linux-kernel - Re: [RFC PATCH v5 2/2] arm64: Create a list of SYM_CODE functions, check return PC against list

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]

Message-ID: <20210604162415.GF4045@sirena.org.uk>
Date:   Fri, 4 Jun 2021 17:24:15 +0100
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, catalin.marinas@....com, will@...nel.org,
        jmorris@...ei.org, pasha.tatashin@...een.com, jthierry@...hat.com,
        linux-arm-kernel@...ts.infradead.org,
        live-patching@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v5 2/2] arm64: Create a list of SYM_CODE functions,
 check return PC against list

On Wed, May 26, 2021 at 04:49:17PM -0500, madvenka@...ux.microsoft.com wrote:

> The unwinder should check if the return PC falls in any function that
> is considered unreliable from an unwinding perspective. If it does,
> mark the stack trace unreliable.

Reviwed-by: Mark Brown <broonie@...nel.org>

However it'd be good for someone else to double check this as it's
entirely possible that I've missed some case here.

> + * Some special cases covered by sym_code_functions[] deserve a mention here:

> + *	- All EL1 interrupt and exception stack traces will be considered
> + *	  unreliable. This is the correct behavior as interrupts and exceptions
> + *	  can happen on any instruction including ones in the frame pointer
> + *	  prolog and epilog. Unless stack metadata is available so the unwinder
> + *	  can unwind through these special cases, such stack traces will be
> + *	  considered unreliable.
> + *

If you're respinning this it's probably also worth noting that we only
ever perform reliable stack trace on either blocked tasks or the current
task which should if my reasoning is correct mean that the fact that
the exclusions here mean that we avoid having to worry about so many
race conditions when entering and leaving functions.  If we got
preempted at the wrong moment for one of them then we should observe the
preemption and mark the trace as unreliable due to that which means that
any confusion the race causes is a non-issue.

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