search for: enque

In case we have to migrate a ballon page to a newpage of another zone, the managed page count of both zones is wrong. Paired with memory offlining (which will adjust the managed page count), we can trigger kernel crashes and all kinds of different symptoms. One way to reproduce: 1. Start a QEMU guest with 4GB, no NUMA 2. Hotplug a 1GB DIMM and only the memory to ZONE_NORMAL 3. Inflate the balloon

In case we have to migrate a ballon page to a newpage of another zone, the managed page count of both zones is wrong. Paired with memory offlining (which will adjust the managed page count), we can trigger kernel crashes and all kinds of different symptoms. One way to reproduce: 1. Start a QEMU guest with 4GB, no NUMA 2. Hotplug a 1GB DIMM and only the memory to ZONE_NORMAL 3. Inflate the balloon

Hi, I implemented queues on some ports with htb by filtering. and to have a constant bw, i defined rate and ceil the same. But while watching the bw''s, I ve realized they may sometimes exceed the defined value for short periods. How could I avoid this situation? I thought that implementing a leaky bucket''d be fine but i m not sure about this and how to do it on tc. I would be

In case we have to migrate a ballon page to a newpage of another zone, the managed page count of both zones is wrong. Paired with memory offlining (which will adjust the managed page count), we can trigger kernel crashes and all kinds of different symptoms. One way to reproduce: 1. Start a QEMU guest with 4GB, no NUMA 2. Hotplug a 1GB DIMM and online the memory to ZONE_NORMAL 3. Inflate the

...e algorithm should handle it pretty well. If an packet does not fit the queue it gets dropped and also the whole communication afterwards till the queue gets freed. And as we know the queue is controlled by the Qdisc structure and it''s mechanisms and not the WRR if ((retvalue=qdisc->enqueue(skb, qdisc)) == ENQUEUE_SUCCESS) { // Successfull sch->stats.packets++; ... if(retvalue!=ENQUEUE_SUCCESS) { // Packet not enqued: sch->stats.drops++; It does not look that the queue could "fill up". Can somebody help (the problem is interresting, im thinking b...

On Tue, Dec 10, 2019 at 02:44:38PM +0100, David Hildenbrand wrote: > On 05.12.19 10:24, David Hildenbrand wrote: > > In case we have to migrate a ballon page to a newpage of another zone, the > > managed page count of both zones is wrong. Paired with memory offlining > > (which will adjust the managed page count), we can trigger kernel crashes > > and all kinds of

On 05.12.19 10:24, David Hildenbrand wrote: > In case we have to migrate a ballon page to a newpage of another zone, the > managed page count of both zones is wrong. Paired with memory offlining > (which will adjust the managed page count), we can trigger kernel crashes > and all kinds of different symptoms. > > One way to reproduce: > 1. Start a QEMU guest with 4GB, no NUMA

Dear List, I am wanting to perform some traffic shaping as the subject of this email suggests. What I am wanting to do is this; I would like to have traffic shaping performed on the following protocols: HTTP, RDP, GRE, PPTP, SIP and IAX. Obviously I would like to have highest priority set for voice packets so much so that the general http traffic does not impede on the voice packets. I