kernel/linux.git/security/integrity/ima/ima_init.c, branch v6.18.21

ima: Suspend PCR extends and log appends when rebooting

2024-12-11T16:55:53+00:00

To avoid the following types of error messages due to a failure by the TPM driver to use the TPM, suspend TPM PCR extensions and the appending of entries to the IMA log once IMA's reboot notifier has been called. This avoids trying to use the TPM after the TPM subsystem has been shut down. [111707.685315][ T1] ima: Error Communicating to TPM chip, result: -19 [111707.685960][ T1] ima: Error Communicating to TPM chip, result: -19 Synchronization with the ima_extend_list_mutex to set ima_measurements_suspended ensures that the TPM subsystem is not shut down when IMA holds the mutex while appending to the log and extending the PCR. The alternative of reading the system_state variable would not provide this guarantee. This error could be observed on a ppc64 machine running SuSE Linux where processes are still accessing files after devices have been shut down. Suspending the IMA log and PCR extensions shortly before reboot does not seem to open a significant measurement gap since neither TPM quoting would work for attestation nor that new log entries could be written to anywhere after devices have been shut down. However, there's a time window between the invocation of the reboot notifier and the shutdown of devices. This includes all subsequently invoked reboot notifiers as well as kernel_restart_prepare() where __usermodehelper_disable() waits for all running_helpers to exit. During this time window IMA could now miss log entries even though attestation would still work. The reboot of the system shortly after may make this small gap insignificant. Signed-off-by: Tushar Sugandhi Signed-off-by: Stefan Berger Reviewed-by: Jarkko Sakkinen Signed-off-by: Mimi Zohar

integrity: Avoid -Wflex-array-member-not-at-end warnings

2024-04-08T11:55:48+00:00

-Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. There is currently an object (`hdr)` in `struct ima_max_digest_data` that contains a flexible structure (`struct ima_digest_data`): struct ima_max_digest_data { struct ima_digest_data hdr; u8 digest[HASH_MAX_DIGESTSIZE]; } __packed; So, in order to avoid ending up with a flexible-array member in the middle of a struct, we use the `__struct_group()` helper to separate the flexible array from the rest of the members in the flexible structure: struct ima_digest_data { __struct_group(ima_digest_data_hdr, hdr, __packed, ... the rest of the members ); u8 digest[]; } __packed; And similarly for `struct evm_ima_xattr_data`. With the change described above, we can now declare an object of the type of the tagged `struct ima_digest_data_hdr`, without embedding the flexible array in the middle of another struct: struct ima_max_digest_data { struct ima_digest_data_hdr hdr; u8 digest[HASH_MAX_DIGESTSIZE]; } __packed; And similarly for `struct evm_digest` and `struct evm_xattr`. We also use `container_of()` whenever we need to retrieve a pointer to the flexible structure. So, with these changes, fix the following warnings: security/integrity/evm/evm.h:64:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/evm/../integrity.h:40:35: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/evm/../integrity.h:68:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/ima/../integrity.h:40:35: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/ima/../integrity.h:68:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/integrity.h:40:35: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/integrity.h:68:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/platform_certs/../integrity.h:40:35: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] security/integrity/platform_certs/../integrity.h:68:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Signed-off-by: Mimi Zohar

ima: Make it independent from 'integrity' LSM

2024-02-16T04:43:47+00:00

Make the 'ima' LSM independent from the 'integrity' LSM by introducing IMA own integrity metadata (ima_iint_cache structure, with IMA-specific fields from the integrity_iint_cache structure), and by managing it directly from the 'ima' LSM. Create ima_iint.c and introduce the same integrity metadata management functions found in iint.c (renamed with ima_). However, instead of putting metadata in an rbtree, reserve space from IMA in the inode security blob for a pointer, and introduce the ima_inode_set_iint()/ima_inode_get_iint() primitives to store/retrieve that pointer. This improves search time from logarithmic to constant. Consequently, don't include the inode pointer as field in the ima_iint_cache structure, since the association with the inode is clear. Since the inode field is missing in ima_iint_cache, pass the extra inode parameter to ima_get_verity_digest(). Prefer storing the pointer instead of the entire ima_iint_cache structure, to avoid too much memory pressure. Use the same mechanism as before, a cache named ima_iint_cache (renamed from iint_cache), to quickly allocate a new ima_iint_cache structure when requested by the IMA policy. Create the new ima_iint_cache in ima_iintcache_init(), called by init_ima_lsm(), during the initialization of the 'ima' LSM. And, register ima_inode_free_security() to free the ima_iint_cache structure, if exists. Replace integrity_iint_cache with ima_iint_cache in various places of the IMA code. Also, replace integrity_inode_get() and integrity_iint_find(), respectively with ima_inode_get() and ima_iint_find(). Finally, move the remaining IMA-specific flags to security/integrity/ima/ima.h, since they are now unnecessary in the common integrity layer. Signed-off-by: Roberto Sassu Reviewed-by: Casey Schaufler Reviewed-by: Stefan Berger Reviewed-by: Mimi Zohar Acked-by: Mimi Zohar Signed-off-by: Paul Moore

ima: define ima_max_digest_data struct without a flexible array variable

2022-02-15T16:52:06+00:00

To support larger hash digests in the 'iint' cache, instead of defining the 'digest' field as the maximum digest size, the 'digest' field was defined as a flexible array variable. The "ima_digest_data" struct was wrapped inside a local structure with the maximum digest size. But before adding the record to the iint cache, memory for the exact digest size was dynamically allocated. The original reason for defining the 'digest' field as a flexible array variable is still valid for the 'iint' cache use case. Instead of wrapping the 'ima_digest_data' struct in a local structure define 'ima_max_digest_data' struct. Reviewed-by: Stefan Berger Signed-off-by: Mimi Zohar

ima: Add digest and digest_len params to the functions to measure a buffer

2021-07-23T13:27:02+00:00

This patch performs the final modification necessary to pass the buffer measurement to callers, so that they provide a functionality similar to ima_file_hash(). It adds the 'digest' and 'digest_len' parameters to ima_measure_critical_data() and process_buffer_measurement(). These functions calculate the digest even if there is no suitable rule in the IMA policy and, in this case, they simply return 1 before generating a new measurement entry. Signed-off-by: Roberto Sassu Reviewed-by: Lakshmi Ramasubramanian Signed-off-by: Mimi Zohar

evm: Load EVM key in ima_load_x509() to avoid appraisal

2021-05-21T16:47:04+00:00

The public builtin keys do not need to be appraised by IMA as the restriction on the IMA/EVM trusted keyrings ensures that a key can be loaded only if it is signed with a key on the builtin or secondary keyrings. However, when evm_load_x509() is called, appraisal is already enabled and a valid IMA signature must be added to the EVM key to pass verification. Since the restriction is applied on both IMA and EVM trusted keyrings, it is safe to disable appraisal also when the EVM key is loaded. This patch calls evm_load_x509() inside ima_load_x509() if CONFIG_IMA_LOAD_X509 is enabled, which crosses the normal IMA and EVM boundary. Signed-off-by: Roberto Sassu Signed-off-by: Mimi Zohar

IMA: Measure kernel version in early boot

2021-01-27T00:06:41+00:00

The integrity of a kernel can be verified by the boot loader on cold boot, and during kexec, by the current running kernel, before it is loaded. However, it is still possible that the new kernel being loaded is older than the current kernel, and/or has known vulnerabilities. Therefore, it is imperative that an attestation service be able to verify the version of the kernel being loaded on the client, from cold boot and subsequent kexec system calls, ensuring that only kernels with versions known to be good are loaded. Measure the kernel version using ima_measure_critical_data() early on in the boot sequence, reducing the chances of known kernel vulnerabilities being exploited. With IMA being part of the kernel, this overall approach makes the measurement itself more trustworthy. To enable measuring the kernel version "ima_policy=critical_data" needs to be added to the kernel command line arguments. For example, BOOT_IMAGE=/boot/vmlinuz-5.11.0-rc3+ root=UUID=fd643309-a5d2-4ed3-b10d-3c579a5fab2f ro nomodeset ima_policy=critical_data If runtime measurement of the kernel version is ever needed, the following should be added to /etc/ima/ima-policy: measure func=CRITICAL_DATA label=kernel_info To extract the measured data after boot, the following command can be used: grep -m 1 "kernel_version" \ /sys/kernel/security/integrity/ima/ascii_runtime_measurements Sample output from the command above: 10 a8297d408e9d5155728b619761d0dd4cedf5ef5f ima-buf sha256:5660e19945be0119bc19cbbf8d9c33a09935ab5d30dad48aa11f879c67d70988 kernel_version 352e31312e302d7263332d31363138372d676564623634666537383234342d6469727479 The above hex-ascii string corresponds to the kernel version (e.g. xxd -r -p): 5.11.0-rc3-16187-gedb64fe78244-dirty Signed-off-by: Raphael Gianotti Signed-off-by: Mimi Zohar

ima: Call ima_calc_boot_aggregate() in ima_eventdigest_init()

2020-06-03T21:20:43+00:00

If the template field 'd' is chosen and the digest to be added to the measurement entry was not calculated with SHA1 or MD5, it is recalculated with SHA1, by using the passed file descriptor. However, this cannot be done for boot_aggregate, because there is no file descriptor. This patch adds a call to ima_calc_boot_aggregate() in ima_eventdigest_init(), so that the digest can be recalculated also for the boot_aggregate entry. Cc: stable@vger.kernel.org # 3.13.x Fixes: 3ce1217d6cd5d ("ima: define template fields library and new helpers") Reported-by: Takashi Iwai Signed-off-by: Roberto Sassu Signed-off-by: Mimi Zohar

ima: Switch to ima_hash_algo for boot aggregate

2020-04-20T02:03:39+00:00

boot_aggregate is the first entry of IMA measurement list. Its purpose is to link pre-boot measurements to IMA measurements. As IMA was designed to work with a TPM 1.2, the SHA1 PCR bank was always selected even if a TPM 2.0 with support for stronger hash algorithms is available. This patch first tries to find a PCR bank with the IMA default hash algorithm. If it does not find it, it selects the SHA256 PCR bank for TPM 2.0 and SHA1 for TPM 1.2. Ultimately, it selects SHA1 also for TPM 2.0 if the SHA256 PCR bank is not found. If none of the PCR banks above can be found, boot_aggregate file digest is filled with zeros, as for TPM bypass, making it impossible to perform a remote attestation of the system. Cc: stable@vger.kernel.org # 5.1.x Fixes: 879b589210a9 ("tpm: retrieve digest size of unknown algorithms with PCR read") Reported-by: Jerry Snitselaar Suggested-by: James Bottomley Signed-off-by: Roberto Sassu Signed-off-by: Mimi Zohar

integrity: Remove duplicate pr_fmt definitions

2020-02-28T19:32:58+00:00

The #define for formatting log messages, pr_fmt, is duplicated in the files under security/integrity. This change moves the definition to security/integrity/integrity.h and removes the duplicate definitions in the other files under security/integrity. With this change, the messages in the following files will be prefixed with 'integrity'. security/integrity/platform_certs/platform_keyring.c security/integrity/platform_certs/load_powerpc.c security/integrity/platform_certs/load_uefi.c security/integrity/iint.c e.g. "integrity: Error adding keys to platform keyring %s\n" And the messages in the following file will be prefixed with 'ima'. security/integrity/ima/ima_mok.c e.g. "ima: Allocating IMA blacklist keyring.\n" For the rest of the files under security/integrity, there will be no change in the message format. Suggested-by: Shuah Khan Suggested-by: Joe Perches Signed-off-by: Tushar Sugandhi Reviewed-by: Lakshmi Ramasubramanian Signed-off-by: Mimi Zohar