On Fri, Nov 22, 2024 at 04:37:09PM +0800, Herbert Xu wrote: > On Fri, Nov 22, 2024 at 03:44:27PM +0800, Herbert Xu wrote: > > > > This is a bug in the API/driver. Users should not be expected to know > > what kind of a virtual pointer is acceptable. > > > > In this particular case, rsassa-pkcs1.c should be fixed to use the > > crypto_akcipher_sync_encrypt interface. > > Lukas, could you please take a look at this regression? Hm, my impression is that this needs to be fixed in arm64's virt_addr_valid() macro. Let's see what arm64 maintainers have to say about it: https://lore.kernel.org/r/90667b2b7f773308318261f96ebefd1a67133c4c.1732464395.git.lukas@xxxxxxxxx/ Thanks, Lukas
