Tried it. It reads well when what is being impaired is a boolean vis:
if (IMPAIR(send_bogus_isakmp_flag)) {
hdr.isa_flags |= ISAKMP_FLAGS_RESERVED_BIT6;
}
however, think it breaks down for more complex values. For instance, this
would look decidedly weird iv IMPAIR() was used.
fixup->impair = (impair.v1_hash_exchange == exchange
? impair.v1_hash_payload : SEND_NORMAL);
So I suspect we're left with impair.send_bogus_isakmp_flag et.al.
On Wed, 1 Apr 2020 at 19:59, Paul Wouters <[email protected]> wrote:
> Works for me
>
> Sent from my iPhone
>
> On Apr 1, 2020, at 19:38, Andrew Cagney <[email protected]> wrote:
>
>
>
>
> On Wed, 1 Apr 2020 at 18:10, Antony Antony <[email protected]> wrote:
>
>> On Mon, Mar 30, 2020 at 12:07:17PM -0400, Andrew Cagney wrote:
>> > I'm cleaning up the impair code.
>> >
>> > Internally, the old style #define names are in upper case vis:
>> > #define IMPAIR_REPLAY_FORWARD ...
>> > and
>> > IMPAIR(REPLAY_FORWARD)
>> > while the new ones (that take parameters) are in lower case vis:
>> > extern enum send_impairment impair_ike_key_length_attribute
>> > I've merged the two but preserved the names.
>> >
>> > Should the names be made consistent?
>>
>> that would be great. go for it.
>>
>>
> so how about:
> IMPAIR(replay_forward)
> ?
>
>
>> thanks,
>> -antony
>>
> _______________________________________________
> Swan-dev mailing list
> [email protected]
> https://lists.libreswan.org/mailman/listinfo/swan-dev
>
>
_______________________________________________
Swan-dev mailing list
[email protected]
https://lists.libreswan.org/mailman/listinfo/swan-dev
