The PD id is masked with 0x7fff, while PD can be 18 bits for GEN2 HW.
Remove the masking as it should not be needed and can cause incorrect PD
id to be used.

Fixes: b48c24c2 ("RDMA/irdma: Implement device supported verb APIs")
Link: https://lore.kernel.org/r/20220225163211.127-4-shiraz.saleem@intel.comSigned-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Using PCI_FUNC macro in a VM, when the device is in passthrough mode does
not provide the real function instance. This means that currently, devices
will not probe unless the instance in the VM matches the instance in the
host.

Fix this by getting the pf_id from the LAN during the probe.

Fixes: 8498a30e ("RDMA/irdma: Register auxiliary driver and implement private channel OPs")
Link: https://lore.kernel.org/r/20220225163211.127-3-shiraz.saleem@intel.comSigned-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Currently, events on vlan netdevs are being ignored. Fix this by finding
the real netdev and processing the notifications for vlan netdevs.

Fixes: 915cc7ac ("RDMA/irdma: Add miscellaneous utility definitions")
Link: https://lore.kernel.org/r/20220225163211.127-2-shiraz.saleem@intel.comSigned-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

The function irdma_create_mg_ctx always returns 0, so make it void and
delete the return value check.

Link: https://lore.kernel.org/r/20220224182832.3896686-1-yanjun.zhu@linux.devSigned-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Acked-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

The union irdma_sockaddr is used frequently. So move it to the header
file.

Link: https://lore.kernel.org/r/20220223024252.3873736-4-yanjun.zhu@linux.devSigned-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Firstly the variable saddr was to check the type of a network. Now the
variable net_type is used to do the same work. So it is removed.

Link: https://lore.kernel.org/r/20220223024252.3873736-3-yanjun.zhu@linux.devSigned-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

The member variable net_type is to check the type of network.

Link: https://lore.kernel.org/r/20220223024252.3873736-2-yanjun.zhu@linux.devSigned-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Finish adding subroutine comment headers to subroutines in
rxe_mcast.c. Make minor api change cleanups.

Link: https://lore.kernel.org/r/20220223230706.50332-5-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Collect cleanup code for struct rxe_mca into a subroutine,
__rxe_cleanup_mca() called in rxe_detach_mcg() in rxe_mcast.c.

Link: https://lore.kernel.org/r/20220223230706.50332-4-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Collect initialization code for struct rxe_mca into a subroutine,
__rxe_init_mca(), to cleanup rxe_attach_mcg() in rxe_mcast.c. Check
limit on total number of attached qp's.

Link: https://lore.kernel.org/r/20220223230706.50332-3-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Print a warning if memory allocated by mcast
is not cleared when the rxe driver is unloaded.

Link: https://lore.kernel.org/r/20220223230706.50332-2-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

As irdma_status_code is replaced with an int, there is no need for two
variables to hold error codes.

Remove the excess variable in functions where this occurs.  Also, remove
any redundant initializations which are no longer needed.

Link: https://lore.kernel.org/r/20220217151851.1518-4-shiraz.saleem@intel.comSigned-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

All functions now return linux error codes. Propagate the return from
these functions as opposed to converting them to generic values.

Link: https://lore.kernel.org/r/20220217151851.1518-3-shiraz.saleem@intel.comSigned-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Replace use of custom irdma_status_code with linux error codes.

Remove enum irdma_status_code and header in which its defined.

Link: https://lore.kernel.org/r/20220217151851.1518-2-shiraz.saleem@intel.comSigned-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Make it more clear what the different ib_srp data structures represent.

Link: https://lore.kernel.org/r/20220215210511.28303-2-bvanassche@acm.orgSigned-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

The mkc is the key for the mkey cache, hence, created in each attempt to
get a cache mkey, while pcie_relaxed_ordering_enabled() is called during
the setting of the mkc, but used only for cases where
IB_ACCESS_RELAXED_ORDERING is set.

pcie_relaxed_ordering_enabled() is an expensive call (26 us). Reorder the
code so the driver will call it only when it is needed.

Link: https://lore.kernel.org/r/684be1366cb1d4f05aa3e78986205e4bc410443a.1644947594.git.leonro@nvidia.comSigned-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Currently, ent->xlt stores the translation table size. This data should
not be stored in the cache entry but be written directly to the mailbox.
Store ndescs instead, and deduce the translation table size from it
according to the access mode.

Link: https://lore.kernel.org/r/e9dbfaa1f279793a6bd28ee5a31cb4f0f0d70f05.1644947594.git.leonro@nvidia.comSigned-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

When allocating a MR from the cache, the driver calls to get_cache_mr(),
and in case of failure, retries with create_cache_mr(). This is the flow
of mlx5_mr_cache_alloc(), so use it instead.

Link: https://lore.kernel.org/r/53c85fcd4de6ec9de0b8e6cbb1bf5d5fe19900c3.1644947594.git.leonro@nvidia.comSigned-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

When an ODP MR cache entry is empty and trying to allocate it, increment
the ent->miss counter and call to queue_adjust_cache_locked() to verify
the entry is balanced.

Fixes: aad719dc ("RDMA/mlx5: Allow MRs to be created in the cache synchronously")
Link: https://lore.kernel.org/r/09503e295276dcacc92cb1d8aef1ad0961c99dc1.1644947594.git.leonro@nvidia.comSigned-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

delayed_cache_work_func() and the cache_work_func() are both wrappers of
__cache_work_func(). Instead of having a special not delayed work, use the
delayed work with delay = 0.

Link: https://lore.kernel.org/r/18b6ae205e75f087aa4a2a05c81ea8b66d8d88dc.1644947594.git.leonro@nvidia.comSigned-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

XRC INI QPs should be able to adjust their local ACK timeout.

Fixes: 2c1619ed ("IB/cma: Define option to set ack timeout and pack tos_set")
Link: https://lore.kernel.org/r/1644421175-31943-1-git-send-email-haakon.bugge@oracle.comSigned-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Suggested-by: Avneesh Pant <avneesh.pant@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Finish removing mcg from rxe pools. Replace rxe pools ref counting by
kref's. Replace rxe_alloc by kzalloc.

Link: https://lore.kernel.org/r/20220208211644.123457-8-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Now that rxe_mcast.c has it's own red-black tree support there is no
longer any requirement for key'ed objects in rxe pools. This patch removes
the key APIs and related code.

Link: https://lore.kernel.org/r/20220208211644.123457-7-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Continuing to decouple mcg from rxe pools. Create red-black tree code in
rxe_mcast.c to hold mcg index. Replace pool key calls by calls to local
red-black routines.

Link: https://lore.kernel.org/r/20220208211644.123457-6-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Replace int num_qp in struct rxe_mcg by atomic_t qp_num.

Link: https://lore.kernel.org/r/20220208211644.123457-5-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Replace 'grp' by 'mcg', 'mce' by 'mca'.  Shorten subroutine names in
rxe_mcast.c.  These name uses are more in line with other object names
used.

Link: https://lore.kernel.org/r/20220208211644.123457-4-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Remove rxe_mca (was rxe_mc_elem) from rxe pools and use kzmalloc and kfree
to allocate and free in rxe_mcast.c. Call kzalloc outside of spinlocks to
avoid having to use GFP_ATOMIC.

Link: https://lore.kernel.org/r/20220208211644.123457-3-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Replace mcg->mcg_lock and mc_grp_pool->pool_lock by rxe->mcg_lock.  This
is the first step of several intended to decouple the mc_grp and mc_elem
objects from the rxe pool code.

Link: https://lore.kernel.org/r/20220208211644.123457-2-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

A previous patch replaced all irqsave locks in rxe with bh locks.  This
ran into problems because rdmacm has a bad habit of calling rdma verbs
APIs while disabling irqs. This is not allowed during spin_unlock_bh()
causing programs that use rdmacm to fail.  This patch reverts the changes
to locks that had this problem or got dragged into the same mess. After
this patch blktests/check -q srp now runs correctly.

Link: https://lore.kernel.org/r/20220215194448.44369-1-rpearsonhpe@gmail.com
Fixes: 21adfa7a ("RDMA/rxe: Replace irqsave locks with bh locks")
Reported-by: Guoqing Jiang <guoqing.jiang@linux.dev>
Reported-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Tested-by: Bart Van Assche <bvanassche@acm.org>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

drivers/infiniband/hw/hfi1/affinity.c code calls cpumask_weight() to check
if any bit of a given cpumask is set. We can do it more efficiently with
cpumask_empty() because cpumask_empty() stops traversing the cpumask as
soon as it finds first set bit, while cpumask_weight() counts all bits
unconditionally.

Link: https://lore.kernel.org/r/20220210224933.379149-20-yury.norov@gmail.comSigned-off-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Shiraz Saleem says:

====================
Add support for DSCP QoS

git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/linux iwl-next
====================

Link: https://lore.kernel.org/r/20220207235921.1303522-1-anthony.l.nguyen@intel.com/Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Add DSCP support for the Intel Ethernet 800 Series devices.  Setup VSI
DSCP info when PCI driver indicates DSCP mode during driver probe or as
notification event.

Link: https://lore.kernel.org/r/20220202191921.1638-4-shiraz.saleem@intel.comSigned-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Rename dcb flag to dcb_vlan_mode in irdma_device struct.  Add a new helper
function, irdma_set_qos_info, to set the VSI QoS information passed by the
PCI driver.

Link: https://lore.kernel.org/r/20220202191921.1638-3-shiraz.saleem@intel.comSigned-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

It's wrong to check the last packet by RXE_COMP_MASK because the flag is
to indicate if responder needs to generate a completion.

Fixes: 9fcd67d1 ("IB/rxe: increment msn only when completing a request")
Fixes: 8700e3e7 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20211229034438.1854908-1-yangx.jy@fujitsu.comSigned-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

The ice driver provides QoS information to auxiliary drivers
through the exported function ice_get_qos_params.  This function
doesn't currently support L3 DSCP QoS.

Add the necessary defines, structure elements and code to support
DSCP QoS through the IIDC functions.
Signed-off-by: Dave Ertman <david.m.ertman@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

Variable nreq is being assigned a value that is never read. The assignment
is redundant and can be removed.

Link: https://lore.kernel.org/r/20220130225747.8114-1-colin.i.king@gmail.comSigned-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Since it is no longer required to cleanup attachments to multicast
groups when a QP is destroyed qp->grp_lock and qp->grp_list are
no longer needed and are removed.

Link: https://lore.kernel.org/r/20220127213755.31697-7-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

With o10-2.2.3 enforced rxe_drop_all_mcast_groups is completely
unnecessary. Remove it and references to it.

Link: https://lore.kernel.org/r/20220127213755.31697-6-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Add code to check if a QP is attached to one or more multicast groups
when destroy_qp is called and return an error if so.

Link: https://lore.kernel.org/r/20220127213755.31697-5-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Rename rxe_mc_grp to rxe_mcg. Rename rxe_mc_elem to rxe_mca.
These can be read 'multicast group' and 'multicast attachment'.
'elem' collided with the use of elem in rxe pools and was a little
confusing.

Link: https://lore.kernel.org/r/20220127213755.31697-4-rpearsonhpe@gmail.comSigned-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>