diff options
Diffstat (limited to 'results/classifier/118/all/808588')
| -rw-r--r-- | results/classifier/118/all/808588 | 126 |
1 files changed, 126 insertions, 0 deletions
diff --git a/results/classifier/118/all/808588 b/results/classifier/118/all/808588 new file mode 100644 index 00000000..10419d80 --- /dev/null +++ b/results/classifier/118/all/808588 @@ -0,0 +1,126 @@ +hypervisor: 0.991 +graphic: 0.990 +assembly: 0.990 +risc-v: 0.989 +device: 0.989 +debug: 0.988 +architecture: 0.987 +mistranslation: 0.987 +arm: 0.987 +register: 0.986 +peripherals: 0.986 +user-level: 0.985 +boot: 0.983 +semantic: 0.983 +virtual: 0.982 +KVM: 0.981 +permissions: 0.979 +ppc: 0.977 +performance: 0.977 +network: 0.976 +kernel: 0.972 +TCG: 0.969 +PID: 0.968 +vnc: 0.965 +socket: 0.964 +x86: 0.960 +VMM: 0.956 +files: 0.951 +i386: 0.856 + +Netperf tests cause i82551 network down + +1. boot up a guest with 82551 nic +# qemu-kvm -net nic,model=i82551 ... +2. launch netperf server in the guest +3.on the host +for b in 32 64 128 256 512 1024 1460 2048 4096 8192 9000 16384 32768 65495 65507 +do ./netperf -t TCP_STREAM -f m -H <guest ip> -P 0 -l 10 -- -m $b +done + +for b in 32 64 128 256 512 1024 1460 2048 4096 8192 9000 16384 32768 65495 65507 +do ./netperf -t UDP_STREAM -f m -H <guest ip> -P 0 -l 10 -- -m $b +done + + +Result: +Guest network becomes down + + +netperf client output: +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 32 + 87380 16384 32 10.97 19.61 +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 64 + 87380 16384 64 11.55 79.68 +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 128 + 87380 16384 128 10.16 14.20 +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 256 + 87380 16384 256 11.17 12.85 +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 512 + 87380 16384 512 10.01 16.38 +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 1024 +Interrupted system call +netperf: remote error 4./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 1460 +establish control: are you sure there is a netserver listening on 10.66.9.39 at port 12865? +establish_control could not establish the control connection from 0.0.0.0 port 0 address family AF_UNSPEC to 10.66.9.39 port 12865 address family AF_UNSPEC +./netperf -t TCP_STREAM -f m -H 10.66.9.39 -P 0 -l 10 -- -m 2048 + + +qemu debug message: +.... +EE100 nic_receive command 0x0000, link 0x3d3e6822, addr 0xffffffff, size 1518 +EE100 nic_can_receive 0x29a0180 +EE100 nic_receive 0x29a0180 received broadcast, len=60 +EE100 nic_receive Receive buffer (0 bytes) too small for data (60 bytes); data truncated +EE100 nic_receive command 0x8000, link 0x37b32022, addr 0xffffffff, size 0 + ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +EE100 nic_receive receive: Running out of frames + ^^^^^^^^^^^^^^^^^^^^^^^^ +EE100 eepro100_write1 addr=Command/Status+1 val=0x20 + + + +kvm version: +qemu-kvm: commit 525e3df73e40290e95743d4c8f8b64d8d9cbe021 +Date: Mon Jul 4 13:36:06 2011 +0300 + +related fix in driver: +http://marc.info/?l=linux-netdev&m=118859393206222&w=3 + + +1. When bug reproduces, we can only capture arp request in the tap device. + +# tcpdump -i tap0 +tcpdump: WARNING: tap0: no IPv4 address assigned +tcpdump: verbose output suppressed, use -v or -vv for full protocol decode +listening on tap0, link-type EN10MB (Ethernet), capture size 65535 bytes +16:14:11.741203 ARP, Request who-has 10.66.8.167 tell 10.66.8.252, length 28 +16:14:12.741186 ARP, Request who-has 10.66.8.167 tell 10.66.8.252, length 28 +16:14:13.741183 ARP, Request who-has 10.66.8.167 tell 10.66.8.252, length 28 + +2. Execute 'system_reset' in qemu monitor, guest could not get ip address. + +# tcpdump -i tap0 +tcpdump: WARNING: tap0: no IPv4 address assigned +tcpdump: verbose output suppressed, use -v or -vv for full protocol decode +listening on tap0, link-type EN10MB (Ethernet), capture size 65535 bytes +16:16:21.508588 IP6 :: > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28 +16:16:21.682585 IP6 :: > ff02::1:ff12:3456: ICMP6, neighbor solicitation, who has fe80::5054:ff:fe12:3456, length 24 +16:16:22.682616 IP6 fe80::5054:ff:fe12:3456 > ff02::2: ICMP6, router solicitation, length 16 +16:16:23.105303 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 52:54:00:12:34:56 (oui Unknown), length 300 +16:16:25.837597 IP6 fe80::5054:ff:fe12:3456 > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28 +16:16:26.682594 IP6 fe80::5054:ff:fe12:3456 > ff02::2: ICMP6, router solicitation, length 16 +16:16:30.111097 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 52:54:00:12:34:56 (oui Unknown), length 300 +16:16:30.682598 IP6 fe80::5054:ff:fe12:3456 > ff02::2: ICMP6, router solicitation, length 16 +16:16:39.112289 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 52:54:00:12:34:56 (oui Unknown), length 300 +16:16:43.107232 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 52:54:00:12:34:56 (oui Unknown), length 300 + + +The patch mentioned in comment #3 had been included in the kernel here: +http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=7734f6e6bcd7ba78b00e93e74a4ddafd9886cdea +So I guess we can close this bug nowadays? Or can you still reproduce this issue with the current kernel and current version of QEMU? + +There hasn't been a reply to my question in the last comment within +months, so I assume nobody cares about this anymore. So I'm closing this +ticket now... + |