From: Eduard Zingerman <eddyz87@gmail.com>
To: Andrii Nakryiko <andrii@kernel.org>,
bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net,
martin.lau@kernel.org
Cc: kernel-team@meta.com
Subject: Re: [PATCH v2 bpf-next 06/10] bpf: preserve constant zero when doing partial register restore
Date: Tue, 21 Nov 2023 18:20:00 +0200 [thread overview]
Message-ID: <d2cc304ac744e2663c8802e3853ed1e948743b32.camel@gmail.com> (raw)
In-Reply-To: <20231121002221.3687787-7-andrii@kernel.org>
On Mon, 2023-11-20 at 16:22 -0800, Andrii Nakryiko wrote:
> Similar to special handling of STACK_ZERO, when reading 1/2/4 bytes from
> stack from slot that has register spilled into it and that register has
> a constant value zero, preserve that zero and mark spilled register as
> precise for that. This makes spilled const zero register and STACK_ZERO
> cases equivalent in their behavior.
>
> Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
> ---
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
[...]
> + if (spill_cnt == size &&
> + tnum_is_const(reg->var_off) && reg->var_off.value == 0) {
> + __mark_reg_const_zero(&state->regs[dst_regno]);
> + /* this IS register fill, so keep insn_flags */
> + } else if (zero_cnt == size) {
> + /* similarly to mark_reg_stack_read(), preserve zeroes */
> + __mark_reg_const_zero(&state->regs[dst_regno]);
> + insn_flags = 0; /* not restoring original register state */
nit: In case if there would be v3, could you please
leave a comment here, something like below:
when check_stack_write_fixed_off() puts STACK_ZERO marks
for writes, not aligned on register boundary, it marks source
registers precise. Thus, additional precision propagation is
necessary in this case and insn_flags could be cleared.
or something along these lines?
> + } else {
> + mark_reg_unknown(env, state->regs, dst_regno);
> + insn_flags = 0; /* not restoring original register state */
> + }
> }
> state->regs[dst_regno].live |= REG_LIVE_WRITTEN;
> } else if (dst_regno >= 0) {
next prev parent reply other threads:[~2023-11-21 16:20 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-21 0:22 [PATCH v2 bpf-next 00/10] Complete BPF verifier precision tracking support for register spills Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 01/10] bpf: support non-r10 register spill/fill to/from stack in precision tracking Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 02/10] selftests/bpf: add stack access precision test Andrii Nakryiko
2023-11-21 0:42 ` Eduard Zingerman
2023-11-21 0:22 ` [PATCH v2 bpf-next 03/10] bpf: fix check for attempt to corrupt spilled pointer Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 04/10] bpf: preserve STACK_ZERO slots on partial reg spills Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 05/10] selftests/bpf: validate STACK_ZERO is preserved on subreg spill Andrii Nakryiko
2023-11-21 0:55 ` Eduard Zingerman
2023-11-21 0:22 ` [PATCH v2 bpf-next 06/10] bpf: preserve constant zero when doing partial register restore Andrii Nakryiko
2023-11-21 16:20 ` Eduard Zingerman [this message]
2023-11-21 18:14 ` Andrii Nakryiko
2023-11-21 20:22 ` Eduard Zingerman
2023-11-21 0:22 ` [PATCH v2 bpf-next 07/10] selftests/bpf: validate zero preservation for sub-slot loads Andrii Nakryiko
2023-11-21 16:20 ` Eduard Zingerman
2023-11-21 18:15 ` Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 08/10] bpf: track aligned STACK_ZERO cases as imprecise spilled registers Andrii Nakryiko
2023-11-21 20:38 ` Alexei Starovoitov
2023-11-21 22:01 ` Andrii Nakryiko
2023-11-21 0:22 ` [PATCH v2 bpf-next 09/10] selftests/bpf: validate precision logic in partial_stack_load_preserves_zeros Andrii Nakryiko
2023-11-21 16:20 ` Eduard Zingerman
2023-11-21 0:22 ` [PATCH v2 bpf-next 10/10] bpf: use common instruction history across all states Andrii Nakryiko
2023-11-21 16:20 ` Eduard Zingerman
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=d2cc304ac744e2663c8802e3853ed1e948743b32.camel@gmail.com \
--to=eddyz87@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=kernel-team@meta.com \
--cc=martin.lau@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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