similar to: xennet windows pv performance

With GPLPV under 2.6.30, GPLPV gets the following from the ring: ring slot n (first buffer): status (length) = 54 bytes offset = 0 flags = NETRXF_extra_info (possibly csum too but not relevant) ring slot n + 1 (extra info) gso.size (mss) = 1460 Because NETRXF_extra_info is not set, that''s all I get for that packet. In the IP header though, the total length is 1544 (which in itself

With GPLPV under 2.6.30, GPLPV gets the following from the ring: ring slot n (first buffer): status (length) = 54 bytes offset = 0 flags = NETRXF_extra_info (possibly csum too but not relevant) ring slot n + 1 (extra info) gso.size (mss) = 1460 Because NETRXF_extra_info is not set, that''s all I get for that packet. In the IP header though, the total length is 1544 (which in itself

Hello, I installed the signed drivers from http://wiki.univention.de/index.php?title=Installing-signed-GPLPV-drivers and I ran into a BSOD on a Windows 2008 Server R2 Enterprise domU with a large number of vcpu''s. The BSOD is related to xennet.sys. After some trials I found that it runs fine up to 15 cores. From 16 or more, the BSOD kicks in when booting the domU. The hardware (4 times

Hello, My testing machine has 2 quad-core CPU (It supports hyperthreading, but i disable it in BIOS). I uses Xen 4.0.1 as the hypervisor. When I use 8 VMs to conduct a test, CPU affinity of the VMs is very strange. Like this: vm_name vcpu_num cpu_affinity Domain-0 8 any VM1 4 1,3,5,7 VM2 4 1,3,5,7 VM3 4 1,3,5,7 VM4 4

Hi all, I am working with libvirt and I am trying to set cpu affinity. Now I can always use      virsh vcpupin <domain_name> <vcpu> <pcpu> to pin vcpus to pcpus. I want to do it using Python API. Now, there is function pinVcpu which is supposed to do that. But this is not working. For example I gave      dom.pinVcpu(0,1) but still my vcpu affinity is for all the pcpus. The

Hello, In http://wiki.xensource.com/xenwiki/Xen4.0 we mention the following: "Better performance and scalability: 128 vcpus per guest, 1 TB of RAM per host, 128 physical CPUs per host (as a default, can be compile-time increased to lots more)." But what''s the max supported amount of memory for a (64bit) guest? -- Pasi _______________________________________________

Hi All, I am new to Xen, and would like to know if anyone can help with a problem I have. I have a dual - Quad Core Intel 5535 VT 2.66 server, with 24G ram running CentOS5 dom0 and domU (both 64 bit). Everything works great and I am SUPER impressed with the efficiency of Xen (i have always been a UML man). I would however, like to run a single domU domain, with a single VCPU, but get the power

Hi. I have the same problem, which described in: http://lists.xensource.com/archives/html/xen-users/2010-07/msg00241.html I''m using Debian Squeeze (2.6.32-5-xen-amd64) in Dom0, Xen 4.0.1-rc6, Windows 2008, 2003 x86_64 in domU After installing drivers from http://www.xen.org/files/XenCloud/Software/0.5/XCP-base-0.5.iso, system stops on progress bar in normal boot, but ok in safe mode. In

Hi, I''ve just attempted several times to install Fedora Linux domU''s on a Fedora 6 [32-bit xen kernel 2.6.20-1.2944], and it keeps crashing at various points when trying to format domU''s virtual disk. I''ve tried allocating 1 vcpu and 2 vcpus, and it still keeps crashing dom0. The machine has a total of two hyperthreaded 3.2ghz Xeon cpus, and 3GB of RAM.

On a selection of boxes, ntpd running in Xen Dom0 reproducibly exhibits extermely high noise/jitter. Switching back to -default, non-xen kernel ntpd runs with very low jitter/noise. Question -- how can I ''tame'' ntpd noise & jitter when running in Dom0? Is the problem a config issue, or a bug? Already reported this downstream; everybody''s "stumped".

On a selection of boxes, ntpd running in Xen Dom0 reproducibly exhibits extermely high noise/jitter. Switching back to -default, non-xen kernel ntpd runs with very low jitter/noise. Question -- how can I ''tame'' ntpd noise & jitter when running in Dom0? Is the problem a config issue, or a bug? Already reported this downstream; everybody''s "stumped".

Hi experts, Q1:What''s the different for "dom0_max_vcpus=4 dom0_vcpus_pin" and "dom0_max_vcpus=4" ? which will get better performance Q2: dom0_max_vcpus=4 means "core0-3 will be just used by dom0" or means "4 cores(not dedicate cores) will be used by dom0, eg: core2-5 or core3-6? Q3.what does mean "nosmp" , xen, dom0,domU, will just use one

Hi experts, Q1:What''s the different for "dom0_max_vcpus=4 dom0_vcpus_pin" and "dom0_max_vcpus=4" ? which will get better performance Q2: dom0_max_vcpus=4 means "core0-3 will be just used by dom0" or means "4 cores(not dedicate cores) will be used by dom0, eg: core2-5 or core3-6? Q3.what does mean "nosmp" , xen, dom0,domU, will just use one

Running GPL PV Windows drivers on Windows 2003 Server Standard Edition SP2 - The current GPL PV Windows xennet driver source found in the repository will BSOD if the option "Large Send Offload" is changed to disabled or if the driver is disabled. Tracked down the issue via WinDbg – If a buffer is in the free list prior to the function XenNet_RxBufferFree() being called via the function

Running GPL PV Windows drivers on Windows 2003 Server Standard Edition SP2 - The current GPL PV Windows xennet driver source found in the repository will BSOD if the option "Large Send Offload" is changed to disabled or if the driver is disabled. Tracked down the issue via WinDbg – If a buffer is in the free list prior to the function XenNet_RxBufferFree() being called via the

Hi there. I have been dealing with a conundrum off and on with a VPS of mine. Windows 2003 Standard 32 bit on a Xen HVM instance. Been trying to get the GPLPV drivers working to help boost performance but always seem to run into one major issue: The PV network device looks like it works but actual connectivity is non-existant. Everything else in the GPLPV driver suite works fine. What I want to

Hi all, This is the second take of a patch that I submitted some time ago for allowing specifying vcpu pinning taking NUMA nodes into account. IOW, something like this: * "nodes:0-3": all pCPUs of nodes 0,1,2,3;  * "nodes:0-3,^node:2": all pCPUS of nodes 0,1,3;  * "1,nodes:1-2,^6": pCPU 1 plus all pCPUs of nodes 1,2    but not pCPU 6; v1 was a single patch, this is

after shutdown many domu, the memory is not release to dom0, before:  host: 8g phy memory 16g swap (4g used) dom0: 600m domu: consumed all memory, no new domu could be created after shutdown many domu:  host: 8g phy memory 16g swap (4g used) dom0: 600m domu: 2g why memory of dom0 is still 600m? how to release more to dom0? pls advise. Yahoo!香港提供網上安全攻略，教你如何防範黑客! 請前往

Hello, Fedora 19 running Xen 4.2.3 rpms, "xl list" after dom0 reboot: [root@localhost ~]# xl list Name ID Mem VCPUs State Time(s) (null) 0 1024 8 r----- 16.3 Why is dom0''s name "(null)" ? It seems also xenstore is quite empty: [root@localhost ~]# xenstore-ls tool =

Hi experts, First I post my log root@intel_5500_server:/> ifconfig eth0 Link encap:Ethernet HWaddr 00:15:17:BB:5E:88 inet addr:128.224.165.205 Bcast:128.224.165.255 Mask:255.255.255.0 inet6 addr: fe80::215:17ff:febb:5e88/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2375 errors:0 dropped:0 overruns:0 frame:0