Commit graph

424 commits

Author SHA1 Message Date
Jason Gunthorpe
50d5c02ab8 RDMA/nldev: Fix locking when accessing mr->pd
Sashiko points out that, due to rereg_mr, the PD is actually variable and
all the touches in nldev are racy.

Use mr->device instead of mr->pd->device.

Getting the PD restrack ID is more tricky. To avoid disturbing all the
happy paths, add an rdma_restrack_sync() operation which is sort of like
flush_workqueue() or synchronize_irq(): after it returns, all the old
nldev touches to the mr are gone and everything sees the new PD. This
makes it safe to reach into the PD pointer.

Fixes: da5c850782 ("RDMA/nldev: add driver-specific resource tracking")
Link: https://patch.msgid.link/r/4-v1-29ebd2c229b5+fd5-ib_mr_pd_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-06-08 14:32:43 -03:00
Jason Gunthorpe
43b57d73eb RDMA/core: Don't make a dummy ib_udata on the stack in create_qp
Sashiko points out the udata for destruction has to be created using
uverbs_get_cleared_udata(). Move it to ib_core_uverbs.c so that the core
qp code can call it. Rework the call chain to pass the struct
uverbs_attr_bundle right up to the driver op callback.

Fixes a possible wild stack reference in drivers during error unwinding,
mlx5 can call rdma_udata_to_drv_context() from destroy_qp() when
destroying a QP.

Fixes: 00a79d6b99 ("RDMA/core: Configure selinux QP during creation")
Link: https://patch.msgid.link/r/1-v1-922fa8e828ba+f7-ib_udata_stack_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-06-03 15:12:43 -03:00
Jiri Pirko
d7a40b5194 RDMA/uverbs: Expose CoCo DMA bounce requirement to userspace
In CoCo guests, guest memory is encrypted and untrusted (T=0) devices
cannot DMA to it directly; such transfers must go through unencrypted
bounce buffers. RDMA registers user pages for direct device access,
bypassing the DMA layer and thus any bouncing, so registered memory does
not work in this configuration.

Until trusted (T=1) device detection is available, conservatively flag
every device attached to a CoCo guest. Expose the condition to userspace
as IB_UVERBS_DEVICE_CC_DMA_BOUNCE in device_cap_flags_ex so applications
can avoid memory registration and fall back to copying buffers through
send/recv.

Link: https://patch.msgid.link/r/20260517141311.2409230-2-jiri@resnulli.us
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-05-29 20:27:29 -03:00
Jiri Pirko
7c5bbaaf44 RDMA/uverbs: Remove legacy umem field from struct ib_cq
Now that all drivers use helper to get umem and manage the lifetime,
legacy umem field in struct ib_cq is no longer needed. Remove it
along with ib_umem_get_cq_tmp() helper that populated it and both
error and destroy paths.

Link: https://patch.msgid.link/r/20260529134312.2836341-12-jiri@resnulli.us
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-05-29 20:19:59 -03:00
Jason Gunthorpe
9733e9f580 RDMA/core: Move flow related functions to ib_uverbs_support.ko
mlx5 uses these as part of the driver implementation, move them to the
support module instead.

Link: https://patch.msgid.link/r/6-v3-43aba1969751+1988-ib_uverbs_support_ko_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-05-26 10:11:43 -03:00
Michael Guralnik
1f3b337af2 RDMA/core: Fix rereg_mr use-after-free race
When a driver creates a new MR during rereg_user_mr, a race window
exists between rdma_alloc_commit_uobject() for the new MR and the point
where the code reads that MR to populate the response keys.

A concurrent rereg_mr or destroy_mr could destroy the MR in this window
and cause UAF in the first thread.

Racing flow between two rereg_mr calls:

 CPU0                           CPU1
 ----                           ----
 rereg_user_mr(mr_handle)
   uobj_get_write(mr_handle) -> mr0
   mr1 = driver→rereg()
   rdma_alloc_commit_uobject(mr1)
   // mr1 replaced mr0 and is unlocked
   uobj_put_destroy(mr0)
                                rereg_user_mr(mr_handle)
                                  uobj_get_write(mr_handle) -> mr1
                                  mr2 = driver→rereg()
                                  rdma_alloc_commit_uobject(mr2)
                                  // mr2 replaced mr1 and is unlocked
                                  uobj_put_destroy(mr1)
                                  // Destroys mr1!

   resp.lkey = mr1->lkey; // UAF - mr1 was freed!
   resp.rkey = mr1->rkey; // UAF - mr1 was freed!

Fix by storing lkey/rkey in local variables before the new MR is
unlocked and using the local variables to set the user response.

Fixes: 6e0954b11c ("RDMA/uverbs: Allow drivers to create a new HW object during rereg_mr")
Link: https://patch.msgid.link/r/20260427-security-bug-fixes-v3-4-4621fa52de0e@nvidia.com
Signed-off-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-04-29 16:37:12 -03:00
Leon Romanovsky
dc76086a2d RDMA: Properly propagate the number of CQEs as unsigned int
Instead of checking whether the number of CQEs is negative or zero, fix the
.resize_user_cq() declaration to use unsigned int. This better reflects the
expected value range. The sanity check is then handled correctly in ib_uvbers.

Link: https://patch.msgid.link/20260319-resize_cq-cqe-v1-1-b78c6efc1def@nvidia.com
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2026-03-30 13:47:44 -04:00
Leon Romanovsky
ce68351be0 RDMA: Clarify that CQ resize is a user‑space verb
The CQ resize operation is used only by uverbs. Make this explicit.

Link: https://patch.msgid.link/20260318-resize_cq-type-v1-2-b2846ed18846@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2026-03-30 13:47:44 -04:00
Jason Gunthorpe
38a6e5579d RDMA: Use copy_struct_from_user() instead of open coding
This entire function is just open coding copy_struct_from_user(), call it
directly, it is faster anyhow.

Link: https://patch.msgid.link/r/1-v3-bd56dd443069+49-bnxt_re_uapi_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2026-03-08 06:20:24 -04:00
Leon Romanovsky
a291758288 RDMA/core: Reject zero CQE count
All drivers already ensure that the number of CQEs is at least 1.
Add this validation to the core so drivers no longer need to repeat it.
Future patches converting to the .create_user_cq() interface will remove
the per‑driver checks.

Link: https://patch.msgid.link/20260213-refactor-umem-v1-8-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2026-02-25 08:15:30 -05:00
Leon Romanovsky
584ec74748 RDMA/core: Prepare create CQ path for API unification
Ensure that .create_cq_umem() and .create_cq() follow the same API
contract, allowing drivers to be gradually migrated to the umem-aware
CQ management flow.

Link: https://patch.msgid.link/20260213-refactor-umem-v1-7-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2026-02-25 08:15:30 -05:00
Leon Romanovsky
25c7410488 RDMA/core: Manage CQ umem in core code
In the current implementation, CQ umem is handled both by ib_core and
the driver. ib_core sometimes creates and destroys it, while the driver
also destroys it.

Store the umem in struct ib_cq and ensure that only ib_core manages
its lifetime, relying solely on its internal reference counter.

Link: https://patch.msgid.link/20260213-refactor-umem-v1-5-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2026-02-25 08:15:30 -05:00
Linus Torvalds
32a92f8c89 Convert more 'alloc_obj' cases to default GFP_KERNEL arguments
This converts some of the visually simpler cases that have been split
over multiple lines.  I only did the ones that are easy to verify the
resulting diff by having just that final GFP_KERNEL argument on the next
line.

Somebody should probably do a proper coccinelle script for this, but for
me the trivial script actually resulted in an assertion failure in the
middle of the script.  I probably had made it a bit _too_ trivial.

So after fighting that far a while I decided to just do some of the
syntactically simpler cases with variations of the previous 'sed'
scripts.

The more syntactically complex multi-line cases would mostly really want
whitespace cleanup anyway.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2026-02-21 20:03:00 -08:00
Linus Torvalds
bf4afc53b7 Convert 'alloc_obj' family to use the new default GFP_KERNEL argument
This was done entirely with mindless brute force, using

    git grep -l '\<k[vmz]*alloc_objs*(.*, GFP_KERNEL)' |
        xargs sed -i 's/\(alloc_objs*(.*\), GFP_KERNEL)/\1)/'

to convert the new alloc_obj() users that had a simple GFP_KERNEL
argument to just drop that argument.

Note that due to the extreme simplicity of the scripting, any slightly
more complex cases spread over multiple lines would not be triggered:
they definitely exist, but this covers the vast bulk of the cases, and
the resulting diff is also then easier to check automatically.

For the same reason the 'flex' versions will be done as a separate
conversion.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2026-02-21 17:09:51 -08:00
Kees Cook
69050f8d6d treewide: Replace kmalloc with kmalloc_obj for non-scalar types
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:

Single allocations:	kmalloc(sizeof(TYPE), ...)
are replaced with:	kmalloc_obj(TYPE, ...)

Array allocations:	kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with:	kmalloc_objs(TYPE, COUNT, ...)

Flex array allocations:	kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with:	kmalloc_flex(*PTR, FAM, COUNT, ...)

(where TYPE may also be *VAR)

The resulting allocations no longer return "void *", instead returning
"TYPE *".

Signed-off-by: Kees Cook <kees@kernel.org>
2026-02-21 01:02:28 -08:00
Yi Liu
58b604dfc7 RDMA/uverbs: Add __GFP_NOWARN to ib_uverbs_unmarshall_recv() kmalloc
Since wqe_size in ib_uverbs_unmarshall_recv() is user-provided and already
validated, but can still be large, add __GFP_NOWARN to suppress memory
allocation warnings for large sizes, consistent with the similar fix in
ib_uverbs_post_send().

Fixes: 67cdb40ca4 ("[IB] uverbs: Implement more commands")
Signed-off-by: Yi Liu <liuy22@mails.tsinghua.edu.cn>
Link: https://patch.msgid.link/20260129094900.3517706-1-liuy22@mails.tsinghua.edu.cn
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2026-02-01 07:39:14 -05:00
Yi Liu
1956f0a74c RDMA/uverbs: Validate wqe_size before using it in ib_uverbs_post_send
ib_uverbs_post_send() uses cmd.wqe_size from userspace without any
validation before passing it to kmalloc() and using the allocated
buffer as struct ib_uverbs_send_wr.

If a user provides a small wqe_size value (e.g., 1), kmalloc() will
succeed, but subsequent accesses to user_wr->opcode, user_wr->num_sge,
and other fields will read beyond the allocated buffer, resulting in
an out-of-bounds read from kernel heap memory. This could potentially
leak sensitive kernel information to userspace.

Additionally, providing an excessively large wqe_size can trigger a
WARNING in the memory allocation path, as reported by syzkaller.

This is inconsistent with ib_uverbs_unmarshall_recv() which properly
validates that wqe_size >= sizeof(struct ib_uverbs_recv_wr) before
proceeding.

Add the same validation for ib_uverbs_post_send() to ensure wqe_size
is at least sizeof(struct ib_uverbs_send_wr).

Fixes: c3bea3d2dc ("RDMA/uverbs: Use the iterator for ib_uverbs_unmarshall_recv()")
Signed-off-by: Yi Liu <liuy22@mails.tsinghua.edu.cn>
Link: https://patch.msgid.link/20260122142900.2356276-2-liuy22@mails.tsinghua.edu.cn
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2026-01-26 07:57:44 -05:00
Yishai Hadas
a272019a46 IB: Extend UVERBS_METHOD_REG_MR to get DMAH
Extend UVERBS_METHOD_REG_MR to get DMAH and pass it to all drivers.

It will be used in mlx5 driver as part of the next patch from the
series.

Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/2ae1e628c0675db81f092cc00d3ad6fbf6139405.1752752567.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-07-23 01:42:11 -04:00
Parav Pandit
28ea058a29 RDMA/nldev: Check CAP_NET_RAW in user namespace for QP modify
Currently, the capability check is done in the default
init_user_ns user namespace. When a process runs in a
non default user namespace, such check fails. Due to this
when a process is running using Podman, it fails to modify
the QP.

Since the RDMA device is a resource within a network namespace,
use the network namespace associated with the RDMA device to
determine its owning user namespace.

Fixes: 0cadb4db79 ("RDMA/uverbs: Restrict usage of privileged QKEYs")
Signed-off-by: Parav Pandit <parav@nvidia.com>
Link: https://patch.msgid.link/099eb263622ccdd27014db7e02fec824a3307829.1750963874.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-07-02 05:11:44 -04:00
Parav Pandit
4b527c23c9 RDMA/uverbs: Check CAP_NET_RAW in user namespace for RAW QP create
Currently, the capability check is done in the default
init_user_ns user namespace. When a process runs in a
non default user namespace, such check fails. Due to this
when a process is running using Podman, it fails to create
the QP.

Since the RDMA device is a resource within a network namespace,
use the network namespace associated with the RDMA device to
determine its owning user namespace.

Signed-off-by: Parav Pandit <parav@nvidia.com>
Link: https://patch.msgid.link/3914ef9702b01de8843a391ce397fca67d0fc7af.1750963874.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-07-02 05:11:44 -04:00
Parav Pandit
0498c2d998 RDMA/uverbs: Check CAP_NET_RAW in user namespace for QP create
Currently, the capability check is done in the default
init_user_ns user namespace. When a process runs in a
non default user namespace, such check fails. Due to this
when a process is running using Podman, it fails to create
the QP.

Since the RDMA device is a resource within a network namespace,
use the network namespace associated with the RDMA device to
determine its owning user namespace.

Fixes: 2dee0e5458 ("IB/uverbs: Enable QP creation with a given source QP number")
Signed-off-by: Parav Pandit <parav@nvidia.com>
Link: https://patch.msgid.link/0e5920d1dfe836817bb07576b192da41b637130b.1750963874.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-07-01 05:21:41 -04:00
Parav Pandit
f458ccd2aa RDMA/uverbs: Check CAP_NET_RAW in user namespace for flow create
Currently, the capability check is done in the default
init_user_ns user namespace. When a process runs in a
non default user namespace, such check fails. Due to this
when a process is running using Podman, it fails to create
the flow resource.

Since the RDMA device is a resource within a network namespace,
use the network namespace associated with the RDMA device to
determine its owning user namespace.

Fixes: 436f2ad05a ("IB/core: Export ib_create/destroy_flow through uverbs")
Signed-off-by: Parav Pandit <parav@nvidia.com>
Suggested-by: Eric W. Biederman <ebiederm@xmission.com>
Link: https://patch.msgid.link/6df6f2f24627874c4f6d041c19dc1f6f29f68f84.1750963874.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-07-01 05:21:27 -04:00
Li Haoran
7bc871af41 RDMA/uverbs: Convert to use ERR_CAST()
As opposed to open-code, using the ERR_CAST macro clearly indicates that
this is a pointer to an error value and a type conversion was performed.

Link: https://patch.msgid.link/r/202504012109233981_YPVbd4wQzmAzP3tA5IG@zte.com.cn
Signed-off-by: Li Haoran <li.haoran7@zte.com.cn>
Signed-off-by: Shao Mingyin <shao.mingyin@zte.com.cn>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2025-04-07 15:12:07 -03:00
Maher Sanalla
81f8f7454a RDMA/uverbs: Propagate errors from rdma_lookup_get_uobject()
Currently, the IB uverbs API calls uobj_get_uobj_read(), which in turn
uses the rdma_lookup_get_uobject() helper to retrieve user objects.
In case of failure, uobj_get_uobj_read() returns NULL, overriding the
error code from rdma_lookup_get_uobject(). The IB uverbs API then
translates this NULL to -EINVAL, masking the actual error and
complicating debugging. For example, applications calling ibv_modify_qp
that fails with EBUSY when retrieving the QP uobject will see the
overridden error code EINVAL instead, masking the actual error.

Furthermore, based on rdma-core commit:
"2a22f1ced5f3 ("Merge pull request #1568 from jakemoroni/master")"
Kernel's IB uverbs return values are either ignored and passed on as is
to application or overridden with other errnos in a few cases.

Thus, to improve error reporting and debuggability, propagate the
original error from rdma_lookup_get_uobject() instead of replacing it
with EINVAL.

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://patch.msgid.link/64f9d3711b183984e939962c2f83383904f97dfb.1740577869.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-03-13 08:26:37 -04:00
Chiara Meiohas
fe9d7822ba RDMA/uverbs: Add support for UCAPs in context creation
Add support for file descriptor array attribute for GET_CONTEXT
commands.

Check that the file descriptor (fd) array represents fds for valid UCAPs.
Store the enabled UCAPs from the fd array as a bitmask in ib_ucontext.

Signed-off-by: Chiara Meiohas <cmeiohas@nvidia.com>
Link: https://patch.msgid.link/ebfb30bc947e2259b193c96a319c80e82599045b.1741261611.git.leon@kernel.org
Reviewed-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2025-03-09 13:13:02 -04:00
Dan Carpenter
d0257e089d RDMA/uverbs: Prevent integer overflow issue
In the expression "cmd.wqe_size * cmd.wr_count", both variables are u32
values that come from the user so the multiplication can lead to integer
wrapping.  Then we pass the result to uverbs_request_next_ptr() which also
could potentially wrap.  The "cmd.sge_count * sizeof(struct ib_uverbs_sge)"
multiplication can also overflow on 32bit systems although it's fine on
64bit systems.

This patch does two things.  First, I've re-arranged the condition in
uverbs_request_next_ptr() so that the use controlled variable "len" is on
one side of the comparison by itself without any math.  Then I've modified
all the callers to use size_mul() for the multiplications.

Fixes: 67cdb40ca4 ("[IB] uverbs: Implement more commands")
Cc: stable@vger.kernel.org
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://patch.msgid.link/b8765ab3-c2da-4611-aae0-ddd6ba173d23@stanley.mountain
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2024-12-04 09:19:39 -05:00
Al Viro
38052c2dd7 deal with the last remaing boolean uses of fd_file()
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2024-11-03 01:28:07 -05:00
Al Viro
88a2f6468d struct fd: representation change
We want the compiler to see that fdput() on empty instance
is a no-op.  The emptiness check is that file reference is NULL,
while fdput() is "fput() if FDPUT_FPUT is present in flags".
The reason why fdput() on empty instance is a no-op is something
compiler can't see - it's that we never generate instances with
NULL file reference combined with non-zero flags.

	It's not that hard to deal with - the real primitives behind
fdget() et.al. are returning an unsigned long value, unpacked by (inlined)
__to_fd() into the current struct file * + int.  The lower bits are
used to store flags, while the rest encodes the pointer.  Linus suggested
that keeping this unsigned long around with the extractions done by inlined
accessors should generate a sane code and that turns out to be the case.
Namely, turning struct fd into a struct-wrapped unsinged long, with
        fd_empty(f) => unlikely(f.word == 0)
	fd_file(f) => (struct file *)(f.word & ~3)
	fdput(f) => if (f.word & 1) fput(fd_file(f))
ends up with compiler doing the right thing.  The cost is the patch
footprint, of course - we need to switch f.file to fd_file(f) all over
the tree, and it's not doable with simple search and replace; there are
false positives, etc.

	Note that the sole member of that structure is an opaque
unsigned long - all accesses should be done via wrappers and I don't
want to use a name that would invite manual casts to file pointers,
etc.  The value of that member is equal either to (unsigned long)p | flags,
p being an address of some struct file instance, or to 0 for an empty fd.

	For now the new predicate (fd_empty(f)) has no users; all the
existing checks have form (!fd_file(f)).  We will convert to fd_empty()
use later; here we only define it (and tell the compiler that it's
unlikely to return true).

	This commit only deals with representation change; there will
be followups.

Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2024-08-12 22:01:05 -04:00
Al Viro
1da91ea87a introduce fd_file(), convert all accessors to it.
For any changes of struct fd representation we need to
turn existing accesses to fields into calls of wrappers.
Accesses to struct fd::flags are very few (3 in linux/file.h,
1 in net/socket.c, 3 in fs/overlayfs/file.c and 3 more in
explicit initializers).
	Those can be dealt with in the commit converting to
new layout; accesses to struct fd::file are too many for that.
	This commit converts (almost) all of f.file to
fd_file(f).  It's not entirely mechanical ('file' is used as
a member name more than just in struct fd) and it does not
even attempt to distinguish the uses in pointer context from
those in boolean context; the latter will be eventually turned
into a separate helper (fd_empty()).

	NOTE: mass conversion to fd_empty(), tempting as it
might be, is a bad idea; better do that piecewise in commit
that convert from fdget...() to CLASS(...).

[conflicts in fs/fhandle.c, kernel/bpf/syscall.c, mm/memcontrol.c
caught by git; fs/stat.c one got caught by git grep]
[fs/xattr.c conflict]

Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2024-08-12 22:00:43 -04:00
Akiva Goldberger
dd6d7f8574 RDMA: Pass entire uverbs attr bundle to create cq function
Changes the create_cq verb signature by sending the entire uverbs attr
bundle as a parameter. This allows drivers to send driver specific attrs
through ioctl for the create_cq verb and access them in their driver
specific code.

Also adds a new enum value for driver specific ioctl attributes for
methods already supporting UHW.

Link: https://lore.kernel.org/r/ed147343987c0d43fd391c1b2f85e2f425747387.1719512393.git.leon@kernel.org
Signed-off-by: Akiva Goldberger <agoldberger@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-06-27 16:28:21 -03:00
Erick Archer
14b526f55b RDMA/uverbs: Remove flexible arrays from struct *_filter
When a struct containing a flexible array is included in another struct,
and there is a member after the struct-with-flex-array, there is a
possibility of memory overlap. These cases must be audited [1]. See:

struct inner {
	...
	int flex[];
};

struct outer {
	...
	struct inner header;
	int overlap;
	...
};

This is the scenario for all the "struct *_filter" structures that are
included in the following "struct ib_flow_spec_*" structures:

struct ib_flow_spec_eth
struct ib_flow_spec_ib
struct ib_flow_spec_ipv4
struct ib_flow_spec_ipv6
struct ib_flow_spec_tcp_udp
struct ib_flow_spec_tunnel
struct ib_flow_spec_esp
struct ib_flow_spec_gre
struct ib_flow_spec_mpls

The pattern is like the one shown below:

struct *_filter {
	...
	u8 real_sz[];
};

struct ib_flow_spec_* {
	...
	struct *_filter val;
	struct *_filter mask;
};

In this case, the trailing flexible array "real_sz" is never allocated
and is only used to calculate the size of the structures. Here the use
of the "offsetof" helper can be changed by the "sizeof" operator because
the goal is to get the size of these structures. Therefore, the trailing
flexible arrays can also be removed.

However, due to the trailing padding that can be induced in structs it
is possible that the:

offsetof(struct *_filter, real_sz) != sizeof(struct *_filter)

This situation happens with the "struct ib_flow_ipv6_filter" and to
avoid it the "__packed" macro is used in this structure. But now, the
"sizeof(struct ib_flow_ipv6_filter)" has changed. This is not a problem
since this size is not used in the code.

The situation now is that "sizeof(struct ib_flow_spec_ipv6)" has also
changed (this struct contains the struct ib_flow_ipv6_filter). This is
also not a problem since it is only used to set the size of the "union
ib_flow_spec", which can store all the "ib_flow_spec_*" structures.

Link: https://lore.kernel.org/r/20240217142913.4285-1-erick.archer@gmx.com
Signed-off-by: Erick Archer <erick.archer@gmx.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-02-21 13:28:52 -04:00
Patrisious Haddad
465d6b42f1 RDMA/core: Add support to set privileged QKEY parameter
Add netlink command that enables/disables privileged QKEY by default.

It is disabled by default, since according to IB spec only privileged
users are allowed to use privileged QKEY.
According to the IB specification rel-1.6, section 3.5.3:
"QKEYs with the most significant bit set are considered controlled
QKEYs, and a HCA does not allow a consumer to arbitrarily specify a
controlled QKEY."

Using rdma tool,
$rdma system set privileged-qkey on

When enabled non-privileged users would be able to use
controlled QKEYs which are considered privileged.

Using rdma tool,
$rdma system set privileged-qkey off

When disabled only privileged users would be able to use
controlled QKEYs.

You can also use the command below to check the parameter state:
$rdma system show
netns shared privileged-qkey off copy-on-fork on

Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Link: https://lore.kernel.org/r/90398be70a9d23d2aa9d0f9fd11d2c264c1be534.1696848201.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2023-10-19 09:31:00 +03:00
Edward Srouji
0cadb4db79 RDMA/uverbs: Restrict usage of privileged QKEYs
According to the IB specification rel-1.6, section 3.5.3:
"QKEYs with the most significant bit set are considered controlled
QKEYs, and a HCA does not allow a consumer to arbitrarily specify a
controlled QKEY."

Thus, block non-privileged users from setting such a QKEY.

Cc: stable@vger.kernel.org
Fixes: bc38a6abdd ("[PATCH] IB uverbs: core implementation")
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://lore.kernel.org/r/c00c809ddafaaf87d6f6cb827978670989a511b3.1685960567.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2023-06-11 11:26:06 +03:00
Daisuke Matsuda
241f9a27e0 IB: Set IOVA/LENGTH on IB_MR in core/uverbs layers
Set 'iova' and 'length' on ib_mr in ib_uverbs and ib_core layers to let all
drivers have the members filled. Also, this commit removes redundancy in
the respective drivers.

Previously, commit 04c0a5fcfc ("IB/uverbs: Set IOVA on IB MR in uverbs
layer") changed to set 'iova', but seems to have missed 'length' and the
ib_core layer at that time.

Fixes: 04c0a5fcfc ("IB/uverbs: Set IOVA on IB MR in uverbs layer")
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Link: https://lore.kernel.org/r/20220921080844.1616883-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-09-22 12:46:39 +03:00
Jason Gunthorpe
e945c653c8 RDMA: Split kernel-only global device caps from uverbs device caps
Split out flags from ib_device::device_cap_flags that are only used
internally to the kernel into kernel_cap_flags that is not part of the
uapi. This limits the device_cap_flags to being the same bitmap that will
be copied to userspace.

This cleanly splits out the uverbs flags from the kernel flags to avoid
confusion in the flags bitmap.

Add some short comments describing which each of the kernel flags is
connected to. Remove unused kernel flags.

Link: https://lore.kernel.org/r/0-v2-22c19e565eef+139a-kern_caps_jgg@nvidia.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-04-06 15:02:13 -03:00
Xiao Yang
f543a3e82b IB/uverbs: Move part of enum ib_device_cap_flags to uapi
1) Part of enum ib_device_cap_flags are used by ibv_query_device(3)
   or ibv_query_device_ex(3), so we define them in
   include/uapi/rdma/ib_user_verbs.h and only expose them to userspace.

2) Reformat enum ib_device_cap_flags by removing the indent before '='.

Link: https://lore.kernel.org/r/20220331032419.313904-2-yangx.jy@fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-04-04 10:58:37 -03:00
Zhu Yanjun
9692407d43 RDMA/uverbs: Remove the unnecessary assignment
The struct member variable create_flags is assigned twice.  Remove the
unnecessary assignment.

Fixes: ece9ca97cc ("RDMA/uverbs: Do not check the input length on create_cq/qp paths")
Link: https://lore.kernel.org/r/20211207064607.541695-1-yanjun.zhu@linux.dev
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Håkon Bugge <haakon.bugge@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-12-06 19:50:35 -04:00
Aharon Landau
f1a090f09f RDMA/core: Require the driver to set the IOVA correctly during rereg_mr
If the driver returns a new MR during rereg it has to fill it with the
IOVA from the proper source. If IB_MR_REREG_TRANS is set then the IOVA is
cmd.hca_va, otherwise the IOVA comes from the old MR. mlx5 for example has
two calls inside rereg_mr:

		return create_real_mr(new_pd, umem, mr->ibmr.iova,
				      new_access_flags);
and
		return create_real_mr(new_pd, new_umem, iova, new_access_flags);

Unconditionally overwriting the iova in the newly allocated MR will
corrupt the iova if the first path is used.

Remove the redundant initializations from ib_uverbs_rereg_mr().

Fixes: 6e0954b11c ("RDMA/uverbs: Allow drivers to create a new HW object during rereg_mr")
Link: https://lore.kernel.org/r/4b0a31bbc372842613286a10d7a8cbb0ee6069c7.1635400472.git.leonro@nvidia.com
Signed-off-by: Aharon Landau <aharonl@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-11-03 09:37:52 -03:00
Leon Romanovsky
d2b10794fc RDMA/core: Create clean QP creations interface for uverbs
Unify create QP creation interface to make clean approach to create
XRC_TGT and regular QPs.

Link: https://lore.kernel.org/r/5cd50e7d8ad9112545a1a61dea62799a5cb3224a.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-08-03 15:26:19 -03:00
Leon Romanovsky
5507f67d08 RDMA/core: Properly increment and decrement QP usecnts
The QP usecnts were incremented through QP attributes structure while
decreased through QP itself. Rely on the ib_creat_qp_user() code that
initialized all QP parameters prior returning to the user and increment
exactly like destroy does.

Link: https://lore.kernel.org/r/25d256a3bb1fc480b77d7fe439817b993de48610.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-08-03 15:26:19 -03:00
Leon Romanovsky
00a79d6b99 RDMA/core: Configure selinux QP during creation
All QP creation flows called ib_create_qp_security(), but differently.
This caused to the need to provide exclusion conditions for the XRC_TGT,
because such QP already had selinux configuration call.

In order to fix it, move ib_create_qp_security() to the general QP
creation routine.

Link: https://lore.kernel.org/r/4d7cd6f5828aca37fb62283e6b126b73ab86b18c.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-08-03 15:26:18 -03:00
Jason Gunthorpe
fdcebbc2ac Linux 5.13-rc7
-----BEGIN PGP SIGNATURE-----
 
 iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAmDPuyMeHHRvcnZhbGRz
 QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGvxgH/RKvSuRPwkJ2Jcp9
 VLi5kCbqtJlYLq6tB6peSJ8otKgxkcRwC0pIY4LlYIAWYboktLQ5RKp/9nB2h2FN
 aMZUMu6AI/lVJyFMI5MnKnJIUiUq+WXR3lSSlw68vwFLFdzqUZFNq+bqeiVvnIy1
 yqA6naj24Tu/RbYffQoPvdSJcU2SLXRMxwD8HRGiU2d51RaFsOvsZvF+P5TVcsEV
 ZmttJeER21CaI/A809eqaFmyGrUOcZZK9roZEbMwanTZOMw18biEsLu/UH4kBX01
 JC4+RlGxcWjQ5YNZgChsgoOK/CHzc6ITztTntdeDWAvwZjQFzV7pCy4/3BWne3O+
 5178yHM=
 =o8cN
 -----END PGP SIGNATURE-----

Merge tag 'v5.13-rc7' into rdma.git for-next

Linux 5.13-rc7

Needed for dependencies in following patches. Merge conflict in rxe_cmop.c
resolved by compining both patches.

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22 14:43:51 -03:00
Wenpeng Liang
3cea7b4a7d RDMA/core: Fix incorrect print format specifier
There are some '%u' for 'int' and '%d' for 'unsigend int', they should be
fixed.

Link: https://lore.kernel.org/r/1623325232-30900-1-git-send-email-liweihang@huawei.com
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-21 15:38:30 -03:00
Maor Gottlieb
2adcb4c5a5 RDMA: Verify port when creating flow rule
Validate port value provided by the user and with that remove no longer
needed validation by the driver.  The missing check in the mlx5_ib driver
could cause to the below oops.

Call trace:
  _create_flow_rule+0x2d4/0xf28 [mlx5_ib]
  mlx5_ib_create_flow+0x2d0/0x5b0 [mlx5_ib]
  ib_uverbs_ex_create_flow+0x4cc/0x624 [ib_uverbs]
  ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0xd4/0x150 [ib_uverbs]
  ib_uverbs_cmd_verbs.isra.7+0xb28/0xc50 [ib_uverbs]
  ib_uverbs_ioctl+0x158/0x1d0 [ib_uverbs]
  do_vfs_ioctl+0xd0/0xaf0
  ksys_ioctl+0x84/0xb4
  __arm64_sys_ioctl+0x28/0xc4
  el0_svc_common.constprop.3+0xa4/0x254
  el0_svc_handler+0x84/0xa0
  el0_svc+0x10/0x26c
 Code: b9401260 f9615681 51000400 8b001c20 (f9403c1a)

Fixes: 436f2ad05a ("IB/core: Export ib_create/destroy_flow through uverbs")
Link: https://lore.kernel.org/r/faad30dc5219a01727f47db3dc2f029d07c82c00.1623309971.git.leonro@nvidia.com
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-10 08:59:33 -03:00
Leon Romanovsky
f974428872 RDMA/core: Sanitize WQ state received from the userspace
The mlx4 and mlx5 implemented differently the WQ input checks.  Instead of
duplicating mlx4 logic in the mlx5, let's prepare the input in the central
place.

The mlx5 implementation didn't check for validity of state input.  It is
not real bug because our FW checked that, but still worth to fix.

Fixes: f213c05272 ("IB/uverbs: Add WQ support")
Link: https://lore.kernel.org/r/ac41ad6a81b095b1a8ad453dcf62cf8d3c5da779.1621413310.git.leonro@nvidia.com
Reported-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-02 15:20:11 -03:00
Wenpeng Liang
f681967ae7 RDMA/core: Remove redundant spaces
Space is not required after '(', before ')', before ',' and between '*'
and symbol name of a definition.

Link: https://lore.kernel.org/r/1617783353-48249-5-git-send-email-liweihang@huawei.com
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-12 14:56:48 -03:00
Mark Bloch
1fb7f8973f 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>
2021-03-26 09:31:21 -03:00
Xiao Yang
efeb973ffc RDMA/uverbs: Don't set rcq for a QP if qp_type is IB_QPT_XRC_INI
An INI QP doesn't require receive CQ, the creation flow sets the recv counts
to zero:

		if (cmd->qp_type == IB_QPT_XRC_INI) {
			cmd->max_recv_wr = 0;
			cmd->max_recv_sge = 0;

The new IOCTL path also does not set the rcq, so make things the same.

Link: https://lore.kernel.org/r/20201216071755.149449-1-yangx.jy@cn.fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-21 10:26:15 -04:00
Jason Gunthorpe
6e0954b11c RDMA/uverbs: Allow drivers to create a new HW object during rereg_mr
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>
2020-12-07 14:06:23 -04:00
Jason Gunthorpe
adac4cb3c1 RDMA/uverbs: Check ODP in ib_check_mr_access() as well
No reason only one caller checks this. This properly blocks ODP
from the rereg flow if the device does not support ODP.

Link: https://lore.kernel.org/r/20201130075839.278575-3-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-12-07 14:06:23 -04:00