Hey Linus,

/* Summary */ Here is an urgent fix for ptrace_may_access() permission checking.

Commit 69f594a38967 ("ptrace: do not audit capability check when outputing /proc/pid/stat") introduced the ability to opt out of audit messages for accesses to various proc files since they are not violations of policy. While doing so it switched the check from ns_capable() to has_ns_capability{_noaudit}(). That means it switched from checking the subjective credentials (ktask->cred) of the task to using the objective credentials (ktask->real_cred). This is appears to be wrong. ptrace_has_cap() is currently only used in ptrace_may_access() And is used to check whether the calling task (subject) has the CAP_SYS_PTRACE capability in the provided user namespace to operate on the target task (object). According to the cred.h comments this means the subjective credentials of the calling task need to be used.

With this pr we switch ptrace_has_cap() to use security_capable() and thus back to using the subjective credentials.

As one example where this might be particularly problematic, Jann pointed out that in combination with the upcoming IORING_OP_OPENAT{2} feature, this bug might allow unprivileged users to bypass the capability checks while asynchronously opening files like /proc/*/mem, because the capability checks for this would be performed against kernel credentials.

To illustrate on the former point about this being exploitable: When io_uring creates a new context it records the subjective credentials of the caller. Later on, when it starts to do work it creates a kernel thread and registers a callback. The callback runs with kernel creds for ktask->real_cred and ktask->cred. To prevent this from becoming a full-blown 0-day io_uring will call override_cred() and override ktask->cred with the subjective credentials of the creator of the io_uring instance. With ptrace_has_cap() currently looking at ktask->real_cred this override will be ineffective and the caller will be able to open arbitray proc files as mentioned above. Luckily, this is currently not exploitable but will turn into a 0-day once IORING_OP_OPENAT{2} land in v5.6. Let's fix it now.

To minimize potential regressions I successfully ran the criu testsuite. criu makes heavy use of ptrace() and extensively hits ptrace_may_access() codepaths and has a good change of detecting any regressions. Additionally, I succesfully ran the ptrace and seccomp kernel tests.

/* Testing */ All patches have seen exposure in linux-next and are based on v5.5-rc6. As mentioned above, the criu test-suite which is one of the test-suits make massive use of ptrace and hitting ptrace_may_access() codepaths successfully passed on a kernel with this fix: ################## ALL TEST(S) PASSED (TOTAL 178/SKIPPED 16) ################### I've posted the full test-log at: https://gitlab.com/snippets/1931214 Additionally, I succesfully ran the ptrace and seccomp kernel tests. We also will add a regression test once IO_URING_OPENAT{2} has landed for v5.6 since this gives us a really easy test.

/* Conflicts */ At the time of creating this PR no merge conflicts were reported from linux-next.

The following changes since commit b3a987b0264d3ddbb24293ebff10eddfc472f653:

Linux 5.5-rc6 (2020-01-12 16:55:08 -0800)

are available in the Git repository at:

git@gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux tags/for-linus-2020-01-18

for you to fetch changes up to 6b3ad6649a4c75504edeba242d3fd36b3096a57f:

ptrace: reintroduce usage of subjective credentials in ptrace_has_cap() (2020-01-18 13:51:39 +0100)

Please consider pulling these changes from the signed for-linus-2020-01-18 tag.

Thanks! Christian

---------------------------------------------------------------- for-linus-2020-01-18

---------------------------------------------------------------- Christian Brauner (1): ptrace: reintroduce usage of subjective credentials in ptrace_has_cap()

kernel/ptrace.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)