On 03/08/18 13:37, Ondrej Mosnacek wrote:
> It turns out I had misunderstood how the x86_match_cpu() function works.
> It evaluates a logical OR of the matching conditions, not logical AND.
> This caused the CPU feature checks for AEGIS to pass even if only SSE2
> (but not AES-NI) was supported (or vice versa), leading to potential
> crashes if something tried to use the registered algs.
>
> This patch switches the checks to a simpler method that is used e.g. in
> the Camellia x86 code.
>
> The patch also removes the MODULE_DEVICE_TABLE declarations which
> actually seem to cause the modules to be auto-loaded at boot, which is
> not desired. The crypto API on-demand module loading is sufficient.
>
> Fixes: 1d373d4e8e15 ("crypto: x86 - Add optimized AEGIS implementations")
> Fixes: 6ecc9d9ff91f ("crypto: x86 - Add optimized MORUS implementations")
> Signed-off-by: Ondrej Mosnacek <omosnace@xxxxxxxxxx>
I tried this patch on x86_64 with AES-NI and also on system with
SSE but without AES-NI and it works as expected now
(module is loaded only on demand and optimized one is used if available).
If it is worth it, add
Tested-by: Milan Broz <gmazyland@xxxxxxxxx>
Any chance it could still reach 4.18?
Without this patch it actually crashes kernel on x86_64 without AES-NI
but with SSE flags, see https://bugzilla.redhat.com/show_bug.cgi?id=1610180#c4
Thanks,
Milan
