The PD allocations in IB/core allows us to simplify drivers and their
error flows in their .alloc_pd() paths. The changes in .alloc_pd() go hand
in had with relevant update in .dealloc_pd().

We will use this opportunity and convert .dealloc_pd() to don't fail, as
it was suggested a long time ago, failures are not happening as we have
never seen a WARN_ON print.
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

All callers to ib_alloc_device() provide a larger size than struct
ib_device and rely on the fact that struct ib_device is embedded in their
driver specific structure as the first member.

Provide a safer variant of ib_alloc_device() that checks and enforces this
approach to make sure the drivers are using it right.
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The commit cited below replaced rdma_create_ah with
mlx4_ib_create_slave_ah when creating AHs for the paravirtualized special
QPs.

However, this change also required replacing rdma_destroy_ah with
mlx4_ib_destroy_ah in the affected flows.

The commit missed 3 places where rdma_destroy_ah should have been replaced
with mlx4_ib_destroy_ah.

As a result, the pd usecount was decremented when the ah was destroyed --
although the usecount was NOT incremented when the ah was created.

This caused the pd usecount to become negative, and resulted in the
WARN_ON stack trace below when the mlx4_ib.ko module was unloaded:

WARNING: CPU: 3 PID: 25303 at drivers/infiniband/core/verbs.c:329 ib_dealloc_pd+0x6d/0x80 [ib_core]
Modules linked in: rdma_ucm rdma_cm iw_cm ib_cm ib_umad mlx4_ib(-) ib_uverbs ib_core mlx4_en mlx4_core nfsv3 nfs fscache configfs xt_conntrack nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 libcrc32c ipt_REJECT nf_reject_ipv4 tun ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter bridge stp llc dm_mirror dm_region_hash dm_log dm_mod dax rndis_wlan rndis_host coretemp kvm_intel cdc_ether kvm usbnet iTCO_wdt iTCO_vendor_support cfg80211 irqbypass lpc_ich ipmi_si i2c_i801 mii pcspkr i2c_core mfd_core ipmi_devintf i7core_edac ipmi_msghandler ioatdma pcc_cpufreq dca acpi_cpufreq nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables ext4 mbcache jbd2 sr_mod cdrom ata_generic pata_acpi mptsas scsi_transport_sas mptscsih crc32c_intel ata_piix bnx2 mptbase ipv6 crc_ccitt autofs4 [last unloaded: mlx4_core]
CPU: 3 PID: 25303 Comm: modprobe Tainted: G        W I       5.0.0-rc1-net-mlx4+ #1
Hardware name: IBM  -[7148ZV6]-/Node 1, System Card, BIOS -[MLE170CUS-1.70]- 09/23/2011
RIP: 0010:ib_dealloc_pd+0x6d/0x80 [ib_core]
Code: 00 00 85 c0 75 02 5b c3 80 3d aa 87 03 00 00 75 f5 48 c7 c7 88 d7 8f a0 31 c0 c6 05 98 87 03 00 01 e8 07 4c 79 e0 0f 0b 5b c3 <0f> 0b eb be 0f 0b eb ab 90 66 2e 0f 1f 84 00 00 00 00 00 66 66 66
RSP: 0018:ffffc90005347e30 EFLAGS: 00010282
RAX: 00000000ffffffea RBX: ffff8888589e9540 RCX: 0000000000000006
RDX: 0000000000000006 RSI: ffff88885d57ad40 RDI: 0000000000000000
RBP: ffff88885b029c00 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000004 R12: ffff8887f06c0000
R13: ffff8887f06c13e8 R14: 0000000000000000 R15: 0000000000000000
FS:  00007fd6743c6740(0000) GS:ffff88887fcc0000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000ed1038 CR3: 00000007e3156000 CR4: 00000000000006e0
Call Trace:
 mlx4_ib_close_sriov+0x125/0x180 [mlx4_ib]
 mlx4_ib_remove+0x57/0x1f0 [mlx4_ib]
 mlx4_remove_device+0x92/0xa0 [mlx4_core]
 mlx4_unregister_interface+0x39/0x90 [mlx4_core]
 mlx4_ib_cleanup+0xc/0xd7 [mlx4_ib]
 __x64_sys_delete_module+0x17d/0x290
 ? trace_hardirqs_off_thunk+0x1a/0x1c
 ? do_syscall_64+0x12/0x180
 do_syscall_64+0x4a/0x180
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Fixes: 5e62d5ff ("IB/mlx4: Create slave AH's directly")
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Introduce and use rdma_device_to_ibdev() API for those drivers which are
registering one sysfs group and also use in ib_core.

In subsequent patch, device->provider_ibdev one-to-one mapping is no
longer holds true during accessing sysfs entries.
Therefore, introduce an API rdma_device_to_ibdev() that provides such
information.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Most provider routines are callback routines which ib core invokes.
_callback suffix doesn't convey information about when such callback is
invoked. Therefore, rename port_callback to init_port.

Additionally, store the init_port function pointer in ib_device_ops, so
that it can be accessed in subsequent patches when binding rdma device to
net namespace.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

As part of an audit process to update drivers to use rdma_restrack_add()
ensure that CQ objects is cleared before access.
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

ib_umem_get() can only be called in a method callback, which always has a
udata parameter. This allows ib_umem_get() to derive the ucontext pointer
directly from the udata without requiring the drivers to find it in some
way or another.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NShamir Rabinovitch <shamir.rabinovitch@oracle.com>

The next patch will add dependency from ib_umem_get in to ib_uverbs so
move the required ib_umem_xxx functionality to it's correct module -
ib_uverbs - and avoid circular dependecy from the form of ib_core ->
ib_uverbs -> ib_core in depmod.

Since this now requires all drivers to be build modular if uverbs is
modular, hoist the test a couple drivers had into the main kconfig and
apply it to all drivers uniformly.
Signed-off-by: NShamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/infiniband/hw/mlx4/qp.c: In function '_mlx4_ib_destroy_qp':
drivers/infiniband/hw/mlx4/qp.c:1612:22: warning:
 variable 'pd' set but not used [-Wunused-but-set-variable]

Fixes: e00b64f7 ("RDMA: Cleanup undesired pd->uobject usage")
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Introduce a 'flags' field to destroy address handle callback and add a
flag that marks whether the callback is executed in an atomic context or
not.

This will allow drivers to wait for completion instead of polling for it
when it is allowed.
Signed-off-by: NGal Pressman <galpress@amazon.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Introduce a 'flags' field to create address handle callback and add a flag
that marks whether the callback is executed in an atomic context or not.

This will allow drivers to wait for completion instead of polling for it
when it is allowed.
Signed-off-by: NGal Pressman <galpress@amazon.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Drivers should be using udata to determine if a method is invoked from
user space or kernel space. A pd does not necessarily say a different
objects is kernel or user.

Transforming the tests to use udata eliminates a large number of uobject
references from the drivers.
Signed-off-by: NShamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The macro MLX4_IB_SQ_HEADROOM calculates the spare room needed to be
left. Use it instead of hard-coding the HW prefetch size.
Signed-off-by: NYuval Shaia <yuval.shaia@oracle.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Make all the required change to start use the ib_device_ops structure.
Signed-off-by: NKamal Heib <kamalheib1@gmail.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Initialize ib_device_ops with the supported operations using
ib_set_device_ops().
Signed-off-by: NKamal Heib <kamalheib1@gmail.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

NULL check for kfree is unnecessary, remove it.

Fixes: b42dde47 ("IB/mlx4: Rework special QP creation error path")
Signed-off-by: NYuval Shaia <yuval.shaia@oracle.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

This fixes a compilation warning in sysfs.c

drivers/infiniband/hw/mlx4/sysfs.c:360:2: warning: 'strncpy' output may be
truncated copying 8 bytes from a string of length 31
[-Wstringop-truncation]

By eliminating the temporary stack buffer.
Signed-off-by: NQian Cai <cai@gmx.us>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Perform CQ initialization in the driver when the capability is supported
by the FW.  When passing the CQ to HW indicate that the CQ buffer has
been pre-initialized.

Doing so decreases CQ creation time.  Testing on P8 showed a single 2048
entry CQ creation time was reduced from ~395us to ~170us, which is
2.3x faster.
Signed-off-by: NDaniel Jurgens <danielj@mellanox.com>
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The mlx4 driver does not trigger an IB_EVENT_PORT_ACTIVE when the RoCE
network interface is activated. When SMC determines the RoCE device port
to be used, it checks the port states. This patch triggers IB events for
NETDEV_UP and NETDEV_DOWN.
Signed-off-by: NUrsula Braun <ubraun@linux.ibm.com>
Acked-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Use rdma_set_device_sysfs_group() to register device attributes and
simplify the driver.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Add said information and make the debug print format consistent.
Signed-off-by: NHåkon Bugge <haakon.bugge@oracle.com>
Acked-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

IB Subnet Management Packets (SMPs) were excluded from debug prints.

Fixed by enabling print even on QP0 MADs.
Signed-off-by: NHåkon Bugge <haakon.bugge@oracle.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

iov sysfs tree is created under ib device at
/sys/class/infiniband/mlx4_0/iov.
And,
ibdev->ports_parent->parent = &ibdev->dev.

Therefore, refer to device's kobject directly instead of
indirect access to it.

Additionally, iov entries are created under device kobject and deleted
before device is removed. There is no need to hold additional reference
to device kobject in provider driver.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

The ll parameter is not used in ib_modify_qp_is_ok(), so remove it.
Signed-off-by: NKamal Heib <kamalheib1@gmail.com>
Reviewed-by: NDennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The current code has two copies of the device name, ibdev->dev and
dev_name(&ibdev->dev), and they are setup at different times, which is
very confusing.

Set them both up at the same time and make dev_name() the lead name, which
is the proper use of the driver core APIs. To make it very clear that the
name is not valid until registration pass it in to the
ib_register_device() call rather than messing with ibdev->name directly.

Also the reorganization now checks that dev_name is unique even if it does
not contain a %.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Acked-by: NAdit Ranadive <aditr@vmware.com>
Reviewed-by: NSteve Wise <swise@opengridcomputing.com>
Acked-by: NDevesh Sharma <devesh.sharma@broadcom.com>
Reviewed-by: NShiraz Saleem <shiraz.saleem@intel.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Reviewed-by: NDennis Dalessandro <dennis.dalessandro@intel.com>
Reviewed-by: NMichael J. Ruhl <michael.j.ruhl@intel.com>

The mlx4 driver produces a link error when it is configured
as built-in while CONFIG_INFINIBAND_USER_ACCESS is set to =m:

drivers/infiniband/hw/mlx4/main.o: In function `mlx4_ib_mmap':
main.c:(.text+0x1af4): undefined reference to `rdma_user_mmap_io'

The same function is called from mlx5, which already has a
dependency to ensure we can call it, and from hns, which
appears to suffer from the same problem.

This adds the same dependency that mlx5 uses to the other two.

Fixes: 6745d356 ("RDMA/hns: Use rdma_user_mmap_io")
Fixes: c282da41 ("RDMA/mlx4: Use rdma_user_mmap_io")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Clang warns when more than one set of parentheses are used in single
conditional statements.

drivers/infiniband/hw/mlx4/mcg.c:676:16: warning: equality comparison
with extraneous parentheses [-Wparentheses-equality]
                        if ((method == IB_MGMT_METHOD_GET_RESP)) {
                             ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/infiniband/hw/mlx4/mcg.c:676:16: note: remove extraneous
parentheses around the comparison to silence this warning
                        if ((method == IB_MGMT_METHOD_GET_RESP)) {
                            ~       ^                         ~
drivers/infiniband/hw/mlx4/mcg.c:676:16: note: use '=' to turn this
equality comparison into an assignment
                        if ((method == IB_MGMT_METHOD_GET_RESP)) {
                                    ^~
                                    =

Remove the unnecessary parentheses to silence this warning.
Reported-by: NNick Desaulniers <ndesaulniers@google.com>
Signed-off-by: NNathan Chancellor <natechancellor@gmail.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Rely on the new core code helper to map BAR memory from the driver.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

In calculating the global maximum number of the Scatter/Gather elements
supported, the following four maximum parameters must be taken into
consideration: max_sg_rq, max_sg_sq, max_desc_sz_rq and max_desc_sz_sq.

However instead of bringing this complexity to query_device, which still
won't be sufficient anyway (the calculations are dependent on QP type),
the safer approach will be to restore old code, which will give us 32
SGEs.

Fixes: 33023fb8 ("IB/core: add max_send_sge and max_recv_sge attributes")
Reported-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

In the current implementation, the driver tries to allocate contiguous
memory, and if it fails, it falls back to 4K fragmented allocation.

Once the memory is fragmented, the first allocation might take a lot
of time, and even fail, which can cause connection failures.

This patch changes the logic to always allocate with 4K granularity,
since it's more robust and more likely to succeed.

This patch was tested with Lustre and no performance degradation
was observed.

Note: This commit eliminates the "shrinking WQE" feature. This feature
depended on using vmap to create a virtually contiguous send WQ.
vmap use was abandoned due to problems with several processors (see the
commit cited below). As a result, shrinking WQE was available only with
physically contiguous send WQs. Allocating such send WQs caused the
problems described above.
Therefore, as a side effect of eliminating the use of large physically
contiguous send WQs, the shrinking WQE feature became unavailable.

Warning example:
worker/20:1: page allocation failure: order:8, mode:0x80d0
CPU: 20 PID: 513 Comm: kworker/20:1 Tainted: G OE ------------
Workqueue: ib_cm cm_work_handler [ib_cm]
Call Trace:
[<ffffffff81686d81>] dump_stack+0x19/0x1b
[<ffffffff81186160>] warn_alloc_failed+0x110/0x180
[<ffffffff8118a954>] __alloc_pages_nodemask+0x9b4/0xba0
[<ffffffff811ce868>] alloc_pages_current+0x98/0x110
[<ffffffff81184fae>] __get_free_pages+0xe/0x50
[<ffffffff8133f6fe>] swiotlb_alloc_coherent+0x5e/0x150
[<ffffffff81062551>] x86_swiotlb_alloc_coherent+0x41/0x50
[<ffffffffa056b4c4>] mlx4_buf_direct_alloc.isra.7+0xc4/0x180 [mlx4_core]
[<ffffffffa056b73b>] mlx4_buf_alloc+0x1bb/0x260 [mlx4_core]
[<ffffffffa0b15496>] create_qp_common+0x536/0x1000 [mlx4_ib]
[<ffffffff811c6ef7>] ? dma_pool_free+0xa7/0xd0
[<ffffffffa0b163c1>] mlx4_ib_create_qp+0x3b1/0xdc0 [mlx4_ib]
[<ffffffffa0b01bc2>] ? mlx4_ib_create_cq+0x2d2/0x430 [mlx4_ib]
[<ffffffffa0b21f20>] mlx4_ib_create_qp_wrp+0x10/0x20 [mlx4_ib]
[<ffffffffa08f152a>] ib_create_qp+0x7a/0x2f0 [ib_core]
[<ffffffffa06205d4>] rdma_create_qp+0x34/0xb0 [rdma_cm]
[<ffffffffa08275c9>] kiblnd_create_conn+0xbf9/0x1950 [ko2iblnd]
[<ffffffffa074077a>] ? cfs_percpt_unlock+0x1a/0xb0 [libcfs]
[<ffffffffa0835519>] kiblnd_passive_connect+0xa99/0x18c0 [ko2iblnd]

Fixes: 73898db0 ("net/mlx4: Avoid wrong virtual mappings")
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Since neither ib_post_send() nor ib_post_recv() modify the data structure
their second argument points at, declare that argument const. This change
makes it necessary to declare the 'bad_wr' argument const too and also to
modify all ULPs that call ib_post_send(), ib_post_recv() or
ib_post_srq_recv(). This patch does not change any functionality but makes
it possible for the compiler to verify whether the
ib_post_(send|recv|srq_recv) really do not modify the posted work request.

To make this possible, only one cast had to be introduce that casts away
constness, namely in rpcrdma_post_recvs(). The only way I can think of to
avoid that cast is to introduce an additional loop in that function or to
change the data type of bad_wr from struct ib_recv_wr ** into int
(an index that refers to an element in the work request list). However,
both approaches would require even more extensive changes than this
patch.
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

When posting a send work request, the work request that is posted is not
modified by any of the RDMA drivers. Make this explicit by constifying
most ib_send_wr pointers in RDMA transport drivers.
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: NSagi Grimberg <sagi@grimberg.me>
Reviewed-by: NSteve Wise <swise@opengridcomputing.com>
Reviewed-by: NDennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The internal flag IP_BASED_GIDS was added to a field that was being used
to hold the port Info CapabilityMask without considering the effects this
will have. Since most drivers just use the value from the HW MAD it means
IP_BASED_GIDS will also become set on any HW that sets the IBA flag
IsOtherLocalChangesNoticeSupported - which is not intended.

Fix this by keeping port_cap_flags only for the IBA CapabilityMask value
and store unrelated flags externally. Move the bit definitions for this to
ib_mad.h to make it clear what is happening.

To keep the uAPI unchanged define a new set of flags in the uapi header
that are only used by ib_uverbs_query_port_resp.port_cap_flags which match
the current flags supported in rdma-core, and the values exposed by the
current kernel.

Fixes: b4a26a27 ("IB: Report using RoCE IP based gids in port caps")
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

rdma_ah_find_type() can reach into ib_device->port_immutable with a
potentially out-of-bounds port number, so check that the port number is
valid first.

Fixes: 44c58487 ("IB/core: Define 'ib' and 'roce' rdma_ah_attr types")
Signed-off-by: NTarick Bedeir <tarick@google.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Since slave GID's do not exist in the core gid table we can no longer use
the core code to help do this without creating inconsistencies. Directly
create the AH using mlx4 internal APIs.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Reviewed-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

This patch follows the logic from ib_core but considers the internal
device state upon executing the involved commands.

Specifically, Upon internal error state modify QP to an error state can
be assumed to be success as each in-progress WR going to be flushed in
error in any case as expected by that modify command.

In addition,
As the drain should never fail the driver makes sure that post_send/recv
will succeed even if the device is already in an internal error state.
As such once the driver will supply the simulated/SW CQEs the CQE for
the drain WR will be handled as well.

In case of an internal error state the CQE for the drain WR may be
completed as part of the main task that handled the error state or by
the task that issued the drain WR.

As the above depends on scheduling the code takes the relevant locks
and actions to make sure that the completion handler for that WR will
always be called after that the post_send/recv were issued but not in
parallel to the other task that handles the error flow.
Signed-off-by: NYishai Hadas <yishaih@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

This patch replaces the ib_device_attr.max_sge with max_send_sge and
max_recv_sge. It allows ulps to take advantage of devices that have very
different send and recv sge depths.  For example cxgb4 has a max_recv_sge
of 4, yet a max_send_sge of 16.  Splitting out these attributes allows
much more efficient use of the SQ for cxgb4 with ulps that use the RDMA_RW
API. Consider a large RDMA WRITE that has 16 scattergather entries.
With max_sge of 4, the ulp would send 4 WRITE WRs, but with max_sge of
16, it can be done with 1 WRITE WR.
Acked-by: NSagi Grimberg <sagi@grimberg.me>
Acked-by: NChristoph Hellwig <hch@lst.de>
Acked-by: NSelvin Xavier <selvin.xavier@broadcom.com>
Acked-by: NShiraz Saleem <shiraz.saleem@intel.com>
Acked-by: NDennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

For UD the drivers were doing a sgid_index lookup into the cache to get
the attrs, however we can now directly access the same attrs stores in
the ib_ah instead and remove the lookup.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

While converting GID index from attribute to that of the HCA, GID
attribute is available from the ah_attr. Make use of GID attribute
to simplify the code and also avoid avoid GID query.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

The core code now ensures that all driver callbacks that receive an
rdma_ah_attrs will have a sgid_attr's pointer if there is a GRH present.

Drivers can use this pointer instead of calling a query function with
sgid_index. This simplifies the drivers and also avoids races where a
gid_index lookup may return different data if it is changed.
Signed-off-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>