Refactor the page fault handler to be more readable and extensible, this
cleanup was triggered by the error reported below. The code structure made
it unclear to the automatic tools to identify that such a flow is not
possible in real life because "requestor != NULL" means that "qp != NULL"
too.

    drivers/infiniband/hw/mlx5/odp.c:1254 mlx5_ib_mr_wqe_pfault_handler()
    error: we previously assumed 'qp' could be null (see line 1230)

Fixes: 08100fad ("IB/mlx5: Add ODP SRQ support")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The InfiniBand Architecture Specification section 10.6.7.2.4 TYPE 2 MEMORY
WINDOWS says that if the CI supports the Base Memory Management Extensions
defined in this specification, the R_Key format for a Type 2 Memory Window
must consist of:

* 24 bit index in the most significant bits of the R_Key, which is owned
  by the CI, and
* 8 bit key in the least significant bits of the R_Key, which is owned by
  the Consumer.

This means that the kernel should compare only the index part of a R_Key
to determine equality with another R_Key.

Fixes: db570d7d ("IB/mlx5: Add ODP support to MW")
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

If page-fault handler spans multiple MRs then the access mask needs to
be reset before each MR handling or otherwise write access will be
granted to mapped pages instead of read-only.

Cc: <stable@vger.kernel.org> # 3.19
Fixes: 7bdf65d4 ("IB/mlx5: Handle page faults")
Reported-by: NJerome Glisse <jglisse@redhat.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The write access of an implicit MR is inherited to all of its children.
Therefore we must set the correct write access to the parent MR.

Pass full access_flags when creating umem to let it calculate write access
correctly.

Fixes: da6a496a ("IB/mlx5: Ranges in implicit ODP MR inherit its write access")
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

When prefetching odp mr it is required to verify that pd of the mr is
identical to the pd for which the advise_mr request arrived with.

This check was missing from synchronous flow and is added now.

Fixes: 813e90b1 ("IB/mlx5: Add advise_mr() support")
Reported-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

When deferring a prefetch request we need to protect against MR or PD
being destroyed while the request is still enqueued.

The first step is to validate that PD owns the lkey that describes the MR
and that the MR that the lkey refers to is owned by that PD.

The second step is to dequeue all requests when MR is destroyed.

Since PD can't be destroyed while it owns MRs it is guaranteed that when a
worker wakes up the request it refers to is still valid.

Now, it is possible to refrain from taking a reference on the device since
it is assured to be present as pd.

While that, replace the dedicated ordered workqueue with the system
unbound workqueue to reuse an existing resource and improve
performance. This will also fix a bug of queueing to the wrong workqueue.

Fixes: 813e90b1 ("IB/mlx5: Add advise_mr() support")
Reported-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Query all per transport caps for XRC and set the appropriate bits in the
per transport field of the advertised struct.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

ODP support in SRQ is per transport capability. Based on device
capabilities set this flag in device structure for future queries.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Add changes to the WQE page-fault handler to

1. Identify that the event is for a SRQ WQE
2. Pass SRQ object instead of a QP to the function that reads the WQE
3. Parse the SRQ WQE with respect to its structure

The rest is handled as for regular RQ WQE.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Reading a WQE from SRQ is almost identical to reading from regular RQ.
The differences are the size of the queue, the size of a WQE and buffer
location.

Make necessary changes to mlx5_ib_read_user_wqe() to let it read a WQE
from a SRQ or RQ by caller choice.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Skip XRC segment in the beginning of a send WQE and fetch ODP XRC
capabilities when QP type is IB_QPT_XRC_INI. The rest of the handling is
the same as in RC QP.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

In the function mlx5_ib_mr_responder_pfault_handler()

1. The parameter wqe is used as read-only so there is no need to pass it
   by reference.
2. Remove the unused argument pfault from list of arguments.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

When handling an ODP event for a receive WQE in SRQ the target QP is
unknown. Therefore, it is wrong to ask if QP has a SRQ in the page-fault
handler.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

QP and SRQ objects are stored in different containers so the action to get
and lock a common resource during ODP event needs to address that.

While here get rid of 'refcount' and 'free' fields in mlx5_core_srq struct
and use the fields with same semantics in common structure.

Fixes: 032080ab ("IB/mlx5: Lock QP during page fault handling")
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NMajd Dibbiny <majd@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

A sub-range in ODP implicit MR should take its write permission from the
MR and not be set always to allow.

Fixes: d07d1d70 ("IB/umem: Update on demand page (ODP) support")
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

This patch avoids that sparse complains about missing function
declarations.

Fixes: c9990ab3 ("RDMA/umem: Move all the ODP related stuff out of ucontext and into per_mm")
Signed-off-by: NBart Van Assche <bvanassche@acm.org>
Acked-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Add support for ODP for DEVX indirection mkey, it includes:
- Recognizing its type as part of the radix tree lookup.
- Use similar flow as done for the MW MKEY type.
Signed-off-by: NYishai Hadas <yishaih@mellanox.com>
Reviewed-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Work must hold a kref on the ib_device otherwise the dev pointer can
become free before the work runs. This can happen because the work is
being pushed onto the system work queue which is not flushed during driver
unregister.

Remove the bogus use of 'reg_state':
 - While in uverbs the reg_state is guaranteed to always be
   REGISTERED
 - Testing reg_state with no locking is bogus. Use ib_device_try_get()
   to get back into a region that prevents unregistration.

For now continue with a flow that is similar to the existing code.

Fixes: 813e90b1 ("IB/mlx5: Add advise_mr() support")
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Reviewed-by: NMoni Shoua <monis@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>

Consolidate various checks if MR is ODP backed to one simple helper and
update call sites to use it.
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

The verb advise_mr() is used to give advice to the kernel about an address
range that belongs to a MR.  Implement the verb and register it on the
device. The current implementation supports the only known advice to date,
prefetch.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Reviewed-by: NGuy Levi <guyle@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Since any page fault may be interrupted by a MMU invalidation and implicit
leaf MR may be released during this process. The check for parent value
is unreliable condition for an implicit MR.
Use other condition that we can rely on to determine if MR is implicit.

Fixes: b4cfe447 ("IB/mlx5: Implement on demand paging by adding support for MMU notifiers")
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Pagefaults occurred in non-ODP MR are completely valid events, so
initialize return variable to 0.

Fixes: 4d5422a3 ("IB/mlx5: Skip non-ODP MR when handling a page fault")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Memory windows are implemented with an indirect MKey, when a page fault
event comes for a MW Mkey we need to find the MR at the end of the list of
the indirect MKeys by iterating on all items from the first to the last.

The offset calculated during this process has to be zeroed after the first
iteration or the next iteration will start from a wrong address, resulting
incorrect ODP faulting behavior.

Fixes: db570d7d ("IB/mlx5: Add ODP support to MW")
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

It is possible that we call pagefault_single_data_segment() with a MKey
that belongs to a memory region which is not on demand (i.e. pinned
pages). This can happen if, for instance, a WQE that points to multiple
MRs where some of them are ODP MRs and some are not.  In this case we
don't need to handle this MR in the ODP context besides reporting success.

Otherwise the code will call pagefault_mr() which will do to_ib_umem_odp()
on a non-ODP MR and thus access out of bounds.

Fixes: 7bdf65d4 ("IB/mlx5: Handle page faults")
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Use the new generic EQ API to move all ODP RDMA data structures and logic
form mlx5 core driver into mlx5_ib driver.
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Reviewed-by: NLeon Romanovsky <leonro@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Acked-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

Add and modify debug messages to ODP related error flows.
In that context, return code EAGAIN is considered less severe and print
level for it is set debug instead of warn.
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

When page fault event for a WQE arrives, the event data contains the
resource (e.g. QP) number which will later be used by the page fault
handler to retrieve the resource. Meanwhile, another context can destroy
the resource and cause use-after-free. To avoid that, take a reference on the
resource when handler starts and release it when it ends.

Page fault events for RDMA operations don't need to be protected because
the driver doesn't need to access the QP in the page fault handler.

Fixes: d9aaed83 ("{net,IB}/mlx5: Refactor page fault handling")
Signed-off-by: NMoni Shoua <monis@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>

Since ODP had a single struct mmu_notifier located in the ucontext it
could only handle a single MM at a time, and this prevented it from using
the new owning_mm system.

With the prior rework it is now simple to let ODP track multiple MMs per
ucontext, finish the job so that the per_mm is allocated on a mm by mm
basis, and freed when the last umem is dropped from the ucontext.

As a side effect the new saner locking removes the lockdep splat about
nesting the umem_rwsem between mmu_notifier_unregister and
ib_umem_odp_release.

It also makes ODP work with multiple processes, across, fork, etc.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

This is the first step to make ODP use the owning_mm that is now part of
struct ib_umem.

Each ODP umem is linked to a single per_mm structure, which in turn, is
linked to a single mm, via the embedded mmu_notifier. This first patch
introduces the structure and reworks eveything to use it.

This also needs to introduce tgid into the ib_ucontext_per_mm, as
get_user_pages_remote() requires the originating task for statistics
tracking.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

This no longer has any use, we can use container_of to get to the
umem_odp, and a simple flag to indicate if this is an odp MR. Remove the
few remaining references to it.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

These two structures are linked together, use the container_of pattern
instead of a double allocation to make the code simpler and easier to
follow.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

All of these functions already require the ODP version of the umem struct,
make this very clear by having the signature require it. This paves the
way to using the container_of() pattern to link umem_odp and umem
together.
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

There are several blockable mmu notifiers which might sleep in
mmu_notifier_invalidate_range_start and that is a problem for the
oom_reaper because it needs to guarantee a forward progress so it cannot
depend on any sleepable locks.

Currently we simply back off and mark an oom victim with blockable mmu
notifiers as done after a short sleep.  That can result in selecting a new
oom victim prematurely because the previous one still hasn't torn its
memory down yet.

We can do much better though.  Even if mmu notifiers use sleepable locks
there is no reason to automatically assume those locks are held.  Moreover
majority of notifiers only care about a portion of the address space and
there is absolutely zero reason to fail when we are unmapping an unrelated
range.  Many notifiers do really block and wait for HW which is harder to
handle and we have to bail out though.

This patch handles the low hanging fruit.
__mmu_notifier_invalidate_range_start gets a blockable flag and callbacks
are not allowed to sleep if the flag is set to false.  This is achieved by
using trylock instead of the sleepable lock for most callbacks and
continue as long as we do not block down the call chain.

I think we can improve that even further because there is a common pattern
to do a range lookup first and then do something about that.  The first
part can be done without a sleeping lock in most cases AFAICS.

The oom_reaper end then simply retries if there is at least one notifier
which couldn't make any progress in !blockable mode.  A retry loop is
already implemented to wait for the mmap_sem and this is basically the
same thing.

The simplest way for driver developers to test this code path is to wrap
userspace code which uses these notifiers into a memcg and set the hard
limit to hit the oom.  This can be done e.g.  after the test faults in all
the mmu notifier managed memory and set the hard limit to something really
small.  Then we are looking for a proper process tear down.

[akpm@linux-foundation.org: coding style fixes]
[akpm@linux-foundation.org: minor code simplification]
Link: http://lkml.kernel.org/r/20180716115058.5559-1-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Acked-by: Christian König <christian.koenig@amd.com> # AMD notifiers
Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx and umem_odp
Reported-by: NDavid Rientjes <rientjes@google.com>
Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Doug Ledford <dledford@redhat.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Mike Marciniszyn <mike.marciniszyn@intel.com>
Cc: Dennis Dalessandro <dennis.dalessandro@intel.com>
Cc: Sudeep Dutt <sudeep.dutt@intel.com>
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Cc: Dimitri Sivanich <sivanich@sgi.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Unconditional locks/list and ODP srcu initialization should be done in
the INIT stage. Remove those from the CAPS stage and move them to the
proper stage.
Signed-off-by: NMark Bloch <markb@mellanox.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Using the ARRAY_SIZE macro improves the readability of the code.

Found with Coccinelle with the following semantic patch:
@r depends on (org || report)@
type T;
T[] E;
position p;
@@
(
 (sizeof(E)@p /sizeof(*E))
|
 (sizeof(E)@p /sizeof(E[...]))
|
 (sizeof(E)@p /sizeof(T))
)
Signed-off-by: NJérémy Lefaure <jeremy.lefaure@lse.epita.fr>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

The extended address vector is the highest bit in be32 variable,
but it was compared with the lowest. This patch fixes the endianness
of that check and removes already declared define.

Fixes: 17d2f88f ("IB/mlx5: Add ODP atomics support")
Reviewed-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

The extended address vector is the highest bit in be32 variable,
but it was compared with the lowest. This patch fixes the endianness
of that check and removes already declared define.

Fixes: 17d2f88f ("IB/mlx5: Add ODP atomics support")
Reviewed-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Internally MW implemented as KLM MKey and filled by userspace UMR
postsends.  Handle pagefault trigered by operations on this MKeys.
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

To make page fault handling code more flexible
split pagefault_single_data_segment() function.
Keep MR resolution in pagefault_single_data_segment() and
move actual updates into pagefault_single_mr().
Signed-off-by: NArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: NLeon Romanovsky <leon@kernel.org>
Signed-off-by: NDoug Ledford <dledford@redhat.com>