[dpdk-dev] Wrong TCP checksum of packets sent by Linux guest (virtIO/vhost)

2014-09-04 Thread Xie, Huawei
> -Original Message- > From: Franck Baudin [mailto:franck.baudin at qosmos.com] > Sent: Wednesday, September 03, 2014 10:13 PM > To: Xie, Huawei; Gray, Mark D; Thomas Monjalon > Cc: dev at dpdk.org; dpdk-ovs at lists.01.org > Subject: Re: [dpdk-dev] Wrong TCP checksum of packets sent by L

[dpdk-dev] [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line

2014-09-04 Thread Yerden Zhumabekov
Hello Bruce, I'm a little bit concerned about performance issues that would arise if these fields would go to the 2nd cache line. For exampe, l2_len and l3_len fields are used by librte_ip_frag to find L3 and L4 headers position inside mbuf data. Thus, these values should be calculated by NIC off

[dpdk-dev] [PATCH] virtio: Fix vring entry number issue

2014-09-04 Thread Ouyang Changchun
Fix one issue in virtio TX: it needs one more vring entry to hold the virtio header when transmitting packets, it is used later to determine whether to free more entries from used vring. Signed-off-by: Changchun Ouyang Reviewed-by: Huawei Xie Tested-by: Jingguo Fu --- lib/librte_pmd_virtio/v

[dpdk-dev] i211 controller with 0 queues per port

2014-09-04 Thread Sergey Mironov
2014-08-26 13:18 GMT+04:00 Sergey Mironov : > Hi. > > I've got a SCB-8973 machine equipped with i212 network controller (6 > ports). My dpdk application (see the code snippet below) returns zeros > for both dev_info.max_tx_queues and dev_info.max_rx_queues > > Have anybody got an experience of addr

[dpdk-dev] i211 controller with 0 queues per port

2014-09-04 Thread Thomas Monjalon
Hi Sergey, 2014-09-04 11:16, Sergey Mironov: > I've attached a patch which adds necessary i211 support, Sorry, some attached file types are forbidden on the list and automatically removed. You should send your patch inline (git send-email is preferred). > but I didn't check it enough to say it i

[dpdk-dev] [PATCH] igb_ethdev.c: complete switches for i211 NC

2014-09-04 Thread Sergey Mironov
Hi! I have got an update for my "i212 problem". First of all, I found that I have made a mistake. My controller is named i211, not i212 :) Next, i211 controller is controlled by the lib_pmd_e1000 driver. Unfortunately, looks like it's support is pure. For example, igb_ethdev.c contains function eth

[dpdk-dev] Wrong TCP checksum of packets sent by Linux guest (virtIO/vhost)

2014-09-04 Thread Franck Baudin
Hi Huawei, On 09/04/14 04:54, Xie, Huawei wrote: > Hi Franck: > I checked your original thread. > root at linux-native:~# tcpdump -vnei eth0 > tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 65535 > bytes > 17:27:09.262926 00:1b:21:b9:9b:2c > 52:54:00:51:51:12, ethertype IPv

[dpdk-dev] [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line

2014-09-04 Thread Bruce Richardson
On Thu, Sep 04, 2014 at 11:08:57AM +0600, Yerden Zhumabekov wrote: > Hello Bruce, > > I'm a little bit concerned about performance issues that would arise if > these fields would go to the 2nd cache line. > > For exampe, l2_len and l3_len fields are used by librte_ip_frag to find > L3 and L4 head

[dpdk-dev] Wrong TCP checksum of packets sent by Linux guest (virtIO/vhost)

2014-09-04 Thread Xie, Huawei
> -Original Message- > From: Franck Baudin [mailto:franck.baudin at qosmos.com] > Sent: Thursday, September 04, 2014 5:24 PM > To: Xie, Huawei; Gray, Mark D; Thomas Monjalon > Cc: dev at dpdk.org; dpdk-ovs at lists.01.org > Subject: Re: [dpdk-dev] Wrong TCP checksum of packets sent by Lin

[dpdk-dev] [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line

2014-09-04 Thread Yerden Zhumabekov
I get your point. I've also read throught the code of various PMDs and have found no indication of setting l2_len/l3_len fields as well. As for testing, we'd be happy to test the patchset but we are now in process of building our testing facilities so we are not ready to provide enough workload fo

[dpdk-dev] [PATCH 11/13] mbuf: move l2_len and l3_len to second cache line

2014-09-04 Thread Bruce Richardson
On Thu, Sep 04, 2014 at 05:00:12PM +0600, Yerden Zhumabekov wrote: > I get your point. I've also read throught the code of various PMDs and > have found no indication of setting l2_len/l3_len fields as well. > > As for testing, we'd be happy to test the patchset but we are now in > process of buil

[dpdk-dev] filter_ctl PMD API idea

2014-09-04 Thread Wu, Jingjing
Hi, all When we develop filters feature in i40e driver for Intel? Ethernet Controller XL710/X710 [Fortville] (For both 10G/40G), we found that there are lots of new filters, there are also some changes on the existing filters, comparing to ixgbe. If we keep the way to add new ops in rte_eth_de

[dpdk-dev] [PATCH v3 4/6] mbuf: remove the rte_pktmbuf structure

2014-09-04 Thread Bruce Richardson
From: Olivier Matz The rte_pktmbuf structure was initially included in the rte_mbuf structure. This was needed when there was 2 types of mbuf (ctrl and packet). As the control mbuf has been removed, we can merge the rte_pktmbuf into the rte_mbuf structure. Advantages of doing this: - the acces

[dpdk-dev] Test-ACL rules file

2014-09-04 Thread Erik Ziegenbalg
Hey everyone, I have a question pertaining to the test-acl app that comes with the dpdk. I noticed that the program requires a rulesf file. Does anyone know the formatting needed for it to parse the rules correctly? I skimmed through the parser code and it seems like it is looking for an '@' to st

[dpdk-dev] TEST-ACL rules file (no html)

2014-09-04 Thread Erik Ziegenbalg
Hey everyone, I have a question pertaining to the test-acl app that comes with the dpdk. I noticed that the program requires a rulesf file. Does anyone know the formatting needed for it to parse the rules correctly? I skimmed through the parser code and it seems like it is looking for an '@' to st