It would seem that FreeBSD is unable to deal correctly with qemu's networking when *more* than one e1000 interface is exported. TX works, but RX seems broken. re(4) in the exact same configuration works fine. Fix: please. How-To-Repeat: # qemu -nographic -hda /vm/hdd-freebsd-i386 \ -net nic -net tap,script=/vm/ifup \ -net nic -net tap,script=/vm/ifup \ -net nic -net tap,script=/vm/ifup \ -net nic -net tap,script=/vm/ifup \ -net nic -net tap,script=/vm/ifup \ -net nic -net tap,script=/vm/ifup Copyright (c) 1992-2012 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD is a registered trademark of The FreeBSD Foundation. FreeBSD 9.0-RELEASE #0: Tue Jan 3 07:15:25 UTC 2012 root@obrian.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC i386 CPU: QEMU Virtual CPU version 0.14.50 (2660.53-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x633 Family = 6 Model = 3 Stepping = 3 Features=0x781abf9<FPU,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,PGE,CMOV,PAT,MMX,FXSR,SSE,SSE2> Features2=0x80800001<SSE3,POPCNT,HV> real memory = 134205440 (127 MB) avail memory = 112107520 (106 MB) [...] em0: <Intel(R) PRO/1000 Legacy Network Connection 1.0.3> port 0xc040-0xc07f mem 0xf2020000-0xf203ffff irq 11 at device 3.0 on pci0 em0: Memory Access and/or Bus Master bits were not set! em0: Ethernet address: 52:54:00:12:34:56 [...] em5: <Intel(R) PRO/1000 Legacy Network Connection 1.0.3> port 0xc180-0xc1bf mem 0xf2160000-0xf217ffff irq 11 at device 8.0 on pci0 em5: Memory Access and/or Bus Master bits were not set! em5: Ethernet address: 52:54:00:12:34:5b [...] # ifconfig em0 10.0.255.2 # ping 10.0.255.1 PING 10.0.255.1 (10.0.255.1): 56 data bytes ping: sendto: Host is down ping: sendto: Host is down ping: sendto: Host is down
Responsible Changed From-To: freebsd-bugs->freebsd-net Over to maintainer(s).
If you get a chance to retest this, can you try this with 10 or head? The driver being used in freebsd is lem(4) and not em(4) for this adapter and I'm curious if its still broken for you.
If re@ has the time, can a short regression for this be run?
We do not test with every knob turned and button pushed for testing. To the OP: Can you please confirm if this is still an issue with 10.2-BETA2 and later?
Tested with the latest FreeBSD11 snapshot and emulators/qemu-devel from ports and this all seems to work just fine now.