Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
From:   Dylan Hatch <dylanbhatch@google.com>
Date:   Mon, 3 Apr 2023 15:26:41 -0700
Message-ID: <CADBMgpxQ+oM_TrtKRiREcZoZSk=AfenV_bqOk_Vt-Ov5FPHMvw@mail.gmail.com>
Subject: RE: [RFC PATCH v3 00/22] arm64: livepatch: Use ORC for dynamic frame
 pointer validation
To:     misono.tomohiro@fujitsu.com
Cc:     broonie@kernel.org, catalin.marinas@arm.com,
        chenzhongjin@huawei.com, jamorris@linux.microsoft.com,
        jpoimboe@redhat.com, linux-arm-kernel@lists.infradead.org,
        linux-kernel@vger.kernel.org, live-patching@vger.kernel.org,
        madvenka@linux.microsoft.com, mark.rutland@arm.com,
        nobuta.keiya@fujitsu.com, peterz@infradead.org, pmladek@suse.com,
        sjitindarsingh@gmail.com, will@kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

> Then, I noticed that invoke_syscall generates instructions to add random offset
> in sp when RANDOMIZE_KSTACK_OFFSET=y, which is true in the above case.

I'm also seeing this behavior when compiling with
RANDOMIZE_KSTACK_OFFSET=y. I wonder if a special hint type
could/should be added to allow for skipping the reliability check for
stack frames with this randomized offset? Forgive me if this is a
naive suggestion.

Thanks,
Dylan