[Xen-bugs] [Bug 1140] New: Xen and Ethernet Padding
Summary: Xen and Ethernet Padding
I made some tests about ethernet padding ( an expected behavior from
ethernet 10/100 on-wire, that needs at least 60 bytes of data to be able to
CRC. NIC chipset has the job of fullfill the packet if it is smaller than this
- also called "padding" ) and I notice a behavior that is not expected (at
least for me).
>>> Ethernet Padding Sender / Receiver Matrix <<<
\RECEIVER | Dom0_vif |FullV_DomU_2|ParaV_DomU_2|SMP_ParaV_DomU_2
Dom0_vif#######| No (OK) | Yes (OK) | No (NOK) | No (NOK)
FullV_DomU#####| No (NOK) | Yes (OK) | No (NOK) | No (NOK)
ParaV_DomU####| No (NOK) | Yes (OK) | No (NOK) | No (NOK)
SMP_ParaV_DomU| No (NOK) | Yes (OK) | No (NOK) | No (NOK)
My setup is a Centos 5.1 (2.6.18-53.1.4.el5xen) / Xen (3.0.3) in a
bridge network configuration.
The results shows that only FullVirtualization do padding, and only when
receives a packet (not when send it).
You could ask "why padding if you dont have a physical transport?". If
you are playing with some L2 protocol (like AoE), this behavior do impact.
A direct related problem: it's not possible to export AoE disks from
Dom0 and other Paravirtualized guests (AoE blade server verifies if the packet
has a minimal size - 60 bytes) to any local VM. I do not known if another
layer2 protocols has the same problem, but I think that is a expected feature
that a virtual NIC works as a physical one.
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
Xen-bugs mailing list
|<Prev in Thread]
||[Next in Thread>|
- [Xen-bugs] [Bug 1140] New: Xen and Ethernet Padding,