Commit a0417fa3 ("net: Make qdisc_skb_cb upper size bound
explicit.") made it possible for a netdev driver to use skb->cb
between its header_ops.create method and its .ndo_start_xmit
method.  Use this in ipoib_hard_header() to stash away the LL address
(GID + QPN), instead of the "ipoib_pseudoheader" hack.  This allows
IPoIB to stop lying about its hard_header_len, which will let us fix
the L2 check for GRO.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: NArun Sharma <asharma@fb.com>
Reviewed-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As a first step in moving from LRO to GRO, revert commit af40da89
("IPoIB: add LRO support").  Also eliminate the ethtool set_flags
callback which isn't needed anymore.  Finally, we need to include
<linux/sched.h> directly to get the declaration of restart_syscall()
(which used to be included implicitly through <linux/inet_lro.h>).

Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: NOr Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Child devices were created without any offload features set, fix this by
moving the code that computes the features into generic function which is
now called through non-child and child device creation.
Signed-off-by: NOr Gerlitz <ogerlitz@voltaire.com>

-- v1 has a bug where the 'result' flag in ipoib_vlan_add may be used uninitialized
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

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: NRoland 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: NYossi Etigin <yosefe@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Increase IPoIB ring sizes to twice their original sizes (RX: 128->256,
TX: 64->128) to act as a shock absorber for high traffic peaks.  With
the current settings, we have seen cases that there are many calls to
netif_stop_queue(), which causes degradation in throughput.  Also,
larger receive buffer sizes help IPoIB in CM mode to avoid experiencing
RNR NAK conditions due to insufficient receive buffers at the SRQ.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Since IPoIB connected mode does not NETIF_F_SG, we only have one DMA
mapping per send, so we don't need a mapping[] array.  Define a new
struct with a single u64 mapping member and use it for the CM tx_ring.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland 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: NRoland 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: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

No need for a mutex around calls to ib_attach_mcast/ib_detach_mcast
since these operations are synchronized at the HW driver layer.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland 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: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The patch tries to solve the problem of device going down and paths being
flushed on an SM change event. The method is to mark the paths as candidates for
refresh (by setting the new valid flag to 0), and wait for an ARP
probe a new path record query.

The solution requires a different and less intrusive handling of SM
change event. For that, the second argument of the flush function
changes its meaning from a boolean flag to a level.  In most cases, SM
failover doesn't cause LID change so traffic won't stop.  In the rare
cases of LID change, the remote host (the one that hadn't changed its
LID) will lose connectivity until paths are refreshed. This is no
worse than the current state.  In fact, preventing the device from
going down saves packets that otherwise would be lost.
Signed-off-by: NMoni Levy <monil@voltaire.com>
Signed-off-by: NMoni Shoua <monis@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Add "ipoib_use_lro" module parameter to enable LRO and an
"ipoib_lro_max_aggr" module parameter to set the max number of packets
to be aggregated.  Make LRO controllable and LRO statistics accessible
through ethtool.
Signed-off-by: NVladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

The connected mode implementation in the IPoIB driver has a large
overhead in the way SKBs are handled in the receive flow.  It usually
allocates an SKB with as big as was used in the currently received SKB
and moves unused fragments from the old SKB to the new one. This
involves a loop on all the remaining fragments and incurs overhead on
the CPU.  This patch, for small SKBs, allocates an SKB just large
enough to contain the received data and copies to it the data from the
received SKB.  The newly allocated SKB is passed to the stack and the
old SKB is reposted.

When running netperf, UDP small messages, without this pach I get:

    UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to
    14.4.3.178 (14.4.3.178) port 0 AF_INET
    Socket  Message  Elapsed      Messages
    Size    Size     Time         Okay Errors   Throughput
    bytes   bytes    secs            #      #   10^6bits/sec

    114688     128   10.00     5142034      0     526.31
    114688           10.00     1130489            115.71

With this patch I get both send and receive at ~315 mbps.

The reason that send performance actually slows down is as follows:
When using this patch, the overhead of the CPU for handling RX packets
is dramatically reduced.  As a result, we do not experience RNR NAK
messages from the receiver which cause the connection to be closed and
reopened again; when the patch is not used, the receiver cannot handle
the packets fast enough so there is less time to post new buffers and
hence the mentioned RNR NACKs.  So what happens is that the
application *thinks* it posted a certain number of packets for
transmission but these packets are flushed and do not really get
transmitted.  Since the connection gets opened and closed many times,
each time netperf gets the CPU time that otherwise would have been
given to IPoIB to actually transmit the packets.  This can be verified
when looking at the port counters -- the output of ifconfig and the
oputput of netperf (this is for the case without the patch):

    tx packets
    ==========
    port counter:   1,543,996
    ifconfig:       1,581,426
    netperf:        5,142,034

    rx packets
    ==========
    netperf         1,1304,089
Signed-off-by: NEli Cohen <eli@mellanox.co.il>

They don't get updated by git and so they're worse than useless.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Commit f56bcd80 ("IPoIB: Use separate CQ for UD send completions")
introduced a bug where the transmit queue could get stopped and never
woken up.  The problem is that send completions are only polled at the
end of the xmit function, so if the send queue fills up and the xmit
path stops the queue, then there is no way for send completions to
ever get polled, and so the transmit queue stays stopped forever.

Fix this by arming the send CQ just before posting the last send
request that fills the send queue.  Then, when the completion event
handler is called, drain the send CQ.  Since it is possible that not
enough send completions are in the CQ, verify that the the net queue
has been woken up after draining the send CQ, and if not arm a timer
and drain again at the timer function.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Use a dedicated CQ for UD send completions. Also, do not arm the UD
send CQ, which reduces the number of interrupts generated.  This patch
farther reduces overhead by not calling poll CQ for every posted send
WR -- it does polls only when there 16 or more outstanding work requests.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland 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: NShirley Ma <xma@us.ibm.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

This can be used to tune at run time the parameters controlling the
event (interrupt) generation rate and thus reduce the overhead
incurred by handling interrupts resulting in better throughput.  Since
IPoIB uses a single CQ for both RX and TX, RX is chosen to dictate
configuration for both RX and TX.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Just add the infrastructure so we can add functionality later.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

For HCAs that support TCP segmentation offload (IB_DEVICE_UD_TSO), set
NETIF_F_TSO and use HW LSO to offload TCP segmentation.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

For HCAs that support checksum offload (ie that set IB_DEVICE_UD_IP_CSUM
in the device capabilities flags), have IPoIB set NETIF_F_IP_CSUM and
use the HCA to generate and verify IP checksums.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

struct ipoib_cm_tx.ibwc is unused since commit 1b524963 ("IPoIB/cm:
Use common CQ for CM send completions"), so remove it.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>

This patch acts as a preparation for using checksum offload for IB
devices capable of inserting/verifying checksum in IP packets.  The
patch does not actaully turn on NETIF_F_SG - we defer that to the
patches adding checksum offload capabilities.

We only add support for send gathers for datagram mode, since existing
HW does not support checksum offload on connected QPs.
Signed-off-by: NMichael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Some HCAs (such as ehca2) support SRQ, but only support fewer than 16 SG
entries for SRQs.  Currently IPoIB/CM implicitly assumes all HCAs will
support 16 SG entries for SRQs (to handle a 64K MTU with 4K pages). This
patch removes that restriction by limiting the maximum MTU in connected
mode to what the maximum number of SRQ SG entries allows.

This patch addresses <https://bugs.openfabrics.org/show_bug.cgi?id=728>
Signed-off-by: NPradeep Satyanarayana <pradeeps@linux.vnet.ibm.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Some IB adapters (notably IBM's eHCA) do not implement SRQs (shared
receive queues).  The current IPoIB connected mode support only works
on devices that support SRQs.

Fix this by adding support for using the receive queue of each
connected mode receive QP.  The disadvantage of this compared to using
an SRQ is that it means a full queue of receives must be posted for
each remote connected mode peer, which means that total memory usage
is potentially much higher than when using SRQs.  To manage this, add
a new module parameter "max_nonsrq_conn_qp" that limits the number of
connections allowed per interface.

The rest of the changes are fairly straightforward: we use a table of
struct ipoib_cm_rx to hold all the active connections, and put the
table index of the connection in the high bits of receive WR IDs.
This is needed because we cannot rely on the struct ib_wc.qp field for
non-SRQ receive completions.  Most of the rest of the changes just
test whether or not an SRQ is available, and post receives or find
received packets in the right place depending on the answer.

Cleaning up dead connections actually becomes simpler, because we do
not have to do the "last WQE reached" dance that is required to
destroy QPs attached to an SRQ.  We just move the QP to the error
state and wait for all pending receives to be flushed.
Signed-off-by: NPradeep Satyanarayana <pradeeps@linux.vnet.ibm.com>

[ Completely rewritten and split up, based on Pradeep's work.  Several
  bugs fixed and no doubt several bugs introduced.  - Roland ]
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Fix whitespace blunders, convert "foo* bar" to "foo *bar", etc.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Use the same CQ for CM send completions as for all other IPoIB
completions.  This means all completions are processed via the same
NAPI polling routine.  This should help reduce the number of
interrupts for bi-directional traffic (such as TCP) and fixes "driver
is hogging interrupts" errors reported for IPoIB send side, e.g.
<https://bugs.openfabrics.org/show_bug.cgi?id=508>

To do this, keep a per-interface counter of outstanding send WRs, and
stop the interface when this counter reaches the send queue size to
avoid CQ overruns.
Signed-off-by: NMichael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: NRoland 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: NRoland Dreier <rdreier@cisco.com>
Signed-off-by: NJeff 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: NRoland Dreier <rolandd@cisco.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Several devices have multiple independant RX queues per net
device, and some have a single interrupt doorbell for several
queues.

In either case, it's easier to support layouts like that if the
structure representing the poll is independant from the net
device itself.

The signature of the ->poll() call back goes from:

	int foo_poll(struct net_device *dev, int *budget)

to

	int foo_poll(struct napi_struct *napi, int budget)

The caller is returned the number of RX packets processed (or
the number of "NAPI credits" consumed if you want to get
abstract).  The callee no longer messes around bumping
dev->quota, *budget, etc. because that is all handled in the
caller upon return.

The napi_struct is to be embedded in the device driver private data
structures.

Furthermore, it is the driver's responsibility to disable all NAPI
instances in it's ->stop() device close handler.  Since the
napi_struct is privatized into the driver's private data structures,
only the driver knows how to get at all of the napi_struct instances
it may have per-device.

With lots of help and suggestions from Rusty Russell, Roland Dreier,
Michael Chan, Jeff Garzik, and Jamal Hadi Salim.

Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra,
Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan.

[ Ported to current tree and all drivers converted.  Integrated
  Stephen's follow-on kerneldoc additions, and restored poll_list
  handling to the old style to fix mutual exclusion issues.  -DaveM ]
Signed-off-by: NStephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: NDavid 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: NOr Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

To support QoS within and between subnets, modify IPoIB to request
specific Traffic Class values with path record queries, using
the value associated with the IPoIB broadcast group.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>

[ See some comments I made on this at v1 and v2 of the posts
  <http://lists.openfabrics.org/pipermail/general/2007-August/039275.html>
  <http://lists.openfabrics.org/pipermail/general/2007-September/040312.html> ]
Reviewed-by: NOr Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

commit 518b1646 ("IPoIB/cm: Fix SRQ WR leak") introduced a severe
performance regression on Mellanox cards, because keeping a QP in the
error state for extended periods of time moves hardware to the slow
path (until the QP is destroyed).  For example, MPI latency goes from
~3 usecs to ~7 usecs.

Fix this by posting a send WR on one of the QPs that are being
flushed, instead of using a separate drain QP that is kept in the
error state.

This fixes bug <https://bugs.openfabrics.org/show_bug.cgi?id=636>,
reported and bisected by Scott Weitzenkamp at Cisco and debugged by
Sasha Mikheev at Voltaire.
Signed-off-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

Since NAPI polling is disabled while ipoib_cm_dev_stop() is running,
ipoib_cm_dev_stop() must poll the CQ itself in order to see the
packets draining.
Signed-off-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

SRQ WR leakage has been observed with IPoIB/CM: e.g. flipping ports on
and off will, with time, leak out all WRs and then all connections
will start getting RNR NAKs.  Fix this in the way suggested by spec:
move the QP being destroyed to the error state, wait for "Last WQE
Reached" event and then post WR on a "drain QP" connected to the same
CQ.  Once we observe a completion on the drain QP, it's safe to call
ib_destroy_qp.
Signed-off-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

SM reconfiguration or failover possibly causes a shuffling of the values
in the P_Key table. Right now, IPoIB only queries for the P_Key index
once when it creates the device QP, and hence there are problems if the
index of a P_Key value changes.  Fix this by using the PKEY_CHANGE event
to trigger a recheck of the P_Key index.
Signed-off-by: NYosef Etigin <yosefe@voltaire.com>
Acked-by: NMichael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>