Re: Re: Re: Re: Re: [bisected regression] e1000e: "Detected Hardware Unit Hang"
From: Eric Dumazet <hidden>
Date: 2015-01-15 18:24:55
From: Eric Dumazet <hidden>
Date: 2015-01-15 18:24:55
On Thu, 2015-01-15 at 18:37 +0100, Thomas Jarosch wrote:
On Thursday, 15. January 2015 09:20:37 Eric Dumazet wrote:quoted
quoted
for the sake of completeness: 1: hangThis is weird : This should have same effect then GRO off (at most one segment per packet)I thought so, too. OTOH the code path was changed from "goto merge" to "return -E2BIG". I didn't look at the code what happens at the "merge" label.
If you leave the "goto merge", then GRO still builds big GRO packets, using a linked list of packets. Then because drivers do not generally support NETIF_F_FRAGLIST, core networking code linearizes such GRO packets before reaching ndo_start_xmit() Check 8a29111c7ca68d928dfab58636f3f6acf0ac04f7 "net: gro: allow to build full sized skb" for details about this.