[strongSwan] libhydra

2016-11-14 Thread Joy Latten
Hi, I seek a bit of clarity while understanding the libhydra code and strongswan modularity. From reading the code, the kernel_interface_t in libhydra is somewhat generic. It cannot really do anything, that is, it cannot really talk to the kernel without a kernel_ipsec_t being defined, implemented

Re: [strongSwan] libhydra

2016-11-14 Thread Tobias Brunner
Hi Joy, > Any new plugin for talking > to the kernel would require a kernel_ipsec_t as well. Is this correct? Yes. Regards, Tobias ___ Users mailing list Users@lists.strongswan.org https://lists.strongswan.org/mailman/listinfo/users

Re: [strongSwan] libhydra

2016-11-15 Thread Joy Latten
Hi Tobias, Thank you! So is it safe to conclude that without a plugin, libhydra does nothing? (I think you get return of "NOT_SUPPORTED".) And libhydra is required in order for charon/ike to manage SAs and SPs with kernel. Again, thanks! regards, Joy On 11/15/2016 01:53 AM, Tobias Brunner wrote

Re: [strongSwan] libhydra

2016-11-15 Thread Andreas Steffen
Hi Joy, kernel_ipsec_t just defines a standardized strongSwan interface, but the actual interface code is in the plugins. So if you have an IPsec ESP implementation of your own, e.g. a HW accelerator then you have to write a plugin of your own. Best regards Andreas BTW - In the latest strongSwa