From: Daniel Borkmann via iovisor-dev <iovisor-dev-9jONkmmOlFHEE9lA1F8Ukti2O/JbrIOy@public•gmane.org>
To: Edward Cree <ecree-s/n/eUQHGBpZroRs9YW3xA@public•gmane.org>,
davem-fT/PcQaiUtIeIZ0/mPfg9Q@public•gmane.org,
Alexei Starovoitov
<alexei.starovoitov-Re5JQEeQqe8AvxtiuMwx3w@public•gmane.org>,
Alexei Starovoitov <ast-b10kYP2dOMg@public•gmane.org>
Cc: netdev-u79uwXL29TY76Z2rM5mHXA@public•gmane.org,
iovisor-dev
<iovisor-dev-9jONkmmOlFHEE9lA1F8Ukti2O/JbrIOy@public•gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public•gmane.org
Subject: Re: [PATCH v3 net-next 02/12] bpf/verifier: rework value tracking
Date: Wed, 28 Jun 2017 19:09:57 +0200 [thread overview]
Message-ID: <5953E2E5.7040106@iogearbox.net> (raw)
In-Reply-To: <2244b48b-f415-3239-6912-cb09f0abc546-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
On 06/27/2017 02:56 PM, Edward Cree wrote:
> Tracks value alignment by means of tracking known & unknown bits.
> Tightens some min/max value checks and fixes a couple of bugs therein.
> If pointer leaks are allowed, and adjust_ptr_min_max_vals returns -EACCES,
> treat the pointer as an unknown scalar and try again, because we might be
> able to conclude something about the result (e.g. pointer & 0x40 is either
> 0 or 0x40).
>
> Signed-off-by: Edward Cree <ecree-s/n/eUQHGBpZroRs9YW3xA@public•gmane.org>
[...]
> +static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
> + struct bpf_insn *insn)
> +{
> + struct bpf_reg_state *regs = env->cur_state.regs, *dst_reg, *src_reg;
> + struct bpf_reg_state *ptr_reg = NULL, off_reg = {0};
> + u8 opcode = BPF_OP(insn->code);
> + int rc;
> +
> + dst_reg = ®s[insn->dst_reg];
> + check_reg_overflow(dst_reg);
> + src_reg = NULL;
> + if (dst_reg->type != SCALAR_VALUE)
> + ptr_reg = dst_reg;
> + if (BPF_SRC(insn->code) == BPF_X) {
> + src_reg = ®s[insn->src_reg];
> + check_reg_overflow(src_reg);
> +
> + if (src_reg->type != SCALAR_VALUE) {
> + if (dst_reg->type != SCALAR_VALUE) {
> + /* Combining two pointers by any ALU op yields
> + * an arbitrary scalar.
> + */
> + if (!env->allow_ptr_leaks) {
> + verbose("R%d pointer %s pointer prohibited\n",
> + insn->dst_reg,
> + bpf_alu_string[opcode >> 4]);
> + return -EACCES;
> + }
> + mark_reg_unknown(regs, insn->dst_reg);
> + return 0;
> + } else {
> + /* scalar += pointer
> + * This is legal, but we have to reverse our
> + * src/dest handling in computing the range
> + */
> + rc = adjust_ptr_min_max_vals(env, insn,
> + src_reg, dst_reg);
> + if (rc == -EACCES && env->allow_ptr_leaks) {
> + /* scalar += unknown scalar */
> + __mark_reg_unknown(&off_reg);
> + return adjust_scalar_min_max_vals(
> + env, insn,
> + dst_reg, &off_reg);
Could you elaborate on this one? If I understand it correctly, then
the scalar += pointer case would mean the following: given I have one
of the allowed pointer types in adjust_ptr_min_max_vals() then the
prior scalar type inherits the ptr type/id. I would then 'destroy' the
pointer value so we get a -EACCES on it. We mark the tmp off_reg as
scalar type, but shouldn't also actual dst_reg be marked as such
like in below pointer += scalar case, such that we undo the prior
ptr_type inheritance?
> + }
> + return rc;
> + }
> + } else if (ptr_reg) {
> + /* pointer += scalar */
> + rc = adjust_ptr_min_max_vals(env, insn,
> + dst_reg, src_reg);
> + if (rc == -EACCES && env->allow_ptr_leaks) {
> + /* unknown scalar += scalar */
> + __mark_reg_unknown(dst_reg);
> + return adjust_scalar_min_max_vals(
> + env, insn, dst_reg, src_reg);
> + }
> + return rc;
> + }
> + } else {
[...]
next prev parent reply other threads:[~2017-06-28 17:09 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-27 12:53 [PATCH v3 net-next 00/12] bpf: rewrite value tracking in verifier Edward Cree via iovisor-dev
[not found] ` <adc11342-737f-4e06-bce3-f0a92b5594a5-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-27 12:56 ` [PATCH v3 net-next 01/12] selftests/bpf: add test for mixed signed and unsigned bounds checks Edward Cree via iovisor-dev
[not found] ` <12c14f90-153c-c2c2-b2e2-834185b7ec3f-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-28 13:51 ` Daniel Borkmann via iovisor-dev
2017-06-27 12:56 ` [PATCH v3 net-next 02/12] bpf/verifier: rework value tracking Edward Cree via iovisor-dev
2017-06-28 15:15 ` Daniel Borkmann
[not found] ` <5953C7FE.1050205-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org>
2017-06-28 16:07 ` Edward Cree via iovisor-dev
[not found] ` <71aae126-352d-d916-d64a-9d4045d0abe9-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-28 19:44 ` Daniel Borkmann via iovisor-dev
[not found] ` <2244b48b-f415-3239-6912-cb09f0abc546-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-28 17:09 ` Daniel Borkmann via iovisor-dev [this message]
[not found] ` <5953E2E5.7040106-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org>
2017-06-28 18:28 ` Edward Cree via iovisor-dev
2017-07-06 20:26 ` Nadav Amit via iovisor-dev
2017-07-06 21:21 ` Nadav Amit via iovisor-dev
[not found] ` <E65DE75F-FC5F-41BC-9F7B-C03D324D8C6F-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-07-07 13:48 ` Edward Cree via iovisor-dev
[not found] ` <abd3c014-f33a-6e86-aead-be38697a2cec-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-07-07 17:45 ` Nadav Amit via iovisor-dev
[not found] ` <3A96914E-3009-4E19-B138-7A636A76D9C8-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-07-08 0:54 ` Nadav Amit via iovisor-dev
2017-07-12 19:13 ` Edward Cree via iovisor-dev
[not found] ` <68a2487e-f706-1b61-5c4c-20ffe6d51127-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-07-12 22:07 ` Nadav Amit via iovisor-dev
[not found] ` <12DCE590-7F67-4639-A825-61A24AC44ED3-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-07-17 17:02 ` Edward Cree via iovisor-dev
2017-06-29 7:48 ` kbuild test robot
2017-06-27 12:57 ` [PATCH v3 net-next 03/12] nfp: change bpf verifier hooks to match new verifier data structures Edward Cree via iovisor-dev
[not found] ` <b8418d35-7d3f-1ee3-2bfa-cdf87e5b7c95-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-28 20:47 ` Daniel Borkmann via iovisor-dev
2017-06-29 3:47 ` Jakub Kicinski via iovisor-dev
2017-06-27 12:57 ` [PATCH v3 net-next 04/12] bpf/verifier: track signed and unsigned min/max values Edward Cree via iovisor-dev
2017-06-27 12:58 ` [PATCH v3 net-next 05/12] bpf/verifier: more concise register state logs for constant var_off Edward Cree via iovisor-dev
2017-06-27 12:58 ` [PATCH v3 net-next 06/12] selftests/bpf: change test_verifier expectations Edward Cree via iovisor-dev
2017-06-27 12:59 ` [PATCH v3 net-next 07/12] selftests/bpf: rewrite test_align Edward Cree via iovisor-dev
2017-06-27 12:59 ` [PATCH v3 net-next 08/12] selftests/bpf: add a test to test_align Edward Cree via iovisor-dev
2017-06-27 12:59 ` [PATCH v3 net-next 09/12] selftests/bpf: add test for bogus operations on pointers Edward Cree via iovisor-dev
2017-06-27 12:59 ` [PATCH v3 net-next 10/12] selftests/bpf: don't try to access past MAX_PACKET_OFF in test_verifier Edward Cree via iovisor-dev
2017-06-27 13:00 ` [PATCH v3 net-next 11/12] selftests/bpf: add tests for subtraction & negative numbers Edward Cree via iovisor-dev
2017-06-27 13:00 ` [PATCH v3 net-next 12/12] selftests/bpf: variable offset negative tests Edward Cree via iovisor-dev
2017-06-28 13:50 ` [PATCH v3 net-next 00/12] bpf: rewrite value tracking in verifier Daniel Borkmann
[not found] ` <5953B436.6030506-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org>
2017-06-28 14:11 ` Edward Cree via iovisor-dev
[not found] ` <788035e1-1974-b48e-3008-d294194a8b05-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-28 20:38 ` Daniel Borkmann via iovisor-dev
[not found] ` <595413AA.40502-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org>
2017-06-28 21:37 ` Alexei Starovoitov via iovisor-dev
2017-06-30 16:44 ` Edward Cree via iovisor-dev
[not found] ` <91267d15-652a-16d9-4ee9-42958bd842aa-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-06-30 17:34 ` [TEST PATCH] bpf/verifier: roll back ptr&const handling, and fix signed bounds Edward Cree via iovisor-dev
2017-06-30 18:15 ` [PATCH v3 net-next 00/12] bpf: rewrite value tracking in verifier Alexei Starovoitov via iovisor-dev
[not found] ` <ff1ef3ef-71ef-4c23-ede4-b8caeac561ee-b10kYP2dOMg@public.gmane.org>
2017-07-04 19:22 ` Edward Cree via iovisor-dev
[not found] ` <5fa61129-fa82-1607-3363-dfad86aecf1e-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-07-04 22:28 ` Daniel Borkmann via iovisor-dev
2017-07-06 18:27 ` Edward Cree
[not found] ` <54b95191-697d-6b15-ec39-438c85e08adc-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
2017-07-07 9:14 ` Daniel Borkmann via iovisor-dev
[not found] ` <595F50F3.2030008-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org>
2017-07-07 12:50 ` Edward Cree via iovisor-dev
2017-07-07 13:05 ` Daniel Borkmann
2017-07-06 14:07 ` Edward Cree via iovisor-dev
2017-07-14 20:03 ` [iovisor-dev] " Y Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5953E2E5.7040106@iogearbox.net \
--to=iovisor-dev-9jonkmmolfhee9la1f8ukti2o/jbrioy@public$(echo .)gmane.org \
--cc=alexei.starovoitov-Re5JQEeQqe8AvxtiuMwx3w@public$(echo .)gmane.org \
--cc=ast-b10kYP2dOMg@public$(echo .)gmane.org \
--cc=daniel-FeC+5ew28dpmcu3hnIyYJQ@public$(echo .)gmane.org \
--cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public$(echo .)gmane.org \
--cc=ecree-s/n/eUQHGBpZroRs9YW3xA@public$(echo .)gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox