IB/hfi1: Fix alloc failure with larger txqueuelen

The following allocation with large txqueuelen will result in the
following warning:

Call Trace:
__alloc_pages_nodemask+0x283/0x2c0
kmalloc_large_node+0x3c/0xa0
__kmalloc_node+0x22a/0x2f0
hfi1_ipoib_txreq_init+0x19f/0x330 [hfi1]
hfi1_ipoib_setup_rn+0xd3/0x1a0 [hfi1]
rdma_init_netdev+0x5a/0x80 [ib_core]
ipoib_intf_init+0x6c/0x350 [ib_ipoib]
ipoib_intf_alloc+0x5c/0xc0 [ib_ipoib]
ipoib_add_one+0xbe/0x300 [ib_ipoib]
add_client_context+0x12c/0x1a0 [ib_core]
ib_register_client+0x147/0x190 [ib_core]
ipoib_init_module+0xdd/0x132 [ib_ipoib]
do_one_initcall+0x46/0x1c3
do_init_module+0x5a/0x220
load_module+0x14c5/0x17f0
__do_sys_init_module+0x13b/0x180
do_syscall_64+0x5b/0x1a0
entry_SYSCALL_64_after_hwframe+0x65/0xca

For ipoib, the txqueuelen is modified with the module parameter
send_queue_size.

Fix by changing to use kv versions of the same allocator to handle the
large allocations. The allocation embeds a hdr struct that is dma mapped.
Change that struct to a pointer to a kzalloced struct.

Cc: stable@vger.kernel.org
Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/1642287756-182313-3-git-send-email-mike.marciniszyn@cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Add ring consumer and producers traces

These traces are used to debugging ring issues.

The ipoib_txreq needed to be moved to a header file to allow access from
the trace header file.

The trace changes include:
- new producer/consumer traces
- new allocation deallocation traces
- additional fidelity for SDMA engine prints

Fixes: 4bd00b55c978 ("IB/hfi1: Add AIP tx traces")
Link: https://lore.kernel.org/r/20210913132852.131370.9664.stgit@awfm-01.cornelisnetworks.com
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Remove atomic completion count

The atomic is not needed.

Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132847.131370.54250.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Tune netdev xmit cachelines

This patch moves fields in the ring and creates a line for the producer
and the consumer.

The adds a consumer side variable that tracks the ring avail so that the
code doesn't have the read the other cacheline to get a count for every
packet. A read now only occurs when the avail is at 0.

Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132842.131370.15636.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Get rid of hot path divide

The pointer math in this statemet does a divide;
struct hfi1_ipoib_txq *txq = &priv->txqs[napi - priv->tx_napis];

Elminate the divide by embedding the struct napi_strut in the txq and
getting the txq with a container_of() using the newly embedded napi.

Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132831.131370.3993.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Remove cache and embed txreq in ring

This patch removes kmem cache allocation and deallocation in favor of
having the ipoib_txreq in the ring.

The consumer is now the packet sending side allocating tx descriptors from
ring and the producer is the napi interrupt handling freeing tx
descriptors.

The locks are now eliminated because the napi tx lock insures a single
consumer and the napi handling insures a single producer.

The napi poll is converted to memory poll looking for items that have been
marked completed.

Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132826.131370.4397.stgit@awfm-01.cornelisnetworks.com
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Remove indirect call to hfi1_ipoib_send_dma()

hfi1_ipoib_send() directly calls hfi1_ipoib_send_dma() with no value add.

Fix by renaming hfi1_ipoib_send_dma() to hfi1_ipoib_send().

Link: https://lore.kernel.org/r/1617026056-50483-6-git-send-email-dennis.dalessandro@cornelisnetworks.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Correct oversized ring allocation

The completion ring for tx is using the wrong size to size the ring,
oversizing the ring by two orders of magniture.

Correct the allocation size and use kcalloc_node() to allocate the ring.
Fix mistaken GFP defines in similar allocations.

Link: https://lore.kernel.org/r/1617026056-50483-4-git-send-email-dennis.dalessandro@cornelisnetworks.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdev

The current rdma_netdev handling in ipoib hooks the tx_timeout handler,
but prints out a totally useless message that prevents effective debugging
especially when multiple transmit queues are being used.

Add a tx_timeout rdma_netdev hook and implement the callback in the hfi1
to print additional information.

The existing non-helpful message is avoided when the driver has presented
a callback.

Link: https://lore.kernel.org/r/1617026056-50483-3-git-send-email-dennis.dalessandro@cornelisnetworks.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA: Support more than 255 rdma ports

Current code uses many different types when dealing with a port of a RDMA
device: u8, unsigned int and u32. Switch to u32 to clean up the logic.

This allows us to make (at least) the core view consistent and use the
same type. Unfortunately not all places can be converted. Many uverbs
functions expect port to be u8 so keep those places in order not to break
UAPIs. HW/Spec defined values must also not be changed.

With the switch to u32 we now can support devices with more than 255
ports. U32_MAX is reserved to make control logic a bit easier to deal
with. As a device with U32_MAX ports probably isn't going to happen any
time soon this seems like a non issue.

When a device with more than 255 ports is created uverbs will report the
RDMA device as having 255 ports as this is the max currently supported.

The verbs interface is not changed yet because the IBTA spec limits the
port size in too many places to be u8 and all applications that relies in
verbs won't be able to cope with this change. At this stage, we are
extending the interfaces that are using vendor channel solely

Once the limitation is lifted mlx5 in switchdev mode will be able to have
thousands of SFs created by the device. As the only instance of an RDMA
device that reports more than 255 ports will be a representor device and
it exposes itself as a RAW Ethernet only device CM/MAD/IPoIB and other
ULPs aren't effected by this change and their sysfs/interfaces that are
exposes to userspace can remain unchanged.

While here cleanup some alignment issues and remove unneeded sanity
checks (mainly in rdmavt),

Link: https://lore.kernel.org/r/20210301070420.439400-1-leon@kernel.org
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: switch to core handling of rx/tx byte/packet counters

Use netdev->tstats instead of a member of hfi1_ipoib_dev_priv for storing
a pointer to the per-cpu counters. This allows us to use core
functionality for statistics handling.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

IB/hfi1: Add atomic triggered sleep/wakeup

When running iperf in a two host configuration the following trace can
occur:

[ 319.728730] NETDEV WATCHDOG: ib0 (hfi1): transmit queue 0 timed out

The issue happens because the current implementation relies on the netif
txq being stopped to control the flushing of the tx list.

There are two resources that the transmit logic can wait on and stop the
txq:
- SDMA descriptors
- Ring space to hold completions

The ring space is tested on the sending side and relieved when the ring is
consumed in the napi tx reaping.

Unfortunately, that reaping can run conncurrently with the workqueue
flushing of the txlist. If the txq is started just before the workitem
executes, the txlist will never be flushed, leading to the txq being
stuck.

Fix by:
- Adding sleep/wakeup wrappers
* Use an atomic to control the call to the netif routines inside the
wrappers

- Use another atomic to record ring space exhaustion
* Only wakeup when the a ring space exhaustion has happened and it
relieved

Add additional wrappers to clarify the ring space resource handling.

Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20200623204327.108092.4024.stgit@awfm-01.aw.intel.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

IB/hfi1: Add rx functions for dummy netdev

This patch adds the rx functions for the dummy netdev:
- Functions to allocate/free the dummy netdev.
- Functions to allocate/free receiving contexts for the netdev.
- Functions to initialize/de-initialize the receive queue.
- Functions to enable/disable the receive queue.

Link: https://lore.kernel.org/r/20200511160643.173205.75087.stgit@awfm-01.aw.intel.com
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Sadanand Warrier <sadanand.warrier@intel.com>
Signed-off-by: Grzegorz Andrejczuk <grzegorz.andrejczuk@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

IB/hfi1: Add functions to receive accelerated ipoib packets

Ipoib netdev will share receive contexts with existing VNIC netdev.
To achieve that, a dummy netdev is allocated with hfi1_devdata to
own the receive contexts, and ipoib and VNIC netdevs will be put
on top of it. Each receive context is associated with a single
NAPI object.

This patch adds the functions to receive incoming packets for
accelerated ipoib.

Link: https://lore.kernel.org/r/20200511160631.173205.54184.stgit@awfm-01.aw.intel.com
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Sadanand Warrier <sadanand.warrier@intel.com>
Signed-off-by: Grzegorz Andrejczuk <grzegorz.andrejczuk@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

IB/hfi1: Add the transmit side of a datagram ipoib RDMA netdev

This implements the transmit side of the multiple transmit queue RDMA
netdev used to accelerate ipoib. The receive side remains the ipoib
internal implementation.

The init/unint/open/stop netdev operations are saved off and called by the
versions within the hfi1 netdev in order to initialize the connected mode
resources present in ipoib thus allowing us to switch modes between
datagram and connected.

The datagram queue pair instantiated by the ipoib ulp is used by this
implementation for its queue pair number and to register with multicast.

The above queue pair is not used on transmit other than its qpn as the
verbs layer is skipped and packets are directly submitted to the sdma
engines.

Link: https://lore.kernel.org/r/20200511160554.173205.1369.stgit@awfm-01.aw.intel.com
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Gary Leshner <Gary.S.Leshner@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

IB/hfi1: Add functions to transmit datagram ipoib packets

This patch implements the mechanism to accelerate the transmit side of
a multiple transmit queue RDMA netdev by submitting the packets to
the SDMA engine directly instead of sending through the verbs layer.

This patch also changes the UD/SEND_ONLY op to output the entropy value
in byte 0 of deth[1]. UD/SEND_ONLY_WITH_IMMEDIATE uses the previous
behavior with no entropy value being output.

The code in the ipoib rdma netdev which submits tx requests upon
successful submission will call trace_sdma_output_ibhdr to output
the ibhdr to the trace buffer.

Link: https://lore.kernel.org/r/20200511160548.173205.45616.stgit@awfm-01.aw.intel.com
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Gary Leshner <Gary.S.Leshner@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>