Hi Alexandre, Could you compile with `make debug`, so that dmesg gets plenty of debugging output? That will let me see if this is related to handshakes or to transport messages. An initial theory, based on the stacktrace there involving fmonty and such, is that the 32-bit implementation of curve25519-donna uses a lot of stack space, and if you're on a kernel that puts the softirq stack and the process stack in the same place, this could be an issue. Also, would you mind sending your kernel's .config?
Jason _______________________________________________ WireGuard mailing list [email protected] https://lists.zx2c4.com/mailman/listinfo/wireguard
