From: Jiri Olsa <jolsa@kernel.org>
To: Oleg Nesterov <oleg@redhat.com>,
Masami Hiramatsu <mhiramat@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Andrii Nakryiko <andrii@kernel.org>
Cc: bpf@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-trace-kernel@vger.kernel.org, x86@kernel.org,
Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
John Fastabend <john.fastabend@gmail.com>,
Hao Luo <haoluo@google.com>, Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@kernel.org>
Subject: [RFC 0/4] uprobe,bpf: Allow to change app registers from uprobe
Date: Fri, 1 Aug 2025 23:02:34 +0200 [thread overview]
Message-ID: <20250801210238.2207429-1-jolsa@kernel.org> (raw)
hi,
we recently had several requests for tetragon to be able to change
user application function return value or divert its execution through
instruction pointer change.
This *RFC* patchset adds support for uprobe program to change app's
registers including instruction pointer.
There's a hiccup with instruction pointer change.. if uprobe handler
changes instruction pointer, the current code will still execute the
original instruction and increment the (new) ip with its length.
This makes the new instruction pointer bogus and application will
likely crash on illegal instruction execution.
I think if user decides to take execution elsewhere, we can skip
original instruction execution (patch#1), but I might be easily
wrong and overlooking something.. hence RFC ;-)
thoughts? thanks,
jirka
---
Jiri Olsa (4):
uprobe: Do not emulate/sstep original instruction when ip is changed
bpf: Allow uprobe program to change context registers
selftests/bpf: Add uprobe context registers changes test
selftests/bpf: Add uprobe context ip register change test
include/linux/bpf.h | 1 +
kernel/events/core.c | 4 +++
kernel/events/uprobes.c | 3 ++
kernel/trace/bpf_trace.c | 3 +-
tools/testing/selftests/bpf/prog_tests/uprobe.c | 162 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
tools/testing/selftests/bpf/progs/test_uprobe.c | 38 ++++++++++++++++++++++++
6 files changed, 208 insertions(+), 3 deletions(-)
next reply other threads:[~2025-08-01 21:02 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-01 21:02 Jiri Olsa [this message]
2025-08-01 21:02 ` [RFC 1/4] uprobe: Do not emulate/sstep original instruction when ip is changed Jiri Olsa
2025-08-02 10:34 ` Oleg Nesterov
2025-08-04 8:12 ` Jiri Olsa
2025-08-08 18:38 ` Jiri Olsa
2025-08-01 21:02 ` [RFC 2/4] bpf: Allow uprobe program to change context registers Jiri Olsa
2025-08-01 21:02 ` [RFC 3/4] selftests/bpf: Add uprobe context registers changes test Jiri Olsa
2025-08-01 21:02 ` [RFC 4/4] selftests/bpf: Add uprobe context ip register change test Jiri Olsa
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=20250801210238.2207429-1-jolsa@kernel.org \
--to=jolsa@kernel.org \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=haoluo@google.com \
--cc=john.fastabend@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=songliubraving@fb.com \
--cc=x86@kernel.org \
--cc=yhs@fb.com \
/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