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

2017-08-15 Thread Andrew Cagney
New commits: commit 3564dd722cc4094c35aa4c5301e1a9213810fa52 Author: Andrew Cagney Date: Tue Aug 15 17:06:33 2017 -0400 ike_alg: "fix" ike_alg_integ_sha2_256_trunkbug's .integ_ikev1_ah_field It has the magic value AUTH_ALGORITHM_HMAC_SHA2_256_TRUNCBUG which needs to match SADB_

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

2017-08-15 Thread Andrew Cagney
New commits: commit 1fb31be88a7f0447e4979268b7a0801d2298ed1e Author: Andrew Cagney Date: Tue Aug 15 14:27:50 2017 -0400 kernel: replace kernel_alg_info() with kernel_alg_encrypt_key_size() Delete some seemingly redundant logic trying to fix an already fixed keylen. Debug-log c

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

2017-08-15 Thread Andrew Cagney
New commits: commit 837ad29dcf018c48ca9c1b59d437563e8640aba9 Author: Andrew Cagney Date: Tue Aug 15 13:31:18 2017 -0400 kernel: eliminate struct kernel_alg_info's .transid and .encryptalg Replace a mashed up switch containg branches with both IKEv1 and IKEv2 enum values with 't

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

2017-08-15 Thread Andrew Cagney
New commits: commit ded27b176cff792e51434509fe7e19f47f503d77 Author: Andrew Cagney Date: Tue Aug 15 15:44:41 2017 -0400 ikev1: when ESP and AEAD, set .integ to &ike_alg_integ_null Fix case missed by 81cde5584b9db8260366c168d5d8218c38bdff8e. Cause of eb3371164753fe12d607fa1e02dc

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

2017-08-15 Thread Andrew Cagney
New commits: commit 350fbaca125ed697f34a21e6ae754535f8cf7c46 Author: Andrew Cagney Date: Tue Aug 15 12:52:38 2017 -0400 kernel: for ESP, eliminate struct kernel_alg_info's .auth and .authalg fields Use ta->integ's .integ_keymat_size and .integ_ikev1_ah_id instead. (and no long

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

2017-08-15 Thread Andrew Cagney
New commits: commit eb3371164753fe12d607fa1e02dc94a23129ffaa Author: Andrew Cagney Date: Tue Aug 15 12:18:25 2017 -0400 kernel: use kernel_alg_is_ok() to confirm ESP algorithm is supported Since the parser uses the exact same check, having it fail suggests things are pretty mes

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

2017-08-15 Thread D. Hugh Redelmeier
New commits: commit 408adf2b00a3dec65f4df7d81dd8560e698e95e4 Author: D. Hugh Redelmeier Date: Tue Aug 15 11:27:38 2017 -0400 kernel_netlink.c: add () to make macros safe ___ Swan-commit mailing list Swan-commit@lists.libreswan.org https://lists.l

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

2017-08-15 Thread Andrew Cagney
New commits: commit 81f1a1cd713f7e187d262808f8bbb5acf1db96ef Merge: c4b85b8 c532e7e Author: Andrew Cagney Date: Tue Aug 15 11:50:31 2017 -0400 kernel: delete kernel_alg_info hardwired_info[] When checking that an ESP/AH crypto-suite was valid, this table would first be searched

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

2017-08-15 Thread Andrew Cagney
New commits: commit c4b85b8c56a87dc523aaec6866fd70fc297e1fdc Merge: 6a92311 d31e6c2 Author: Andrew Cagney Date: Tue Aug 15 11:37:37 2017 -0400 ike_alg: disable IKEv1 IKE AEAD algorithms (AES_CCM AES_GCM) so that the parser rejects them. The parser was using IKEv1 'numbers'