Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 28 Sep 2013 18:37:23 -0500
From:      Andrew Berg <robotsondrugs@gmail.com>
To:        freebsd-current@freebsd.org
Subject:   Issues with Hyper-V network driver on ALPHA2/3
Message-ID:  <52476833.4050100@gmail.com>

next in thread | raw e-mail | index | archive | help
FreeBSD talkingzebra 10.0-ALPHA3 FreeBSD 10.0-ALPHA3 #0 r255918: Fri Sep 27 17:29:41 CDT 2013
toor@talkingzebra:/usr/obj/usr/src/sys/HYPERV  amd64

Sometimes after a reboot, the interfaces get an IP (statically set), but there is no connectivity. Starting or restarting the netif service
causes this to appear every time:
--- syscall (54, FreeBSD ELF64, sys_ioctl), rip = 0x80119b9ca, rsp = 0x7fffffffd488, rbp = 0x7fffffffd4d0 ---
Waiting on "rndis sema" with the following non-sleepable locks held:
exclusive sleep mutex NetVSCLock (network driver) r = 0 (0xfffff80001c4f2b0) locked @ /usr/src/sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c:904
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe00b989a410
kdb_backtrace() at kdb_backtrace+0x39/frame 0xfffffe00b989a4c0
witness_warn() at witness_warn+0x4a8/frame 0xfffffe00b989a580
_cv_timedwait_sbt() at _cv_timedwait_sbt+0x94/frame 0xfffffe00b989a600
_sema_timedwait() at _sema_timedwait+0x74/frame 0xfffffe00b989a640
hv_rf_set_packet_filter() at hv_rf_set_packet_filter+0x184/frame 0xfffffe00b989a680
hv_rf_on_open() at hv_rf_on_open+0x46/frame 0xfffffe00b989a6a0
hn_ifinit() at hn_ifinit+0xbe/frame 0xfffffe00b989a6e0
ether_ioctl() at ether_ioctl+0x8a/frame 0xfffffe00b989a710
hn_ioctl() at hn_ioctl+0x28a/frame 0xfffffe00b989a770
in_control() at in_control+0xbe5/frame 0xfffffe00b989a820
ifioctl() at ifioctl+0xfc7/frame 0xfffffe00b989a8e0
kern_ioctl() at kern_ioctl+0x2ca/frame 0xfffffe00b989a950
sys_ioctl() at sys_ioctl+0x142/frame 0xfffffe00b989a9a0
amd64_syscall() at amd64_syscall+0x265/frame 0xfffffe00b989aab0
Xfast_syscall() at Xfast_syscall+0xfb/frame 0xfffffe00b989aab0
--- syscall (54, FreeBSD ELF64, sys_ioctl), rip = 0x80119b9ca, rsp = 0x7fffffffd488, rbp = 0x7fffffffd4d0 ---

Every time, everything looks good according to ifconfig, but pings to IPs in the same network always time out. If I reboot again, it will
generally come back with networking working just fine, and I have found no way to get the interfaces to work other than rebooting again.
This is happening on ALPHA3 (r255918), but I have noticed it in a previous revision as well (I can't recall the revision, but it was after
the hyperv device was removed from GENERIC). I have the entire dmesg saved (before reboot) from that session if that would help. Also, this
is on client Hyper-V on Windows 8, if that matters.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52476833.4050100>