On Fri, Aug 1, 2014 at 3:42 PM, Kir Kolyshkin wrote:
> On 08/01/2014 01:13 PM, Harry Koris wrote:
>
>Hi,
>
> I'm looking to install AMD 13.25 APU Driver so that I can access the
> on-board graphics from within a Centos-x86_64 container. I've added the PCI
> slot that the GPU comes up on usin
On 08/01/2014 01:13 PM, Harry Koris wrote:
> Hi,
>
> I'm looking to install AMD 13.25 APU Driver so that I can access the
> on-board graphics from within a Centos-x86_64 container. I've added
> the PCI slot that the GPU comes up on using the '--pci_add' option but
> the driver still reads out that
Greetings,
- Original Message -
> I'm looking to install AMD 13.25 APU Driver so that I can access the
> on-board graphics from within a Centos-x86_64 container. I've added
> the PCI slot that the GPU comes up on using the '--pci_add' option
> but the driver still reads out that no compati
2014-08-02 0:13 GMT+04:00 Harry Koris :
> Hi,
>
> I'm looking to install AMD 13.25 APU Driver so that I can access the
> on-board graphics from within a Centos-x86_64 container. I've added the PCI
> slot that the GPU comes up on using the '--pci_add' option but the driver
> still reads out that no
Hi,
I'm looking to install AMD 13.25 APU Driver so that I can access the
on-board graphics from within a Centos-x86_64 container. I've added the PCI
slot that the GPU comes up on using the '--pci_add' option but the driver
still reads out that no compatible adapters were found. If I run 'lspci' in
The best course of action is to file a bug in bugzilla.openvz.org.
On 07/31/2014 07:26 AM, ml-openvz-e...@kuszelas.eu wrote:
> Does anyone else encounter this:
>
>
> [ 1091.907646] kobject_add_internal failed for net with -EEXIST, don't try to
> register things with the same name in the same dire
Hi Pavel!
We were using venet and the benchmark is NetPIPE, which measures the RTT
(pingpong) between 2 machines. Most distros have netpipe in their repos.
Available also from here:
http://www.scl.ameslab.gov/netpipe/code/NetPIPE-3.7.1.tar.gz
Measuring your current performance is straightforward:
Hello!
Could you provide performance test mentioned above: "we have been able
to increase significantly the performance among CT, from peak transfer
rates of 1.3 Gbps (originally) up to 26 Gbps"?
What kind of OpenVZ interfaces did you use? venet or veth?
On Fri, Aug 1, 2014 at 11:50 AM, Guillerm
Joseph,
Thanks for your interest!
Under the hood, we always establish the "standard" socket and after that,
we try to open an alternative path which rely on IPC resources. This way,
if both peers have UFS then a lightweight path is activated, sending
traffic in userland. The kernel is unchanged a
