>>Ye, this is OpenSUSE 12.3 kernel (both Dom0/DomU is
>>kernel-xen-3.9.2-8.1.g04040b9.x86_64) from 
>>>>http://download.opensuse.org/repositories/Kernel:/HEAD/standard

Sorry, misspelled,

I tested kernel-xen-3.9.2-8.1.g04040b9.x86_64 and kernel-
xen-3.9.2-4.756ee56.x86_64 kernels - results are the same.
-- 
*/Best regards,/*
/Eugene Istomin/




Ye, this is OpenSUSE 12.3 kernel (both Dom0/DomU is kernel 
xen-3.9.2-8.1.g04040b9.x86_64) from 
http://download.opensuse.org/repositories/Kernel:/HEAD/standard


First VM:

template:/home/local # iperf -s 
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 10.251.2.202 port 5001 connected with 10.251.2.201 port 
38196


##
#(after iperf -c 10.251.2.202 -i 2 -f m from the second VM)
##

[   38.447860] BUG: unable to handle kernel paging request at 
ffff88007928b000
[   38.447898] IP: [<ffffffffa001a75c>] netif_poll+0x49c/0xe80 [xennet]
[   38.447927] PGD a83067 PUD a93067 PMD 7fc28067 PTE 
801000007928b065
[   38.447955] Oops: 0003 [#1] SMP 
[   38.447970] Modules linked in: af_packet hwmon domctl crc32_pclmul 
crc32c_intel ghash_clmulni_intel aesni_intel ablk_helper cryptd lrw 
aes_x86_64 xts gf128mul joydev autofs4 scsi_dh_emc scsi_dh_alua 
scsi_dh_rdac scsi_dh_hp_sw scsi_dh xenblk cdrom xennet ata_generic 
ata_piix
[   38.448091] CPU 0 
[   38.448100] Pid: 0, comm: swapper/0 Not tainted 3.9.2-4.756ee56-xen 
#1  
[   38.448125] RIP: e030:[<ffffffffa001a75c>]  [<ffffffffa001a75c>] 
netif_poll+0x49c/0xe80 [xennet]
[   38.448158] RSP: e02b:ffff88007b403d18  EFLAGS: 00010286
[   38.448176] RAX: ffff88007da68cd0 RBX: ffff88007928aec0 RCX: 
ffff88007928b000




This trace is viewed only using xl console, DomUs had no records in logs. 
You are right, may be this is Dom0 trace.

Here is xl-test1.log from Dom0:

libxl: debug: libxl_event.c:503:watchfd_callback: watch w=0x1944130 
wpath=@releaseDomain token=3/0: event epath=@releaseDomain
libxl: debug: libxl.c:998:domain_death_xswatch_callback: 
[evg=0x19435e0:2] from domid=2 nentries=1 rc=1
libxl: debug: libxl.c:1009:domain_death_xswatch_callback: 
[evg=0x19435e0:2]   got=domaininfos[0] got->domain=2
libxl: debug: libxl.c:1036:domain_death_xswatch_callback:  exists 
shutdown_reported=0 dominf.flags=30004
libxl: debug: libxl.c:1048:domain_death_xswatch_callback:  shutdown 
reporting
libxl: debug: libxl.c:1002:domain_death_xswatch_callback: [evg=0] all 
reported
libxl: debug: libxl.c:1066:domain_death_xswatch_callback: domain death 
search done
Domain 2 has shut down, reason code 3 0x3
Action for shutdown reason code 3 is restart
Domain 2 needs to be cleaned up: destroying the domain
libxl: debug: libxl.c:1250:libxl_domain_destroy: ao 0x19438d0: create: 
how=(nil) callback=(nil) poller=0x19435a0
libxl: debug: libxl_dm.c:1266:libxl__destroy_device_model: Device Model 
signaled


-- 
*/Best regards,/*
/Eugene Istomin/


> Moving discussion to Xen-devel
> 
> On Thu, May 16, 2013 at 10:29:56PM +0300, Eugene Istomin wrote:
> > Hello,
> > 
> > I tried to use 3.9.2 kernel with xen 4.2.2/4.3rc1 and in both variants
> > leads to this error in network-intensive load (such as iperf, 100 nginx
> > parallel
> > requests to 1M files and so on):
> It would be more helpful if you can provide info on your configurations
> (Dom0 and DomU), your workload, how to reproduce the bug.
> 
> I run iperf and NFS to test Xen network, but never see any crash like
> this myself.
> 
> > BUG: unable to handle kernel paging request at ffff8800795a3000
> > [   60.246945] IP: [<ffffffffa001a75c>] netif_poll+0x49c/0xe80 [xennet]
> > [   60.246975] PGD a8a067 PUD a9a067 PMD 7fc27067 PTE
> > 80100000795a3065
> > [   60.247004] Oops: 0003 [#1] SMP
> > [   60.247020] Modules linked in: af_packet hwmon domctl 
crc32_pclmul
> > crc32c_intel ghash_clmulni_intel aesni_intel ablk_helper cryptd lrw
> > aes_x86_64 joydev xts gf128mul autofs4 scsi_dh_emc scsi_dh_alua
> > scsi_dh_rdac scsi_dh_hp_sw scsi_dh xenblk cdrom xennet ata_generic
> > ata_piix
> > [   60.247144] CPU 0
> > [   60.247154] Pid: 0, comm: swapper/0 Not tainted 3.9.2-1.g04040b9-
xen
> > #1
> > [   60.247179] RIP: e030:[<ffffffffa001a75c>]  [<ffffffffa001a75c>]
> > netif_poll+0x49c/0xe80 [xennet]
> > ...
> 
> Could you provide fuul stack trace? AFAICT there is no netif_poll in Xen
> netfront/back.
> 
> Presumably this is Dom0 log? (from the domctl module)
> 
> > We have couple of production hypervisors on 3.4 kernels with high-
> > throughput  internal network (VM-to-VM in one Dom0), iperf on them is
> > working well:
> > [  3]  0.0- 2.0 sec  3357 MBytes  14080 Mbits/sec
> > [  3]  2.0- 4.0 sec  2880 MBytes  12077 Mbits/sec
> > [  3]  4.0- 6.0 sec  2909 MBytes  12202 Mbits/sec
> > [  3]  6.0- 8.0 sec  2552 MBytes  10702 Mbits/sec
> > [  3]  8.0-10.0 sec  3616 MBytes  15166 Mbits/sec
> > [  3] 10.0-12.0 sec  3415 MBytes  14324 Mbits/sec
> > 
> > 
> > Seems like a kernel bug, is this related to one of this fixes in
> > linux-next or i need to create new bugreport?