NFS client READ performance on -current

[Russell L. Carter]

On 07/03/14 17:51, Rick Macklem wrote:

> Well, I took a quick look at the driver and it does use m_defrag(), but
> I think that the "retry:" label it does a goto after doing so might be in
> the wrong place.
> 
> The attached untested patch might fix this.
> 
> Is it convenient to build a kernel with this patch applied and then try
> it with TSO enabled?

Patch applied to both client and server (both nics use if_em),
net.inet.tcp.tso=1

With a cold 5GB transfer, I see a fairly steady mid 60s MB/s reading
on the client.  I'm happy BTW.  The throughput is now sufficient for
my application.

HTH,
Russell

> 
> rick
> ps: It does have the transmit segment limit set to 32. I have no idea if
>     this is a hardware limitation.
>