[Swan-commit] Changes to ref refs/heads/main

[Andrew Cagney]

New commits:
commit cbfd173fb580eb20de6f16caa6b323b17de89a41
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Nov 7 22:54:17 2020 -0500

    Revert "ikev2: in ikev2_child_outI() use submit_ke_and_nonce()"
    
    Again notice how this is using .st_oakley.ta_dh, and not the pfs group.
    
    -       case STATE_V2_REKEY_IKE_I0:
    -               request_ke_and_nonce("IKE REKEY Initiator KE and nonce ni",
    -                                    st, st->st_oakley.ta_dh,
    -                                    ikev2_child_outI_continue);
    -               break; /* return STF_SUSPEND; */
    
    This reverts commit 3ebd6bef643485d43c48fdc019fa120d00d296d9.

commit 3b96a696267a0d78cc1f638b6835742abf3eb457
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Nov 7 22:51:29 2020 -0500

    Revert "ikev2: in ikev2_child_inIoutR() use submit_ke_and_nonce()"
    
    Notice how the old code checks if .st_pfs_group is non-NULL, but then
    passes in .st_oakley.ta_dh:
    
    -               if (child->sa.st_pfs_group != NULL) {
    -                       request_ke_and_nonce("Child Responder KE and nonce nr",
    -                                            &child->sa, child->sa.st_oakley.ta_dh,
    -                                            ikev2_child_inIoutR_continue);
    -               } else {
    -                       request_nonce("Child Responder nonce nr",
    -                                     &child->sa, ikev2_child_inIoutR_continue);
    -               }
    
    Something subtle going on that the patch didn't mimic.
    
    This reverts commit 77fab9906b5b20800f35a696e344611e28a5ec4a.