On 06/25/2015 08:22 PM, zhuyj wrote: > sorry. The patch is in the attachment. Please check it. > > Thanks a lot. > Zhu Yanjun
Still no patch. Any attachment is likely being stripped by the mailing list so I would recommend including it inline instead. Also your explanation doesn't make much sense. It isn't as though the layout of the frame should have an impact on the ICV result unless there is a bug in the IPSec protocol implementation that is causing it to pull in invalid memory. I have tested IPSec tunnels over the ixgbe driver using AH and ESP and not seen any issues when sending larger frames. Are you perhaps using a third party protocol implementation? Is it that implementation that requires frames be linear? What we need is reproduction steps. If you have some test that can be run to reproduce this issue then it would be much easier for us to understand it and find if your solution will work for any other cases that might be present. - Alex ------------------------------------------------------------------------------ Monitor 25 network devices or servers for free with OpManager! OpManager is web-based network management software that monitors network devices and physical & virtual servers, alerts via email & sms for fault. Monitor 25 devices for free with no restriction. Download now http://ad.doubleclick.net/ddm/clk/292181274;119417398;o _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired