From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.8 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8E1FAC433E5 for ; Mon, 13 Jul 2020 12:53:57 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 636F62075B for ; Mon, 13 Jul 2020 12:53:57 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (1024-bit key) header.d=kernel.org header.i=@kernel.org header.b="NU7xlNRI" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 636F62075B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=kernel.org Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id DD6E86E4CF; Mon, 13 Jul 2020 12:53:52 +0000 (UTC) Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by gabe.freedesktop.org (Postfix) with ESMTPS id D4BDF6E202 for ; Mon, 13 Jul 2020 12:17:49 +0000 (UTC) Received: from quaco.ghostprotocols.net (unknown [177.158.141.203]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id 5CCDB206F0; Mon, 13 Jul 2020 12:17:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1594642669; bh=wc2dhxkIG9GjfHmWSmkltugECChmqIjO4+eGKEz9U8U=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=NU7xlNRIpfzoyfaed7t2MB9wYUrmC/Wy9Zlmr0S7A2G8ByKqE2GKBqMXDhAYt24gG bdyovVnMvdmfyTmoH1pgspO72I5SfzqbEO+2pPZSiZR+npt9idpq9b8Nf+K4jH7Avf KtG4JjEQf3O/2GuwLHU1ceEofeuxDiN5LocZDN5k= Received: by quaco.ghostprotocols.net (Postfix, from userid 1000) id 125C7405FF; Mon, 13 Jul 2020 09:17:46 -0300 (-03) Date: Mon, 13 Jul 2020 09:17:46 -0300 From: Arnaldo Carvalho de Melo To: Alexey Budankov , Peter Zijlstra Message-ID: <20200713121746.GA7029@kernel.org> References: <76718dc6-5483-5e2e-85b8-64e70306ee1f@linux.ibm.com> <7776fa40-6c65-2aa6-1322-eb3a01201000@linux.intel.com> <20200710170911.GD7487@kernel.org> <0d2e2306-22b2-a730-dc3f-edb3538b6561@linux.intel.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <0d2e2306-22b2-a730-dc3f-edb3538b6561@linux.intel.com> X-Url: http://acmel.wordpress.com X-Mailman-Approved-At: Mon, 13 Jul 2020 12:53:51 +0000 Subject: Re: [Intel-gfx] [PATCH v8 00/12] Introduce CAP_PERFMON to secure system performance monitoring and observability X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Ravi Bangoria , Song Liu , Andi Kleen , linux-man@vger.kernel.org, "linux-doc@vger.kernel.org" , "selinux@vger.kernel.org" , linux-kernel , "intel-gfx@lists.freedesktop.org" , Igor Lubashev , Alexei Starovoitov , Stephane Eranian , James Morris , "linux-security-module@vger.kernel.org" , Ingo Molnar , Namhyung Kim , Thomas Gleixner , Jiri Olsa , Serge Hallyn Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Em Mon, Jul 13, 2020 at 12:48:25PM +0300, Alexey Budankov escreveu: > = > 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 f= or > >>>> 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 r= ole > >>>> for performance monitoring and observability subsystems of the kerne= l. > >>> 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 li= ke > >> kind of expected since currently CAP_PERFMON takes over the related pa= rt > >> 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 ... Everybody should read it, sure :-) = > > I.e., this: > > = > > $ ./perf stat ls > > =A0 Error: > > =A0 Access to performance monitoring and observability operations is li= mited. > > $ > > = > > Could become: > > = > > $ ./perf stat ls > > =A0 Error: > > =A0 Access to performance monitoring and observability operations is li= mited. > > 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 outpu= t. So add a 3rd line: $ ./perf stat ls =A0 Error: =A0 Access to performance monitoring and observability operations is limite= d. Right now only CAP_PERFMON is granted, you may need CAP_SYS_PTRACE. Please read the 'Perf events and tool security' document: https://www.kernel.org/doc/html/latest/admin-guide/perf-security.html > Looks like extending ptrace_may_access() check for perf_events with CAP_P= ERFMON You mean the following? diff --git a/kernel/events/core.c b/kernel/events/core.c index 856d98c36f56..a2397f724c10 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -11595,7 +11595,7 @@ SYSCALL_DEFINE5(perf_event_open, * perf_event_exit_task() that could imply). */ err =3D -EACCES; - if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS)) + if (!perfmon_capable() && !ptrace_may_access(task, PTRACE_MODE_READ_REAL= CREDS)) goto err_cred; } > 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? I tend to agree, Peter? = > 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.h= tml > >> > >>> > >>> Without setting CAP_PERFMON: > >>> > >>> =A0 $ getcap ./perf > >>> =A0 $ ./perf stat -a ls > >>> =A0=A0=A0 Error: > >>> =A0=A0=A0 Access to performance monitoring and observability operatio= ns is limited. > >>> =A0 $ ./perf stat ls > >>> =A0=A0=A0 Performance counter stats for 'ls': > >>> =A0=A0 =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 2.06 msec tas= k-clock:u=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.418 CPUs util= ized > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 context-switches:u=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/sec > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 cpu-migrations:u=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/sec > >>> > >>> With CAP_PERFMON: > >>> > >>> =A0 $ getcap ./perf > >>> =A0=A0=A0 ./perf =3D cap_perfmon+ep > >>> =A0 $ ./perf stat -a ls > >>> =A0=A0=A0 Performance counter stats for 'system wide': > >>> =A0=A0 =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 142.42 msec cpu-clo= ck=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0 25.062 CPUs util= ized > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 182=A0=A0=A0=A0= =A0 context-switches=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.001 M/sec > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 48=A0=A0=A0=A0= =A0 cpu-migrations=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.337 K/sec > >>> =A0 $ ./perf stat ls > >>> =A0=A0=A0 Error: > >>> =A0=A0=A0 Access to performance monitoring and observability operatio= ns is limited. > >>> > >>> Am I missing something silly? > >>> > >>> Analysis: > >>> --------- > >>> A bit more analysis lead me to below kernel code fs/exec.c: > >>> > >>> =A0 begin_new_exec() > >>> =A0 { > >>> =A0=A0=A0=A0=A0=A0=A0 ... > >>> =A0=A0=A0=A0=A0=A0=A0 if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_N= ONDUMP || > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 !(uid_eq(current_euid(), current_ui= d()) && > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 gid_eq(current_egid(), curren= t_gid()))) > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 set_dumpable(current->m= m, suid_dumpable); > >>> =A0=A0=A0=A0=A0=A0=A0 else > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 set_dumpable(current->m= m, SUID_DUMP_USER); > >>> > >>> =A0=A0=A0=A0=A0=A0=A0 ... > >>> =A0=A0=A0=A0=A0=A0=A0 commit_creds(bprm->cred); > >>> =A0 } > >>> > >>> When I execute './perf stat ls', it's going into else condition and t= hus sets > >>> dumpable flag as SUID_DUMP_USER. Then in commit_creds(): > >>> > >>> =A0 int commit_creds(struct cred *new) > >>> =A0 { > >>> =A0=A0=A0=A0=A0=A0=A0 ... > >>> =A0=A0=A0=A0=A0=A0=A0 /* dumpability changes */ > >>> =A0=A0=A0=A0=A0=A0=A0 if (... > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 !cred_cap_issubset(old, new)) { > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 if (task->mm) > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0= set_dumpable(task->mm, suid_dumpable); > >>> =A0 } > >>> > >>> !cred_cap_issubset(old, new) fails for perf without any capability an= d 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) wi= th > >>> 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: > >>> > >>> =A0 perf_event_open() > >>> =A0=A0=A0 ptrace_may_access() > >>> =A0=A0=A0=A0=A0 __ptrace_may_access() { > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 ... > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 if (mm && > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 ((get_dumpa= ble(mm) !=3D SUID_DUMP_USER) && > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 !ptrace_= has_cap(cred, mm->user_ns, mode))) > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 return -EPE= RM; > >>> =A0=A0=A0=A0=A0 } > >>> > >>> This if condition passes for perf with CAP_PERFMON and thus it return= s -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 trie= s 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) sol= ves the > >>> issue. > >>> > >>> =A0 # echo 1 > /proc/sys/fs/suid_dumpable > >>> =A0 $ getcap ./perf > >>> =A0=A0=A0 ./perf =3D cap_perfmon+ep > >>> =A0 $ ./perf stat ls > >>> =A0=A0=A0 Performance counter stats for 'ls': > >>> =A0=A0 =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 1.47 msec tas= k-clock=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.806 CPUs = utilized > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 context-switches=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/sec > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 cpu-migrations=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/s= ec > >>> > >>> Workaround2: > >>> Using CAP_SYS_PTRACE along with CAP_PERFMON solves the issue. > >>> > >>> =A0 $ cat /proc/sys/fs/suid_dumpable > >>> =A0=A0=A0 2 > >>> =A0 # setcap "cap_perfmon,cap_sys_ptrace=3Dep" ./perf > >>> =A0 $ ./perf stat ls > >>> =A0=A0=A0 Performance counter stats for 'ls': > >>> =A0=A0 =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 1.41 msec tas= k-clock=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.826 CPUs = utilized > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 context-switches=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/sec > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 cpu-migrations=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/s= ec > >>> > >>> Workaround3: > >>> Adding CAP_PERFMON to parent of perf (/bin/bash) also solves the issu= e. > >>> > >>> =A0 $ cat /proc/sys/fs/suid_dumpable > >>> =A0=A0=A0 2 > >>> =A0 # setcap "cap_perfmon=3Dep" /bin/bash > >>> =A0 # setcap "cap_perfmon=3Dep" ./perf > >>> =A0 $ bash > >>> =A0 $ ./perf stat ls > >>> =A0=A0=A0 Performance counter stats for 'ls': > >>> =A0=A0 =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 1.47 msec tas= k-clock=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.806 CPUs = utilized > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 context-switches=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/sec > >>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0= =A0=A0 cpu-migrations=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 #=A0=A0=A0 0.000 K/s= ec > >>> > >>> - Ravi > > = -- = - Arnaldo _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx