mlx5 has an ugly flow where it tries to allocate a new MR and replace the
existing MR in the same memory during rereg. This is very complicated and
buggy. Instead of trying to replace in-place inside the driver, provide
support from uverbs to change the entire HW object assigned to a handle
during rereg_mr.
Since destroying a MR is allowed to fail (ie if a MW is pointing at it)
and can't be detected in advance, the algorithm creates a completely new
uobject to hold the new MR and swaps the IDR entries of the two objects.
The old MR in the temporary IDR entry is destroyed, and if it fails
rereg_mr succeeds and destruction is deferred to FD release. This
complexity is why this cannot live in a driver safely.
Link: https://lore.kernel.org/r/20201130075839.278575-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
All FD object destroy implementations return 0, so declare this callback
void.
Link: https://lore.kernel.org/r/20201104144556.3809085-3-leon@kernel.org
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The header files in RDMA subsystem are dual licensed and can be
described by simple SPDX tag, so replace all of them at once
together with making them use the same coding style for header
guard defines.
Link: https://lore.kernel.org/r/20200719072521.135260-1-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
While creating a uobject every create reaches a point where the uobject is
fully initialized. For ioctls that go on to copy_to_user this means they
need to open code the destruction of a fully created uobject - ie the
RDMA_REMOVE_DESTROY sort of flow.
Open coding this creates bugs, eg the CQ does not properly flush the
events list when it does its error unwind.
Provide a uverbs_finalize_uobj_create() function which indicates that the
uobject is fully initialized and that abort should call to destroy_hw to
destroy the uobj->object and related.
Methods can call this function if they go on to have error cases after
setting uobj->object. Once done those error cases can simply do return,
without an error unwind.
Link: https://lore.kernel.org/r/20200519072711.257271-2-leon@kernel.org
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Now that all callers provide a non-NULL attrs the ufile is redundant.
Adjust things so that the context handling is done inside alloc_uobj,
and the ib_uverbs_get_ucontext_file() is avoided if we already have the
context.
Link: https://lore.kernel.org/r/1578504126-9400-13-git-send-email-yishaih@mellanox.com
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This is a left over from an earlier version that creates a lot of
complexity for error unwind, particularly for FD uobjects.
The only reason this was done is so that anon_inode_get_file() could be
called with the final fops and a fully setup uobject. Both need to be
setup since unwinding anon_inode_get_file() via fput will call the
driver's release().
Now that the driver does not provide release, we no longer need to worry
about this complicated sequence, simply create the struct file at the
start and allow the core code's release function to deal with the abort
case.
This allows all the confusing error paths around commit to be removed.
Link: https://lore.kernel.org/r/1578504126-9400-5-git-send-email-yishaih@mellanox.com
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
FD uobjects have a weird split between the struct file and uobject
world. Simplify this to make them pure uobjects and use a generic release
method for all struct file operations.
This fixes the control flow so that mlx5_cmd_cleanup_async_ctx() is always
called before erasing the linked list contents to make the concurrancy
simpler to understand.
For this to work the uobject destruction must fence anything that it is
cleaning up - the design must not rely on struct file lifetime.
Only deliver_event() relies on the struct file to when adding new events
to the queue, add a is_destroyed check under lock to block it.
Link: https://lore.kernel.org/r/1578504126-9400-3-git-send-email-yishaih@mellanox.com
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
dispatch_event_fd() runs from a notifier with minimal locking, and relies
on RCU and a file refcount to keep the uobject and eventfd alive.
As the next patch wants to remove the file_operations release function
from the drivers, re-organize things so that the devx_event_notifier()
path uses the existing RCU to manage the lifetime of the uobject and
eventfd.
Move the refcount puts to a call_rcu so that the objects are guaranteed to
exist and remove the indirect file refcount.
Link: https://lore.kernel.org/r/1578504126-9400-2-git-send-email-yishaih@mellanox.com
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
After device disassociation the uapi_objects are destroyed and freed,
however it is still possible that core code can be holding a kref on the
uobject. When it finally goes to uverbs_uobject_free() via the kref_put()
it can trigger a use-after-free on the uapi_object.
Since needs_kfree_rcu is a micro optimization that only benefits file
uobjects, just get rid of it. There is no harm in using kfree_rcu even if
it isn't required, and the number of involved objects is small.
Link: https://lore.kernel.org/r/20200113143306.GA28717@ziepe.ca
Signed-off-by: Michael Guralnik <michaelgur@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Pass uverbs_attr_bundle down the uobject destroy path. The next patch will
use this to eliminate the dependecy of the drivers in ib_x->uobject
pointers.
Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
the Attempt to use the below commit to initialize the ucontext for the
uobject destroy path has shown that the below commit is incomplete.
Parts were reverted and the ucontext set up in the uverbs_attr_bundle was
moved to rdma_lookup_get_uobject which is called from the uobj_get_XXX
macros and rdma_alloc_begin_uobject which is called when uobject is
created.
Fixes: 3d9dfd0603 ("IB/uverbs: Add ib_ucontext to uverbs_attr_bundle sent from ioctl and cmd flows")
Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Introduce MLX5_IB_OBJECT_DEVX_ASYNC_CMD_FD and its initial implementation.
This object is from type class FD and will be used to read DEVX async
commands completion.
The core layer should allow the driver to set object from type FD in a
safe mode, this option was added with a matching comment in place.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Currently the struct uverbs_obj_type stored in the ib_uobject is part of
the .rodata segment of the module that defines the object. This is a
problem if drivers define new uapi objects as we will be left with a
dangling pointer after device disassociation.
Switch the uverbs_obj_type for struct uverbs_api_object, which is
allocated memory that is part of the uverbs_api and is guaranteed to
always exist. Further this moves the 'type_class' into this memory which
means access to the IDR/FD function pointers is also guaranteed. Drivers
cannot define new types.
This makes it safe to continue to use all uobjects, including driver
defined ones, after disassociation.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The disassociate function was broken by design because it failed all
commands. This prevents userspace from calling destroy on a uobject after
it has detected a device fatal error and thus reclaiming the resources in
userspace is prevented.
This fix is now straightforward, when anything destroys a uobject that is
not the user the object remains on the IDR with a NULL context and object
pointer. All lookup locking modes other than DESTROY will fail. When the
user ultimately calls the destroy function it is simply dropped from the
IDR while any related information is returned.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
After all the recent structural changes this is now straightfoward, hoist
the hw_destroy_rwsem up out of rdma_destroy_explicit and wrap it around
the uobject write lock as well as the destroy.
This is necessary as obtaining a write lock concurrently with
uverbs_destroy_ufile_hw() will cause malfunction.
After this change none of the destroy callbacks require the
disassociate_srcu lock to be correct.
This requires introducing a new lookup mode, UVERBS_LOOKUP_DESTROY as the
IOCTL interface needs to hold an unlocked kref until all command
verification is completed.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
There are several flows that can destroy a uobject and each one is
minimized and sprinkled throughout the code base, making it difficult to
understand and very hard to modify the destroy path.
Consolidate all of these into uverbs_destroy_uobject() and call it in all
cases where a uobject has to be destroyed.
This makes one change to the lifecycle, during any abort (eg when
alloc_commit is not called) we always call out to alloc_abort, even if
remove_commit needs to be called to delete a HW object.
This also renames RDMA_REMOVE_DURING_CLEANUP to RDMA_REMOVE_ABORT to
clarify its actual usage and revises some of the comments to reflect what
the life cycle is for the type implementation.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The ridiculous dance with uobj_remove_commit() is not needed, the write
path can follow the same flow as ioctl - lock and destroy the HW object
then use the data left over in the uobject to form the response to
userspace.
Two helpers are introduced to make this flow straightforward for the
caller.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Allocating the struct file during alloc_begin creates this strange
asymmetry with IDR, where the FD has two krefs pointing at it during the
pre-commit phase. In particular this makes the abort process for FD very
strange and confusing.
For instance abort currently calls the type's destroy_object twice, and
the fops release once if abort is done. This is very counter intuitive. No
fops should be called until alloc_commit succeeds, and destroy_object
should only ever be called once.
Moving the struct file allocation to the alloc_commit is now simple, as we
already support failure of rdma_alloc_commit_uobject, with all the
required rollback pieces.
This creates an understandable symmetry with IDR and simplifies/fixes the
abort handling for FD types.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The ioctl framework already does this correctly, but the write path did
not. This is trivially fixed by simply using a standard pattern to return
uobj_alloc_commit() as the last statement in every function.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Our ABI for write() uses a s32 for FDs and a u32 for IDRs, but internally
we ended up implicitly casting these ABI values into an 'int'. For ioctl()
we use a s64 for FDs and a u64 for IDRs, again casting to an int.
The various casts to int are all missing range checks which can cause
userspace values that should be considered invalid to be accepted.
Fix this by making the generic lookup routine accept a s64, which does not
truncate the write API's u32/s32 or the ioctl API's s64. Then push the
detailed range checking down to the actual type implementations to be
shared by both interfaces.
Finally, change the copy of the uobj->id to sign extend into a s64, so eg,
if we ever wish to return a negative value for a FD it is carried
properly.
This ensures that userspace values are never weirdly interpreted due to
the various trunctations and everything that is really out of range gets
an EINVAL.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The only purpose for this structure was to hold the ib_uobject_file
pointer, but now that is part of the standard ib_uobject the structure
no longer makes any sense, so get rid of it.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
The correct handle to refer to the idr/etc is ib_uverbs_file, revise all
the core APIs to use this instead. The user API are left as wrappers
that automatically convert a ucontext to a ufile for now.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Improve uverbs_cleanup_ucontext algorithm to work properly when the
topology graph of the objects cannot be determined at compile time. This
is the case with objects created via the devx interface in mlx5.
Typically uverbs objects must be created in a strict topologically sorted
order, so that LIFO ordering will generally cause them to be freed
properly. There are only a few cases (eg memory windows) where objects can
point to things out of the strict LIFO order.
Instead of using an explicit ordering scheme where the HW destroy is not
allowed to fail, go over the list multiple times and allow the destroy
function to fail. If progress halts then a final, desperate, cleanup is
done before leaking the memory. This indicates a driver bug.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When some objects are destroyed, we need to extract their status at
destruction. After object's destruction, this status
(e.g. events_reported) relies in the uobject. In order to have the
latest and correct status, the underlying object should be destroyed,
but we should keep the uobject alive and read this information off the
uobject. We introduce a rdma_explicit_destroy function. This function
destroys the class type object (for example, the IDR class type which
destroys the underlying object as well) and then convert the uobject
to be of a null class type. This uobject will then be destroyed as any
other uobject once uverbs_finalize_object[s] is called.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Switch all uverbs_type_attrs_xxxx with DECLARE_UVERBS_OBJECT
macros. This will be later used in order to embed the object
specific methods in the objects as well.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
We rename the "write" flags to "exclusive", as it's used for both
WRITE and DESTROY actions.
Fixes: 3832125624 ('IB/core: Add support for idr types')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This patch adds the standard fd based type - completion_channel.
The completion_channel is now prefixed with ib_uobject, similarly
to the rest of the uobjects.
This requires a few changes:
(1) We define a new completion channel fd based object type.
(2) completion_event and async_event are now two different types.
This means they use different fops.
(3) We release the completion_channel exactly as we release other
idr based objects.
(4) Since ib_uobjects are already kref-ed, we only add the kref to the
async event.
A fd object requires filling out several parameters. Its op pointer
should point to uverbs_fd_ops and its size should be at least the
size if ib_uobject. We use a macro to make the type declaration
easier.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The completion channel we use in verbs infrastructure is FD based.
Previously, we had a separate way to manage this object. Since we
strive for a single way to manage any kind of object in this
infrastructure, we conceptually treat all objects as subclasses
of ib_uobject.
This commit adds the necessary mechanism to support FD based objects
like their IDR counterparts. FD objects release need to be synchronized
with context release. We use the cleanup_mutex on the uverbs_file for
that.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This patch adds the standard idr based types. These types are
used in downstream patches in order to initialize, destroy and
lookup IB standard objects which are based on idr objects.
An idr object requires filling out several parameters. Its op pointer
should point to uverbs_idr_ops and its size should be at least the
size of ib_uobject. We add a macro to make the type declaration easier.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The new ioctl infrastructure supports driver specific objects.
Each such object type has a hot unplug function, allocation size and
an order of destruction.
When a ucontext is created, a new list is created in this ib_ucontext.
This list contains all objects created under this ib_ucontext.
When a ib_ucontext is destroyed, we traverse this list several time
destroying the various objects by the order mentioned in the object
type description. If few object types have the same destruction order,
they are destroyed in an order opposite to their creation.
Adding an object is done in two parts.
First, an object is allocated and added to idr tree. Then, the
command's handlers (in downstream patches) could work on this object
and fill in its required details.
After a successful command, the commit part is called and the user
objects become ucontext visible. If the handler failed, alloc_abort
should be called.
Removing an uboject is done by calling lookup_get with the write flag
and finalizing it with destroy_commit. A major change from the previous
code is that we actually destroy the kernel object itself in
destroy_commit (rather than just the uobject).
We should make sure idr (per-uverbs-file) and list (per-ucontext) could
be accessed concurrently without corrupting them.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Jason Gunthorpe