linux-kernel - Re: [PATCH v7 tip 4/8] tracing: allow BPF programs to call bpf_trace

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 19 Mar 2015 08:51:29 -0700
From:	Alexei Starovoitov <ast@...mgrid.com>
To:	Steven Rostedt <rostedt@...dmis.org>
CC:	Ingo Molnar <mingo@...nel.org>, Namhyung Kim <namhyung@...nel.org>,
	Arnaldo Carvalho de Melo <acme@...radead.org>,
	Jiri Olsa <jolsa@...hat.com>,
	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
	"David S. Miller" <davem@...emloft.net>,
	Daniel Borkmann <daniel@...earbox.net>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	linux-api@...r.kernel.org, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 tip 4/8] tracing: allow BPF programs to call bpf_trace_printk()

On 3/19/15 8:29 AM, Steven Rostedt wrote:
>> +	/* check format string for allowed specifiers */
>> +	for (i = 0; i < fmt_size; i++)
>
> Even though there's only a single "if" statement after the "for", it is
> usually considered proper to add the brackets if the next line is
> complex (more than one line). Which it is in this case.

ok.

>> +			} else if (fmt[i] == 'p') {
>> +				mod_l[fmt_cnt] = true;
>> +				fmt_cnt++;
>
> So you also allow pointer conversions like "%pS" and "%pF"?

good catch. it's a bug. We shouldn't allow things like pV, pD, etc
Something like pK and pS may be ok, but pF is not because of arch
dependencies. So instead of analyzing all possibilities. I'll allow
%p only. bpf_trace_printk is debug only anyway.

>> +	return __trace_printk(1/* fake ip will not be printed */, fmt,
>> +			      mod_l[0] ? r3 : (u32) r3,
>> +			      mod_l[1] ? r4 : (u32) r4,
>> +			      mod_l[2] ? r5 : (u32) r5);
>
> Does the above work on 32 bit machines? as "%ld" would be (u32), but
> here you are passing in u64.

another great catch. it wouldn't crash, but would print junk. will fix.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/