[ewg] [PATCH] RDMA/nes: Fix for TSO low nic throughput - linux-next-pending

[Nikolova, Tatyana E]

I am testing with linux kernel 3.5 and the return type of the function skb_is_gso is bool in linux 3.5. 
http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux.git&a=search&h=28a33cbc24e4256c143dce96c7d93bf423229f92&st=commit&s=skb_is_gso

> static inline bool skb_is_gso(const struct sk_buff *skb)
> {
>         return skb_shinfo(skb)->gso_size;
> }

> its bool indeed, but does return gso_size

Only the return type of skb_is_gso() is changed, up to linux 3.3 the return type was int

While investigating low L2 throughput with TSO on (for kernel 3.5), my testing shows that skb_is_gso() returns 1 in the NES TSO handling code.
Getting the MSS using skb_shinfo(skb)->gso_size solves the problem.      

>>> (the patch is linux-next-pending)
> I don't see it in net-next, through which exact tree are you merging it?

I meant for Vlad to have my patch ([PATCH] RDMA/nes: Fix for TSO low nic throughput) under OFED-3.5 linux-next-pending, because I will submit the fix upstream.


-----Original Message-----
From: Or Gerlitz [mailto:ogerlitz at mellanox.com] 
Sent: Sunday, September 09, 2012 1:54 AM
To: Nikolova, Tatyana E
Cc: vlad at dev.mellanox.co.il; ewg at lists.openfabrics.org
Subject: Re: [ewg] [PATCH] RDMA/nes: Fix for TSO low nic throughput - linux-next-pending

On 08/09/2012 03:37, Tatyana Nikolova wrote:
> Fix for TSO low nic throughput with linux-3.5
> skb_is_gso() is changed to bool and returns 1 instead of MSS.
> The gso_size from skb_shared_info is now used to pass MSS to hardware.

here's the form of this function from net-next
>
> static inline bool skb_is_gso(const struct sk_buff *skb)
> {
>         return skb_shinfo(skb)->gso_size;
> }

its bool indeed, but does return gso_size

>
> (the patch is linux-next-pending)

I don't see it in net-next, through which exact tree are you merging it?