Spartan 6（SP605）GTP错误计数

GTP向导旨在围绕GTP创建一个非常基本的包装器以包含在设计中，当选择Aurora选项时，它将以与Aurora核心使用它相同的方式配置GTP。
然而，这只是第一步，因为它仍然要求设计人员发送与GTP配置方式兼容的正确数据和信号完整性选项（TX输出摆幅，预加重，RX输入端接，均衡）
）适用于正在使用的系统通道。
CoreGen中的Aurora IP也将包含GTP块，因此不需要使用GTP向导单独创建它。
------您是否尝试在Google中输入问题？
如果没有，你应该在发布之前。太多结果？
尝试添加网站：www.xilinx.com
在原帖中查看解决方案



以下为原文

The GTP Wizard is intended to create a very basic wrapper around the GTP for inclusion in a design and when the Aurora option is select it will configure the GTP in the same way that the Aurora core would use it.   However, this is only the first step as it still requires that the designer send proper data that is compatible with the way that the GTP has been configured and that the signal integrity options (TX output swing, pre-emphasis, RX input termination, equalization) are appropriate for the system channel that is being used.
 
The Aurora IP in CoreGen will include the GTP block as well so there is no requirement to create this separately with the GTP wizard.
------Have you tried typing your question into Google?  If not you should before posting.
Too many results?  Try adding site:www.xilinx.comView solution in original post

我将Aurora运行的速度降低到不到1 Gbps。
现在我可以更好地看到每次检测到逗号时，似乎都会增加错误计数。
为什么会这样？
任何帮助表示赞赏。



以下为原文

I decreased the speed that Aurora runs at to less then 1 Gbps.
 
Now I can see better that each time a comma detected, it seems to increase the error count..
 
Why would this happen?
 
Any help is appreciated. 

您应该只启用一次逗号对齐，然后在检测到它后将其禁用。
在您之前的线程中，您似乎使用K28.5字符进行逗号对齐。
您是否也将其用于时钟校正序列？
------您是否尝试在Google中输入问题？
如果没有，你应该在发布之前。太多结果？
尝试添加网站：www.xilinx.com



以下为原文

You should only enable the comma alignment once and then disable it after it has been detected.   In your prior thread it looks like you are using the K28.5 character for the comma alignment.  Are you also using this for a clock correction sequence?
------Have you tried typing your question into Google?  If not you should before posting.
Too many results?  Try adding site:www.xilinx.com

谢谢你的回答mcgett;
我试过你推荐的。
在示例设计中，RXCOMMADETUSE与vcc绑定。
我删除了它，我更改了一个语句，该语句将1分配给RXCOMMADET，如果RTBYTEISALIGNED为0，则为0分配RXCOMMADETUSE以禁用它。
这让我得到了很多差异错误等。我做错了吗？
你是对的，我使用K28.5字符。
但是Seq.1 Byte 1和Byte 2的时钟校正序列是11110111.这是一个问题吗？
问候，



以下为原文

Thanks for your answer mcgett; 
 
 I tried what you recommended. 
 
 In the example design, RXCOMMADETUSE was tied to vcc. I removed it and I changed it with a statement which assigns 1 to RXCOMMADETUSE if the RXBYTEISALIGNED is 0 else it 0 is assigned RXCOMMADETUSE to disable it.
 
This created me to get a lot of disparity errors etc. Am I doing something wrong?
 
You are right that I use K28.5 character. But the clock correction sequence is 11110111 for both Seq.1 Byte 1 and Byte 2. Is this a problem?
 
Regards,