Às 2:52 PM de 4/12/2017, Andrew Lunn escreveu:
>>>> My setup is FPGA based, so it will have lower performance values.
>>>> Iperf results with
>>>>   "Drop Transmit Status" set: ~650Mbps.
>>>>   "Drop Transmit Status" unset: ~450Mbps.
>>>
>>> What percentage of your customers use FPGAs? When i look at the users
>>> of this driver, i see ST, Allwinner, Rockchip, Meson, etc. So silicon,
>>> not FPGA. Does it make sense to do performance measurements on FPGA,
>>> when you say it has lower performance?
>>
>> I don't understand your question. Synopsys is an IP vendor, so all recent IPs
>> are available for prototyping as you can understand and so early development 
>> is
>> done using a FPGA.
> 
> Sure, early development on FPGA makes sense. But my guess is, > 95% of
> the devices running this driver are silicon, not FPGA. The customers
> you sell the IP to want to know that the driver is going to work well
> on their silicon, not your internal FPGA development setup. The kernel
> community want a warm fuzzy feeling that you care about the real
> devices out in the wild using this driver. So if you could say, "I
> tested on a STM DISCO devel board, and performance went up 20%, an
> RK3288 devel board and got 18% performance boost, and a Merrii A80
> Optimus Board showed 22% improvements", we would have a lot better
> feeling about these patches.

Understand your point, but for now our development and testing setup will be
based on the IP Prototyping Kit, consisting of a FPGA + PHY.

Thanks,
Joao

> 
>       Andrew
> 

Reply via email to