Re: [PATCH v8 00/12] Introduce CAP_PERFMON to secure system performance monitoring and observability

[Alexey Budankov]

On 10.07.2020 20:09, Arnaldo Carvalho de Melo wrote:
> Em Fri, Jul 10, 2020 at 05:30:50PM +0300, Alexey Budankov escreveu:
>> On 10.07.2020 16:31, Ravi Bangoria wrote:
>>>> Currently access to perf_events, i915_perf and other performance
>>>> monitoring and observability subsystems of the kernel is open only for
>>>> a privileged process [1] with CAP_SYS_ADMIN capability enabled in the
>>>> process effective set [2].
> 
>>>> This patch set introduces CAP_PERFMON capability designed to secure
>>>> system performance monitoring and observability operations so that
>>>> CAP_PERFMON would assist CAP_SYS_ADMIN capability in its governing role
>>>> for performance monitoring and observability subsystems of the kernel.
>  
>>> I'm seeing an issue with CAP_PERFMON when I try to record data for a
>>> specific target. I don't know whether this is sort of a regression or
>>> an expected behavior.
>  
>> Thanks for reporting and root causing this case. The behavior looks like
>> kind of expected since currently CAP_PERFMON takes over the related part
>> of CAP_SYS_ADMIN credentials only. Actually Perf security docs [1] say
>> that access control is also subject to CAP_SYS_PTRACE credentials.
> 
> I think that stating that in the error message would be helpful, after
> all, who reads docs? 8-)

At least those who write it :D ...

> 
> I.e., this:
> 
> $ ./perf stat ls
> Â Error:
> Â Access to performance monitoring and observability operations is limited.
> $
> 
> Could become:
> 
> $ ./perf stat ls
> Â Error:
> Â Access to performance monitoring and observability operations is limited.
>   Right now only CAP_PERFMON is granted, you may need CAP_SYS_PTRACE.
> $

It would better provide reference to perf security docs in the tool output.
Looks like extending ptrace_may_access() check for perf_events with CAP_PERFMON
makes monitoring simpler and even more secure to use since Perf tool need
not to start/stop/single-step and read/write registers and memory and so on
like a debugger or strace-like tool. What do you think?

Alexei

> 
> - Arnaldo
>  
>> CAP_PERFMON could be used to extend and substitute ptrace_may_access()
>> check in perf_events subsystem to simplify user experience at least in
>> this specific case.
>>
>> Alexei
>>
>> [1] https://www.kernel.org/doc/html/latest/admin-guide/perf-security.html
>>
>>>
>>> Without setting CAP_PERFMON:
>>>
>>> Â $ getcap ./perf
>>> Â $ ./perf stat -a ls
>>> ÂÂÂ Error:
>>> ÂÂÂ Access to performance monitoring and observability operations is limited.
>>> Â $ ./perf stat ls
>>> ÂÂÂ Performance counter stats for 'ls':
>>> ÂÂ ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 2.06 msec task-clock:uÂÂÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.418 CPUs utilized
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ context-switches:uÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ cpu-migrations:uÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>>
>>> With CAP_PERFMON:
>>>
>>> Â $ getcap ./perf
>>> ÂÂÂ ./perf = cap_perfmon+ep
>>> Â $ ./perf stat -a ls
>>> ÂÂÂ Performance counter stats for 'system wide':
>>> ÂÂ ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 142.42 msec cpu-clockÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ #ÂÂ 25.062 CPUs utilized
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 182ÂÂÂÂÂ context-switchesÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.001 M/sec
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 48ÂÂÂÂÂ cpu-migrationsÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.337 K/sec
>>> Â $ ./perf stat ls
>>> ÂÂÂ Error:
>>> ÂÂÂ Access to performance monitoring and observability operations is limited.
>>>
>>> Am I missing something silly?
>>>
>>> Analysis:
>>> ---------
>>> A bit more analysis lead me to below kernel code fs/exec.c:
>>>
>>> Â begin_new_exec()
>>> Â {
>>> ÂÂÂÂÂÂÂ ...
>>> ÂÂÂÂÂÂÂ if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP ||
>>> ÂÂÂÂÂÂÂÂÂÂÂ !(uid_eq(current_euid(), current_uid()) &&
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂ gid_eq(current_egid(), current_gid())))
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ set_dumpable(current->mm, suid_dumpable);
>>> ÂÂÂÂÂÂÂ else
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ set_dumpable(current->mm, SUID_DUMP_USER);
>>>
>>> ÂÂÂÂÂÂÂ ...
>>> ÂÂÂÂÂÂÂ commit_creds(bprm->cred);
>>> Â }
>>>
>>> When I execute './perf stat ls', it's going into else condition and thus sets
>>> dumpable flag as SUID_DUMP_USER. Then in commit_creds():
>>>
>>> Â int commit_creds(struct cred *new)
>>> Â {
>>> ÂÂÂÂÂÂÂ ...
>>> ÂÂÂÂÂÂÂ /* dumpability changes */
>>> ÂÂÂÂÂÂÂ if (...
>>> ÂÂÂÂÂÂÂÂÂÂÂ !cred_cap_issubset(old, new)) {
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ if (task->mm)
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ set_dumpable(task->mm, suid_dumpable);
>>> Â }
>>>
>>> !cred_cap_issubset(old, new) fails for perf without any capability and thus
>>> it doesn't execute set_dumpable(). Whereas that condition passes for perf
>>> with CAP_PERFMON and thus it overwrites old value (SUID_DUMP_USER) with
>>> suid_dumpable in mm_flags. On an Ubuntu, suid_dumpable default value is
>>> SUID_DUMP_ROOT. On Fedora, it's SUID_DUMP_DISABLE. (/proc/sys/fs/suid_dumpable).
>>>
>>> Now while opening an event:
>>>
>>> Â perf_event_open()
>>> ÂÂÂ ptrace_may_access()
>>> ÂÂÂÂÂ __ptrace_may_access() {
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ ...
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ if (mm &&
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ ((get_dumpable(mm) != SUID_DUMP_USER) &&
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ !ptrace_has_cap(cred, mm->user_ns, mode)))
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ return -EPERM;
>>> ÂÂÂÂÂ }
>>>
>>> This if condition passes for perf with CAP_PERFMON and thus it returns -EPERM.
>>> But it fails for perf without CAP_PERFMON and thus it goes ahead and returns
>>> success. So opening an event fails when perf has CAP_PREFMON and tries to open
>>> process specific event as normal user.
>>>
>>> Workarounds:
>>> ------------
>>> Based on above analysis, I found couple of workarounds (examples are on
>>> Ubuntu 18.04.4 powerpc):
>>>
>>> Workaround1:
>>> Setting SUID_DUMP_USER as default (in /proc/sys/fs/suid_dumpable) solves the
>>> issue.
>>>
>>> Â # echo 1 > /proc/sys/fs/suid_dumpable
>>> Â $ getcap ./perf
>>> ÂÂÂ ./perf = cap_perfmon+ep
>>> Â $ ./perf stat ls
>>> ÂÂÂ Performance counter stats for 'ls':
>>> ÂÂ ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 1.47 msec task-clockÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.806 CPUs utilized
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ context-switchesÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ cpu-migrationsÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>>
>>> Workaround2:
>>> Using CAP_SYS_PTRACE along with CAP_PERFMON solves the issue.
>>>
>>> Â $ cat /proc/sys/fs/suid_dumpable
>>> ÂÂÂ 2
>>> Â # setcap "cap_perfmon,cap_sys_ptrace=ep" ./perf
>>> Â $ ./perf stat ls
>>> ÂÂÂ Performance counter stats for 'ls':
>>> ÂÂ ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 1.41 msec task-clockÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.826 CPUs utilized
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ context-switchesÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ cpu-migrationsÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>>
>>> Workaround3:
>>> Adding CAP_PERFMON to parent of perf (/bin/bash) also solves the issue.
>>>
>>> Â $ cat /proc/sys/fs/suid_dumpable
>>> ÂÂÂ 2
>>> Â # setcap "cap_perfmon=ep" /bin/bash
>>> Â # setcap "cap_perfmon=ep" ./perf
>>> Â $ bash
>>> Â $ ./perf stat ls
>>> ÂÂÂ Performance counter stats for 'ls':
>>> ÂÂ ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 1.47 msec task-clockÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.806 CPUs utilized
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ context-switchesÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ 0ÂÂÂÂÂ cpu-migrationsÂÂÂÂÂÂÂÂÂÂÂ #ÂÂÂ 0.000 K/sec
>>>
>>> - Ravi
>