Linux RDMA and InfiniBand development
 help / color / mirror / Atom feed
* [PATCH v2 11/11] infiniband: reorder capability check last
       [not found] <20250302160657.127253-1-cgoettsche@seltendoof.de>
@ 2025-03-02 16:06 ` Christian Göttsche
  2025-03-03 19:04   ` Leon Romanovsky
  0 siblings, 1 reply; 2+ messages in thread
From: Christian Göttsche @ 2025-03-02 16:06 UTC (permalink / raw)
  Cc: Christian Göttsche, Serge Hallyn, Jan Kara, Julia Lawall,
	Nicolas Palix, linux-kernel, linux-security-module, cocci,
	Leon Romanovsky, Jason Gunthorpe, linux-rdma

From: Christian Göttsche <cgzones@googlemail.com>

capable() calls refer to enabled LSMs whether to permit or deny the
request.  This is relevant in connection with SELinux, where a
capability check results in a policy decision and by default a denial
message on insufficient permission is issued.
It can lead to three undesired cases:
  1. A denial message is generated, even in case the operation was an
     unprivileged one and thus the syscall succeeded, creating noise.
  2. To avoid the noise from 1. the policy writer adds a rule to ignore
     those denial messages, hiding future syscalls, where the task
     performs an actual privileged operation, leading to hidden limited
     functionality of that task.
  3. To avoid the noise from 1. the policy writer adds a rule to permit
     the task the requested capability, while it does not need it,
     violating the principle of least privilege.

Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
Reviewed-by: Serge Hallyn <serge@hallyn.com>
---
 drivers/infiniband/hw/mlx5/devx.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index 4186884c66e1..39304cae5b10 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -136,12 +136,14 @@ int mlx5_ib_devx_create(struct mlx5_ib_dev *dev, bool is_user)
 		return -EINVAL;
 
 	uctx = MLX5_ADDR_OF(create_uctx_in, in, uctx);
-	if (is_user && capable(CAP_NET_RAW) &&
-	    (MLX5_CAP_GEN(dev->mdev, uctx_cap) & MLX5_UCTX_CAP_RAW_TX))
+	if (is_user &&
+	    (MLX5_CAP_GEN(dev->mdev, uctx_cap) & MLX5_UCTX_CAP_RAW_TX) &&
+	    capable(CAP_NET_RAW))
 		cap |= MLX5_UCTX_CAP_RAW_TX;
-	if (is_user && capable(CAP_SYS_RAWIO) &&
+	if (is_user &&
 	    (MLX5_CAP_GEN(dev->mdev, uctx_cap) &
-	     MLX5_UCTX_CAP_INTERNAL_DEV_RES))
+	     MLX5_UCTX_CAP_INTERNAL_DEV_RES) &&
+	    capable(CAP_SYS_RAWIO))
 		cap |= MLX5_UCTX_CAP_INTERNAL_DEV_RES;
 
 	MLX5_SET(create_uctx_in, in, opcode, MLX5_CMD_OP_CREATE_UCTX);
-- 
2.47.2


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH v2 11/11] infiniband: reorder capability check last
  2025-03-02 16:06 ` [PATCH v2 11/11] infiniband: reorder capability check last Christian Göttsche
@ 2025-03-03 19:04   ` Leon Romanovsky
  0 siblings, 0 replies; 2+ messages in thread
From: Leon Romanovsky @ 2025-03-03 19:04 UTC (permalink / raw)
  To: cgzones
  Cc: Serge Hallyn, Jan Kara, Julia Lawall, Nicolas Palix, linux-kernel,
	linux-security-module, cocci, Jason Gunthorpe, linux-rdma

On Sun, Mar 02, 2025 at 05:06:47PM +0100, Christian Göttsche wrote:
> From: Christian Göttsche <cgzones@googlemail.com>
> 
> capable() calls refer to enabled LSMs whether to permit or deny the
> request.  This is relevant in connection with SELinux, where a
> capability check results in a policy decision and by default a denial
> message on insufficient permission is issued.
> It can lead to three undesired cases:
>   1. A denial message is generated, even in case the operation was an
>      unprivileged one and thus the syscall succeeded, creating noise.
>   2. To avoid the noise from 1. the policy writer adds a rule to ignore
>      those denial messages, hiding future syscalls, where the task
>      performs an actual privileged operation, leading to hidden limited
>      functionality of that task.
>   3. To avoid the noise from 1. the policy writer adds a rule to permit
>      the task the requested capability, while it does not need it,
>      violating the principle of least privilege.
> 
> Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
> Reviewed-by: Serge Hallyn <serge@hallyn.com>
> ---
>  drivers/infiniband/hw/mlx5/devx.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)

Thanks, applied.
https://web.git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/commit/?h=wip/leon-for-next&id=3745242ad1e1c07d5990b33764529eb13565db44

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2025-03-03 19:04 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <20250302160657.127253-1-cgoettsche@seltendoof.de>
2025-03-02 16:06 ` [PATCH v2 11/11] infiniband: reorder capability check last Christian Göttsche
2025-03-03 19:04   ` Leon Romanovsky

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox