On Fri, Mar 25, 2016 at 22:43 +0100, Mark Kettenis wrote: > > From: Mike Belopuhov <m...@belopuhov.com> > > Date: Thu, 24 Mar 2016 21:33:25 +0100 > > > > On 24 March 2016 at 20:53, Mark Kettenis <mark.kette...@xs4all.nl> wrote: > > > Diff below makes aesni crypto "mpsafe". It adds a CRYPTOCAP_F_MPSAFE > > > flag that makes the crypto framework dispatch to an mpsafe taskq if it > > > is set. In order to make the aesni_process() function that does the > > > actual crypto work mpsafe, it moves the code over to a per-session > > > working buffer. This obviously increases the cost of setting up a > > > session a bit. Directly invoked crypto operations remain locked. My > > > main laptop seem to do ipsec fine with this, but I'm not sure if there > > > is an actual performance gain. So it would probably be good if > > > somebody could benchmark this. > > > > > > Comments? > > > > It should be fairly simple to use SRP for the session list so that you > > don't have to take the mutex in the aesni_process. Other than that I > > think this should go in once tests are made and people a happy with > > the results. > > Actually, I'm not sure that's possible. I think aesni_newsession() > and aesni_freesession() can be called from interrupt context or at > least are not supposed to sleep.
I thought that it's not possible, but it looks like at least tdb_free can be called anywhere and it will trigger freesession, which is a tad unfortunate IMO, yet understandable. > They would call > SRPL_INSERT_HEAD_LOCKED() and SRPL_REMOVE_LOCKED(), which can only be > called from process context. > Not process context though, but under a lock, which wouldn't save us the mtx_enter anyway. > > Another question I have is why are you using an IPL_HIGH mutex? > > The crypto code may be called from any subsystem, so that is the only > safe choice. Well, it's not called from all subsystems, just those that use IPL_BIO and IPL_NET for interrupts and IPL_HIGH just includes AUDIO and CLOCK which don't call any crypto(9) code. So what are you trying to protect it from or is it just a placeholder value? I'm fine if it is, just curious.
