user/sven/linux.git/security/keys, branch v5.1.15

KEYS: trusted: fix -Wvarags warning

2019-04-08T22:58:54Z

Fixes the warning reported by Clang: security/keys/trusted.c:146:17: warning: passing an object that undergoes default argument promotion to 'va_start' has undefined behavior [-Wvarargs] va_start(argp, h3); ^ security/keys/trusted.c:126:37: note: parameter of type 'unsigned char' is declared here unsigned char *h2, unsigned char h3, ...) ^ Specifically, it seems that both the C90 (4.8.1.1) and C11 (7.16.1.4) standards explicitly call this out as undefined behavior: The parameter parmN is the identifier of the rightmost parameter in the variable parameter list in the function definition (the one just before the ...). If the parameter parmN is declared with ... or with a type that is not compatible with the type that results after application of the default argument promotions, the behavior is undefined. Link: https://github.com/ClangBuiltLinux/linux/issues/41 Link: https://www.eskimo.com/~scs/cclass/int/sx11c.html Suggested-by: David Laight Suggested-by: Denis Kenzior Suggested-by: James Bottomley Suggested-by: Nathan Chancellor Signed-off-by: Nick Desaulniers Reviewed-by: Nathan Chancellor Tested-by: Nathan Chancellor Reviewed-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen Signed-off-by: James Morris

KEYS: trusted: allow trusted.ko to initialize w/o a TPM

2019-04-08T22:58:53Z

Allow trusted.ko to initialize w/o a TPM. This commit also adds checks to the exported functions to fail when a TPM is not available. Fixes: 240730437deb ("KEYS: trusted: explicitly use tpm_chip structure...") Cc: James Morris Reported-by: Dan Williams Tested-by: Dan Williams Signed-off-by: Jarkko Sakkinen Signed-off-by: James Morris

Merge branch 'next-tpm' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security

2019-03-11T00:37:29Z

Pull tpm updates from James Morris: - Clean up the transmission flow Cleaned up the whole transmission flow. Locking of the chip is now done in the level of tpm_try_get_ops() and tpm_put_ops() instead taking the chip lock inside tpm_transmit(). The nested calls inside tpm_transmit(), used with the resource manager, have been refactored out. Should make easier to perform more complex transactions with the TPM without making the subsystem a bigger mess (e.g. encrypted channel patches by James Bottomley). - PPI 1.3 support TPM PPI 1.3 introduces an additional optional command parameter that may be needed for some commands. Display the parameter if the command requires such a parameter. Only command 23 (SetPCRBanks) needs one. The PPI request file will show output like this then: # echo "23 16" > request # cat request 23 16 # echo "5" > request # cat request 5 - Extend all PCR banks in IMA Instead of static PCR banks array, the array of available PCR banks is now allocated dynamically. The digests sizes are determined dynamically using a probe PCR read without relying crypto's static list of hash algorithms. This should finally make sealing of measurements in IMA safe and secure. - TPM 2.0 selftests Added a test suite to tools/testing/selftests/tpm2 previously outside of the kernel tree: https://github.com/jsakkine-intel/tpm2-scripts * 'next-tpm' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (37 commits) tpm/ppi: Enable submission of optional command parameter for PPI 1.3 tpm/ppi: Possibly show command parameter if TPM PPI 1.3 is used tpm/ppi: Display up to 101 operations as define for version 1.3 tpm/ppi: rename TPM_PPI_REVISION_ID to TPM_PPI_REVISION_ID_1 tpm/ppi: pass function revision ID to tpm_eval_dsm() tpm: pass an array of tpm_extend_digest structures to tpm_pcr_extend() KEYS: trusted: explicitly use tpm_chip structure from tpm_default_chip() tpm: move tpm_chip definition to include/linux/tpm.h tpm: retrieve digest size of unknown algorithms with PCR read tpm: rename and export tpm2_digest and tpm2_algorithms tpm: dynamically allocate the allocated_banks array tpm: remove @flags from tpm_transmit() tpm: take TPM chip power gating out of tpm_transmit() tpm: introduce tpm_chip_start() and tpm_chip_stop() tpm: remove TPM_TRANSMIT_UNLOCKED flag tpm: use tpm_try_get_ops() in tpm-sysfs.c. tpm: remove @space from tpm_transmit() tpm: move TPM space code out of tpm_transmit() tpm: move tpm_validate_commmand() to tpm2-space.c tpm: clean up tpm_try_transmit() error handling flow ...

Merge branch 'next-integrity' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security

2019-03-11T00:32:04Z

Pull integrity updates from James Morris: "Mimi Zohar says: 'Linux 5.0 introduced the platform keyring to allow verifying the IMA kexec kernel image signature using the pre-boot keys. This pull request similarly makes keys on the platform keyring accessible for verifying the PE kernel image signature. Also included in this pull request is a new IMA hook that tags tmp files, in policy, indicating the file hash needs to be calculated. The remaining patches are cleanup'" * 'next-integrity' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: evm: Use defined constant for UUID representation ima: define ima_post_create_tmpfile() hook and add missing call evm: remove set but not used variable 'xattr' encrypted-keys: fix Opt_err/Opt_error = -1 kexec, KEYS: Make use of platform keyring for signature verify integrity, KEYS: add a reference to platform keyring

Merge branch 'next-general' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security

2019-03-07T19:44:01Z

Pull security subsystem updates from James Morris: - Extend LSM stacking to allow sharing of cred, file, ipc, inode, and task blobs. This paves the way for more full-featured LSMs to be merged, and is specifically aimed at LandLock and SARA LSMs. This work is from Casey and Kees. - There's a new LSM from Micah Morton: "SafeSetID gates the setid family of syscalls to restrict UID/GID transitions from a given UID/GID to only those approved by a system-wide whitelist." This feature is currently shipping in ChromeOS. * 'next-general' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (62 commits) keys: fix missing __user in KEYCTL_PKEY_QUERY LSM: Update list of SECURITYFS users in Kconfig LSM: Ignore "security=" when "lsm=" is specified LSM: Update function documentation for cap_capable security: mark expected switch fall-throughs and add a missing break tomoyo: Bump version. LSM: fix return value check in safesetid_init_securityfs() LSM: SafeSetID: add selftest LSM: SafeSetID: remove unused include LSM: SafeSetID: 'depend' on CONFIG_SECURITY LSM: Add 'name' field for SafeSetID in DEFINE_LSM LSM: add SafeSetID module that gates setid calls LSM: add SafeSetID module that gates setid calls tomoyo: Allow multiple use_group lines. tomoyo: Coding style fix. tomoyo: Swicth from cred->security to task_struct->security. security: keys: annotate implicit fall throughs security: keys: annotate implicit fall throughs security: keys: annotate implicit fall through capabilities:: annotate implicit fall through ...

keys: fix missing __user in KEYCTL_PKEY_QUERY

2019-03-04T23:48:37Z

The arg5 of KEYCTL_PKEY_QUERY should have a __user pointer tag on it as it is a user pointer. This clears the following sparse warning for this: security/keys/keyctl.c:1755:43: warning: incorrect type in argument 3 (different address spaces) security/keys/keyctl.c:1755:43: expected struct keyctl_pkey_query [noderef] * security/keys/keyctl.c:1755:43: got struct keyctl_pkey_query * Signed-off-by: Ben Dooks Acked-by: Serge Hallyn Signed-off-by: James Morris

KEYS: always initialize keyring_index_key::desc_len

2019-02-22T18:11:34Z

syzbot hit the 'BUG_ON(index_key->desc_len == 0);' in __key_link_begin() called from construct_alloc_key() during sys_request_key(), because the length of the key description was never calculated. The problem is that we rely on ->desc_len being initialized by search_process_keyrings(), specifically by search_nested_keyrings(). But, if the process isn't subscribed to any keyrings that never happens. Fix it by always initializing keyring_index_key::desc_len as soon as the description is set, like we already do in some places. The following program reproduces the BUG_ON() when it's run as root and no session keyring has been installed. If it doesn't work, try removing pam_keyinit.so from /etc/pam.d/login and rebooting. #include #include #include int main(void) { int id = add_key("keyring", "syz", NULL, 0, KEY_SPEC_USER_KEYRING); keyctl_setperm(id, KEY_OTH_WRITE); setreuid(5000, 5000); request_key("user", "desc", "", id); } Reported-by: syzbot+ec24e95ea483de0a24da@syzkaller.appspotmail.com Fixes: b2a4df200d57 ("KEYS: Expand the capacity of a keyring") Signed-off-by: Eric Biggers Signed-off-by: David Howells Cc: stable@vger.kernel.org Signed-off-by: James Morris

keys: Timestamp new keys

2019-02-15T22:12:09Z

Set the timestamp on new keys rather than leaving it unset. Fixes: 31d5a79d7f3d ("KEYS: Do LRU discard in full keyrings") Signed-off-by: David Howells Signed-off-by: James Morris

keys: Fix dependency loop between construction record and auth key

2019-02-15T22:12:09Z

In the request_key() upcall mechanism there's a dependency loop by which if a key type driver overrides the ->request_key hook and the userspace side manages to lose the authorisation key, the auth key and the internal construction record (struct key_construction) can keep each other pinned. Fix this by the following changes: (1) Killing off the construction record and using the auth key instead. (2) Including the operation name in the auth key payload and making the payload available outside of security/keys/. (3) The ->request_key hook is given the authkey instead of the cons record and operation name. Changes (2) and (3) allow the auth key to naturally be cleaned up if the keyring it is in is destroyed or cleared or the auth key is unlinked. Fixes: 7ee02a316600 ("keys: Fix dependency loop between construction record and auth key") Signed-off-by: David Howells Signed-off-by: James Morris

KEYS: allow reaching the keys quotas exactly

2019-02-15T22:12:08Z

If the sysctl 'kernel.keys.maxkeys' is set to some number n, then actually users can only add up to 'n - 1' keys. Likewise for 'kernel.keys.maxbytes' and the root_* versions of these sysctls. But these sysctls are apparently supposed to be *maximums*, as per their names and all documentation I could find -- the keyrings(7) man page, Documentation/security/keys/core.rst, and all the mentions of EDQUOT meaning that the key quota was *exceeded* (as opposed to reached). Thus, fix the code to allow reaching the quotas exactly. Fixes: 0b77f5bfb45c ("keys: make the keyring quotas controllable through /proc/sys") Cc: stable@vger.kernel.org Signed-off-by: Eric Biggers Signed-off-by: David Howells Signed-off-by: James Morris