Date: Sat, 28 Mar 2015 22:49:59 +0300 From: Slawa Olhovchenkov <slw@zxy.spb.ru> To: Adrian Chadd <adrian@freebsd.org> Cc: "freebsd-hackers@freebsd.org" <freebsd-hackers@freebsd.org> Subject: Re: irq cpu binding Message-ID: <20150328194959.GE23643@zxy.spb.ru> In-Reply-To: <CAJ-Vmo=R5J=LT9ntgZJ4xHZwn8_WEAUC=3wSqPKn3=z8ki_N6A@mail.gmail.com> References: <20150328112035.GZ23643@zxy.spb.ru> <CAJ-Vmo=sxtLgn1HABbWchu1Hm8kZ4Qy%2BscPeD7ezkkBzr%2BQijA@mail.gmail.com> <20150328154031.GA23643@zxy.spb.ru> <CAJ-VmokpWNy-arZ_=4hNbyAvzBYgDWu4F=vYSDwqPgpjaGHk=w@mail.gmail.com> <20150328181026.GB23643@zxy.spb.ru> <CAJ-Vmok=0=C%2BHLvLXEM5pbYp_zP6C498wudAfEA40s4AVpGa_w@mail.gmail.com> <20150328183147.GC23643@zxy.spb.ru> <CAJ-VmokPdnOSASLcrzHtqs_r6GfUW6Q5ap7r9Tnp1OvKhDh20A@mail.gmail.com> <20150328192505.GD23643@zxy.spb.ru> <CAJ-Vmo=R5J=LT9ntgZJ4xHZwn8_WEAUC=3wSqPKn3=z8ki_N6A@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 28, 2015 at 12:33:52PM -0700, Adrian Chadd wrote: > That's done deferred by the bus interrupt wiring. That's something > John's been looking into as part of the general NUMA work (and I'm > trying to debug right now, on dual-socket boxes with ixgbe. :-) > > Look at bus_bind_intr() and the twisty path to intr_event_bind(), then > x86/x86/intr_machdep.c:intr_assign_cpu(), then intr_shuffle_cpus() at > boot, versus what happens via calls to pic_assign_cpu to setup the > wiring. This is very complex to me -- I am not familar with current x86 hardware. Hmm. I see intr_setaffinity and intr_bind. And I don't see using this nor ixgbe nor cxgbe. Anyway, I am re-set CPUs compared to initial assigned in ixgbe/cxgbe drivers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150328194959.GE23643>