lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Date:	Fri, 4 Jul 2014 09:37:26 +0000
From:	David Laight <David.Laight@...LAB.COM>
To:	Network Development <netdev@...r.kernel.org>
Subject: e1000e (82579LM) management mode under KVM/QEMU

We are seeing some issues with an e1000e (82579LM) interface if
the 'management mode' stuff is enabled to allow VNC+ console access.
This is an Ivy bridge cpu, so will be a Panther point (PCH7) chipset.

The 82579LM interface and a second 82574L one are 'passed through'
to a KVM/qemu VM and then bonded together.

The link is only coming up at 10M HDX (sometimes FDX) even after
replugging the cable.
(ping -s 60000 confirms that the reported speed is correct.)

Forcing n-way negotiation with ethtool results in the correct
100M/1G speed.

The failing system is running an ubuntu built 3.11 kernel.
I don't see the same problem on a non-VM with 3.14(ish) kernel
on the same motherboard.
Running a more recent kernel on the VM is a bit tricky.

I can't see any obvious changes since 3.11, but the driver code
has a lot of conditions based on the PCH (etc) type.
I'm not sure whether the VM kernel is using the correct functions.
I can't see any trace that indicates which host controller the
e1000e driver thinks the device is connected to.

Any hints on how to determine what is actually going on?

	David



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists