Sujet : Re: Another security vulnerability
De : anton (at) *nospam* mips.complang.tuwien.ac.at (Anton Ertl)
Groupes : comp.archDate : 26. Mar 2024, 17:36:26
Autres entêtes
Organisation : Institut fuer Computersprachen, Technische Universitaet Wien
Message-ID : <2024Mar26.173626@mips.complang.tuwien.ac.at>
References : 1 2 3 4 5 6
User-Agent : xrn 10.11
scott@slp53.sl.home (Scott Lurndal) writes:
Lawrence D'Oliveiro <ldo@nz.invalid> writes:
On Mon, 25 Mar 2024 17:07:16 GMT, Scott Lurndal wrote:
>
Run it in non-cacheable memory. Slow but safe.
...
Running the crypto algorithms (when not offloaded to
on-chip accelerators) using non-cacheable memory as a workaround
until the hardware issues are ameliorated doesn't imply that
all other code needs to run non-cachable.
Then your crypto code is slow and unsafe. The attacker will use the
rest of the application to extract the crypto keys, whether through
the cache side-channel of Spectre, or a prefetcher-based side channel.
- anton
-- 'Anyone trying for "industrial quality" ISA should avoid undefined behavior.' Mitch Alsup, <c17fcd89-f024-40e7-a594-88a85ac10d20o@googlegroups.com>