linux-kernel - Re: [PATCH v13 04/11] arm64: Split unwind

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:   Wed, 2 Feb 2022 18:44:03 +0000
From:   Mark Brown <broonie@...nel.org>
To:     madvenka@...ux.microsoft.com
Cc:     mark.rutland@....com, jpoimboe@...hat.com, ardb@...nel.org,
        nobuta.keiya@...itsu.com, sjitindarsingh@...il.com,
        catalin.marinas@....com, will@...nel.org, jmorris@...ei.org,
        linux-arm-kernel@...ts.infradead.org,
        live-patching@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v13 04/11] arm64: Split unwind_init()

On Mon, Jan 17, 2022 at 08:56:01AM -0600, madvenka@...ux.microsoft.com wrote:

> +/*
> + * TODO: document requirements here.
> + */
> +static inline void unwind_init_from_regs(struct unwind_state *state,
> +					 struct pt_regs *regs)

> +/*
> + * TODO: document requirements here.
> + *
> + * Note: this is always inlined, and we expect our caller to be a noinline
> + * function, such that this starts from our caller's caller.
> + */
> +static __always_inline void unwind_init_from_current(struct unwind_state *state)

> +/*
> + * TODO: document requirements here.
> + *
> + * The caller guarantees that the task is not running.
> + */
> +static inline void unwind_init_from_task(struct unwind_state *state,
> +					 struct task_struct *task)

Other than the obvious gap this looks good to me.  For _current() I
don't think we've got any particular requirements other than what's
documented.  For the others I think the main thing is that trying to
walk the stack of a task that is actively executing is going to be a bad
idea so we should say that the task shouldn't be running, but in general
given that one of the main use cases is printing diagnostics on error
we shouldn't have too many *requirements* for calling these.

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)