Re: [Swan-dev] text_said() calls in create_xfrm_migrate_sa() puzzle me

2021-06-21 Thread Antony Antony
Hugh, you spotted a bug in debug output. I think the idea is to log @ reqid=. either dst or src would change. I also recollect trying to log the ports when there is encap. debug output is in: https://testing.libreswan.org/v4.4-483-g292ec75828-main/ikev2-mobike-05-gcm/OUTPUT/north.pluto.log.gz

Re: [Swan-dev] text_said() calls in create_xfrm_migrate_sa() puzzle me

2021-06-21 Thread Paul Wouters
On Mon, 21 Jun 2021, Andrew Cagney wrote: So now we've got a double polarity reversal with an SPI twist.  11 points! Before things are getting "fixed", please do note that we have passing interop testcases:

Re: [Swan-dev] text_said() calls in create_xfrm_migrate_sa() puzzle me

2021-06-21 Thread Andrew Cagney
On Sat, 19 Jun 2021 at 15:35, D. Hugh Redelmeier wrote: > I don't understand this code but I'm "improving" it, based on symmetry. > I've hit a snag. Advice welcome! > > There are four calls to text_said. Here's what they look like, with > details removed: > > I'm a little surprised and

[Swan-dev] text_said() calls in create_xfrm_migrate_sa() puzzle me

2021-06-19 Thread D. Hugh Redelmeier
I don't understand this code but I'm "improving" it, based on symmetry. I've hit a snag. Advice welcome! There are four calls to text_said. Here's what they look like, with details removed: if (endpoint_is_specified(st->st_mobike_local_endpoint)) { if (dir ==