sides that
want to communicate create 2 connections.
5. Each active (TX) connection has a separate CQ for send completions -
this keeps the code simple without CQ resize and other tricks
6. To detect stale passive side connections (where the remote side is
down), we keep an LRU list of passive connections (updated once per
second per connection) and destroy a connection after it has been
unused for several seconds. The LRU rule makes it possible to avoid
scanning connections that have recently been active.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ipoib_neigh_free() is sometimes called while neighbour is still alive,
so it might still have queued skbs. Fix skb leak in this case.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
RFC 4391 ("Transmission of IP over InfiniBand (IPoIB)") says:
If the IB multicast group does not already exist, one must be
created first with the IPoIB link MTU. The MGID MUST use the same
P_Key, Q_Key, SL, MTU, and HopLimit as those used in the
broadcast-GID. The rest of attributes SHOULD follow the values used
in the broadcast-GID as well.
However, the current IPoIB driver is only setting the attributes
required by the InfiniBand spec to create a multicast group, so in
particular the MTU and HopLimit are not being set. Add these
attributes when creating MCGs, and also set the Rate attribute, since
IPoIB pays attention to that attribute as well.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Require users to register with SA module, to prevent the sa_query
module text from going away while an SA query callback is still
running. Update all in-tree users for the new interface.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Remove some trailing whitespace that has snuck in despite the best
efforts of whitespace=error-all. Also fix a few other whitespace
bogosities.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When a send-only multicast group join fails, mcast->query must be set
to NULL. Otherwise, IPoIB will never retry the join and the multicast
group will never be reachable.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Need to set mcast->ah before debug code dereferences it.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch converts the combination of list_del(A) and list_add(A, B) to
list_move(A, B) under drivers/.
Acked-by: Corey Minyard <minyard@mvista.com>
Cc: Ben Collins <bcollins@debian.org>
Acked-by: Roland Dreier <rolandd@cisco.com>
Cc: Alasdair Kergon <dm-devel@redhat.com>
Cc: Gerd Knorr <kraxel@bytesex.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frank Pavlic <fpavlic@de.ibm.com>
Acked-by: Matthew Wilcox <matthew@wil.cx>
Cc: Andrew Vasquez <linux-driver@qlogic.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Greg Kroah-Hartman <greg@kroah.com>
Signed-off-by: Akinobu Mita <mita@miraclelinux.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Various drivers use xmit_lock internally to synchronise with their
transmission routines. They do so without setting xmit_lock_owner.
This is fine as long as netpoll is not in use.
With netpoll it is possible for deadlocks to occur if xmit_lock_owner
isn't set. This is because if a printk occurs while xmit_lock is held
and xmit_lock_owner is not set can cause netpoll to attempt to take
xmit_lock recursively.
While it is possible to resolve this by getting netpoll to use
trylock, it is suboptimal because netpoll's sole objective is to
maximise the chance of getting the printk out on the wire. So
delaying or dropping the message is to be avoided as much as possible.
So the only alternative is to always set xmit_lock_owner. The
following patch does this by introducing the netif_tx_lock family of
functions that take care of setting/unsetting xmit_lock_owner.
I renamed xmit_lock to _xmit_lock to indicate that it should not be
used directly. I didn't provide irq versions of the netif_tx_lock
functions since xmit_lock is meant to be a BH-disabling lock.
This is pretty much a straight text substitution except for a small
bug fix in winbond. It currently uses
netif_stop_queue/spin_unlock_wait to stop transmission. This is
unsafe as an IRQ can potentially wake up the queue. So it is safer to
use netif_tx_disable.
The hamradio bits used spin_lock_irq but it is unnecessary as
xmit_lock must never be taken in an IRQ handler.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix misaligned access faults on ia64: never cast a misaligned
neighbour->ha + 4 pointer to union ib_gid type; pass a void * pointer
instead. The memcpy was being optimized to use full word accesses
because the compiler thought that union ib_gid is always aligned.
The cast in IPOIB_GID_ARG is safe, since it is fixed to access each
byte separately.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ipoib_mcast_restart_task() might free an mcast object while a join
request is still outstanding, leading to an oops when the query
completes. Fix this by waiting for query to complete, similar to what
ipoib_stop_thread() is doing. The wait for mcast completion code is
consolidated in wait_for_mcast_join().
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Push translation of static rate to HCA format into low-level drivers,
where it belongs. For static rate encoding, use encoding of rate
field from IB standard PathRecord, with addition of value 0, for
backwards compatibility with current usage. The changes are:
- Add enum ib_rate to midlayer includes.
- Get rid of static rate translation in IPoIB; just use static rate
directly from Path and MulticastGroup records.
- Update mthca driver to translate absolute static rate into the
format used by hardware. This also fixes mthca's static rate
handling for HCAs that are capable of 4X DDR.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Consolidate IPoIB's private neighbour data handling into
ipoib_neigh_alloc() and ipoib_neigh_free(). This will make it easier
to keep track of the neighbour structures that IPoIB is handling, and
is a nice cleanup of the code:
add/remove: 2/1 grow/shrink: 1/8 up/down: 100/-178 (-78)
function old new delta
ipoib_neigh_alloc - 61 +61
ipoib_neigh_free - 36 +36
ipoib_mcast_join_finish 1288 1291 +3
path_rec_completion 575 573 -2
ipoib_mcast_join_task 664 660 -4
ipoib_neigh_destructor 101 92 -9
ipoib_neigh_setup_dev 14 3 -11
ipoib_neigh_setup 17 - -17
path_free 238 215 -23
ipoib_mcast_free 329 306 -23
ipoib_mcast_send 718 684 -34
neigh_add_path 705 650 -55
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix the IPoIB build (which is broken in net-2.6.17 because of my
screw-up, which left out this chunk in ipoib_multicast.c).
The neighbour destructor is now in neigh_params, so we don't
need to clear it in the ops structure.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ipoib_mcast_stop_thread currently tests mcast->query and if it is
NULL, does not perform wait_for_completion on the mcast and frees the
mcast object directly.
However, since both operations are done without locking, it is
possible that ipoib_mcast_join_complete is in progress on this mcast
object and has set mcast->query to NULL already.
Solve this by:
- taking priv->lock before we change mcast->query in ipoib_mcast_join_complete,
and keeping it until we no longer need the mcast object
- taking priv->lock around mcast->query test in ipoib_mcast_stop_thread
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ipoib_mcast_send() tests mcast->ah twice. If this value is changed
between these two points, we leak an skb. However,
ipoib_mcast_join_finish() sets mcast->ah with no locking, so it could
race against ipoib_mcast_send().
As a solution, take priv->lock around assignment to mcast->ah thus
making sure ipoib_mcast_send() (which also takes priv->lock) is not in
flight.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Even after the last fix, it's still possible for a send-only join to
start before the join for the broadcast group has finished. This
could cause us to create a multicast group using attributes from the
broadcast group that haven't been initialized yet, so we would use
garbage for the Q_Key, etc. Fix this by waiting until the broadcast
group's attached flag is set before starting send-only joins.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Further, there's an additional issue that I saw in testing:
ipoib_mcast_send may get called when priv->broadcast is NULL (e.g. if
the device was downed and then upped internally because of a port
event).
If this happends and the send-only join request gets completed before
priv->broadcast is set, we get an oops.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix the following race scenario:
- Device is up.
- Port event or set mcast list triggers ipoib_mcast_stop_thread,
this cancels the query and waits on mcast "done" completion.
- Completion is called and "done" is set.
- Meanwhile, ipoib_mcast_send arrives and starts a new query,
re-initializing "done".
Fix this by adding a "multicast started" bit and checking it before
starting a send-only join.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Avoid corrupting mcast->pkt_queue by serializing access with
priv->tx_lock. Also, update dropped packet statistics to count
multicast packets removed from pkt_queue as dropped.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
semaphore to mutex conversion by Ingo and Arjan's script.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
[ Sanity-checked on real IB hardware ]
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The current handling of multicast groups in IPoIB ends up never
freeing send-only multicast groups. It turns out the logic was much
more complicated than it needed to be; we can fix this bug and
completely kill ipoib_mcast_dev_down() at the same time.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
dev->mc_list accesses must be protected by dev->xmit_lock.
Found by Eli Cohen <eli@mellanox.co.il>.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Roland Dreier