From: Daniel Jurgens <danielj@xxxxxxxxxxxx> Currently there is no way to provide granular access control to an Infiniband fabric. By providing an ability to restrict user access to specific virtual subfabrics administrators can limit access to bandwidth and isolate users on the fabric. The approach for controlling access for Infiniband is to control access to partitions. A partition is similar in concept to a VLAN where each data packet carries the partition key (PKey) in its header and isolation is enforced by the hardware. The partition key is not a cryptographic key, it's a 16 bit number identifying the partition. By controlling access to PKeys users can be isolated on the fabric. All Infiniband fabrics must have a subnet manager. The subnet manager provisions the partitions and configures the end nodes. Each end port has a PKey table containing all the partitions it can access. In order to enforce access to partitions the subnet management interface (SMI) must also be controlled to prevent unauthorized changes to the fabric configuration. In order to support this there must be a capability to provide security contexts for two new types of objects - PKeys and SMIs. A PKey label consists of a subnet prefix and a range of PKey values and is similar to the labeling mechanism for netports. Each port of an infiniband device can reside on a different subnet, labeling the PKey values for specific subnet prefixes provides the user maximum flexibility. There is a single access vector for PKeys, called "access". An Infiniband device (ibdev) is labeled by name and port number. There is a single access vector for ibdevs as well, called "smi". Because RDMA allows for kernel bypass all enforcement must be done during connection setup. To communicate over RDMA requires a send and receive queue called a queue pair (QP). During the creation of a QP it is initialized before it can be used to send or receive data. During initialization the user must provide the PKey and port the QP will use, at this time access can be enforced. Because there is a possibility that the enforcement settings or security policy can change, a means of notifying the ib_core module of such changes is required. To facilitate this a single LSM hook is provided, ib_core will register for this hook and when called will recheck the PKey access for all existing QPs. Because frequent accesses to the same PKey's SID is expected a cache is implemented which is very similar to the netport cache. Daniel Jurgens (7): security: Add LSM hooks for Infiniband security selinux: Create policydb version for Infiniband support selinux: Call infiniband_flush LSM hook on AVC reset selinux: Allocate and free infiniband security hooks selinux: Implement Infiniband PKey "Access" access vector selinux: Implement IB Device SMI access vector selinux: Add a cache for quicker retreival of PKey SIDs include/linux/lsm_audit.h | 15 ++ include/linux/lsm_hooks.h | 43 ++++- include/linux/security.h | 37 ++++ security/Kconfig | 9 + security/security.c | 52 +++++ security/selinux/Makefile | 2 +- security/selinux/hooks.c | 87 +++++++++- security/selinux/include/classmap.h | 4 + security/selinux/include/initial_sid_to_string.h | 2 + security/selinux/include/objsec.h | 12 ++ security/selinux/include/pkey.h | 33 ++++ security/selinux/include/security.h | 7 +- security/selinux/pkey.c | 220 ++++++++++++++++++++++ security/selinux/ss/policydb.c | 129 +++++++++++-- security/selinux/ss/policydb.h | 13 ++- security/selinux/ss/services.c | 84 ++++++++ 16 files changed, 723 insertions(+), 26 deletions(-) create mode 100644 security/selinux/include/pkey.h create mode 100644 security/selinux/pkey.c -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html