完善资料让更多小伙伴认识你,还能领取20积分哦, 立即完善>
我正在实施Xilinx CPRI核心v4.1
核心可以在运行时切换线路速率,核心逻辑运行速度更快,以获得更高的线路速率。 理想情况下,NGC约束是针对最快的核心逻辑时钟速率。 用于6Gbps操作的307.2MHz 我没有使用这个速度,只能达到使用245.76MHz时钟的4.9Gbps 我的设计无法在这个更快的时钟上计时: 要求:3.255ns数据路径延迟:3.401ns(逻辑电平= 1)时钟路径偏移:-0.223ns(1.213 - 1.436)源时钟:CPRI_CORE_wrapper / clk_i在0.000ns上升目标时钟:CPRI_CORE_wrapper / clk_i在3.255ns时钟上升 不确定度:0.057ns 如您所见,此设计将传递PERALOD约束为4.069ns。 该约束是在Xilinx IP内核的ngc中引入的,因此我无法弄清楚如何编辑它。 是否可以编辑或覆盖它? 以上来自于谷歌翻译 以下为原文 I am Implementing the Xilinx CPRI core v4.1 The core can switch line rates at run time and the core logic runs faster for higher line rates. Sensibly, the NGC constraint is for the fastest core logic clock rate. 307.2MHz for 6Gbps operation I am not using this speed, and only go up to 4.9Gbps which uses a 245.76MHz clock My design fails timing on this faster clock like so: Requirement: 3.255ns Data Path Delay: 3.401ns (Levels of Logic = 1) Clock Path Skew: -0.223ns (1.213 - 1.436) Source Clock: CPRI_CORE_wrapper/clk_i rising at 0.000ns Destination Clock: CPRI_CORE_wrapper/clk_i rising at 3.255ns Clock Uncertainty: 0.057ns As you can see, this design would pass with a PERIOD constraint of 4.069ns. The constraint is introduced in an ngc of the Xilinx IP core so I can't work out how to edit it. Is it possible to either edit or override it? |
|
相关推荐
4个回答
|
|
嗨,杰森,
你确定NGC内有限制吗? 用户指南不包含此信息,相反提供了应设置的约束以及在具有核心的UCF中传递的约束。 (所以不在网表内) #参考时钟约束 NET“cpri_block_i / gt_and_clocks_i / txoutclk”TNM_NET =“refclkout”; TIMESPEC“TS_refclk”= PERIOD“refclkout”307.2 MHz; #辅助时钟约束 NET“aux_clk”TNM_NET =“aux_clk”; TIMESPEC“TS_aux_clk”= PERIOD“aux_clk”125.00 MHz; #接收时钟 #这是接收时钟分频器cct的输入 NET“recclk”TNM_NET =“recclk”; TIMESPEC“TS_recclk”= PERIOD“recclk_int”307.2 MHz; 我见过其他内核提供的NCF约束。 你的目录中有没有这些? 我认为ISE项目目录中的任何NCF文件都会被该工具自动拾取。 对于调试:分析ngdbuild(翻译)BLD报告文件:验证哪些约束文件被拾取和分析,以及这些文件是否与您的设置和您期望的匹配。 否则,是否可以为4.9Gbps重新生成内核? 或者是否可以使用更新版本的核心? (最新是v8.1) 最好的祝福, 德赖斯 -------------------------------------------------- -------------------------------------------------- ----------------如果提供的信息有用,请将答案标记为“接受为解决方案”。给予您认为有用的帖子并通过点击星标回复导向 在帖子旁边。 以上来自于谷歌翻译 以下为原文 Hi Jason, are you sure that there are constraints inside the NGC? The userguide does not contain this information and on the contrary supplies the constraints that should be set and that are delivered in the UCF with the core. (so not inside the netlist) # Reference clock constraintNET "cpri_block_i/gt_and_clocks_i/txoutclk" TNM_NET = "refclkout";TIMESPEC "TS_refclk" = PERIOD "refclkout" 307.2 MHz;# Auxiliary clock constraintNET “aux_clk” TNM_NET = “aux_clk”;TIMESPEC “TS_aux_clk” = PERIOD “aux_clk” 125.00 MHz;# Receive clock# This is the input to the receive clock divider cctNET “recclk” TNM_NET = “recclk”;TIMESPEC “TS_recclk” = PERIOD “recclk_int” 307.2 MHz; I have seen NCF constraints supplied with other cores. Do you have any of those in your directories? I think any NCF file in your ISE project directory will automatically be picked up by the tool. For debugging: analyze the ngdbuild (translate) BLD report file: verify which constraint files are picked up and analyzed and if these match with your settings and what you expect. Otherwise, is it possible to regenerate the core for 4.9Gbps? or is it possible to use a newer version of the core? (latest is v8.1) Best regards, Dries -------------------------------------------------------------------------------------------------------------------- Please mark the Answer as "Accept as solution" if the information provided is helpful. Give Kudos to a post which you think is helpful and reply oriented by clicking the star next to the post. |
|
|
|
谢谢Dries,
看看失败的约束我认为我犯了一个错误 - 这看起来像是来自MMCM的推断约束,因此我需要为永远不会达到的情况编辑MMCM分隔符... 那有意义吗? 时序约束:TS_CPRI_CORE_wrapper_cpri_block_i_gt_and_clocks_i_tx_clk_gen_i_clkout0 = PERIOD TIMEGRP“CPRI_CORE_wrapper_cpri_block_i_gt_and_clocks_i_tx_clk_gen_i_clkout0”TS_refclk * 1.25 HIGH 50%; 以上来自于谷歌翻译 以下为原文 Thanks Dries, Looking at the constraint that's failing I think I've made an error - This looks like an inferred constraint from an MMCM and therefore I need to edit the MMCM divider for the case that's never reached... does that make sense? Timing constraint: TS_CPRI_CORE_wrapper_cpri_block_i_gt_and_clocks_i_tx_clk_gen_i_clkout0 = PERIOD TIMEGRP "CPRI_CORE_wrapper_cpri_block_i_gt_and_clocks_i_tx_clk_gen_i_clkout0" TS_refclk * 1.25 HIGH 50%; |
|
|
|
嗨,杰森,
是的,这更有意义:) 最好的祝福 德赖斯 -------------------------------------------------- -------------------------------------------------- ----------------如果提供的信息有用,请将答案标记为“接受为解决方案”。给予您认为有用的帖子并通过点击星标回复导向 在帖子旁边。 以上来自于谷歌翻译 以下为原文 Hi Jason, yes, that makes more sense :) Best regards Dries -------------------------------------------------------------------------------------------------------------------- Please mark the Answer as "Accept as solution" if the information provided is helpful. Give Kudos to a post which you think is helpful and reply oriented by clicking the star next to the post. |
|
|
|
简要更新:
驱动用户时钟的MMCM使用动态重新配置,其默认值列在hdl上,工具使用这些来推断最大usrclk速率 - 将默认值编辑为使用最快的配置允许约束降低。 谢谢你的帮助 以上来自于谷歌翻译 以下为原文 Brief update: The MMCM driving the user clock uses dynamic reconfiguration, its default values are listed on the hdl and the tool uses these to inferr the max usrclk rate - editing the defaults to the fastest used configuration allows the constraint to come down. Thanks for your help |
|
|
|
只有小组成员才能发言,加入小组>>
2383 浏览 7 评论
2800 浏览 4 评论
Spartan 3-AN时钟和VHDL让ISE合成时出现错误该怎么办?
2263 浏览 9 评论
3336 浏览 0 评论
如何在RTL或xilinx spartan fpga的约束文件中插入1.56ns延迟缓冲区?
2430 浏览 15 评论
有输入,但是LVDS_25的FPGA内部接收不到数据,为什么?
756浏览 1评论
请问vc707的电源线是如何连接的,我这边可能出现了缺失元件的情况导致无法供电
546浏览 1评论
求一块XILINX开发板KC705,VC707,KC105和KCU1500
368浏览 1评论
1964浏览 0评论
683浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2024-11-23 22:02 , Processed in 1.459060 second(s), Total 84, Slave 67 queries .
Powered by 电子发烧友网
© 2015 bbs.elecfans.com
关注我们的微信
下载发烧友APP
电子发烧友观察
版权所有 © 湖南华秋数字科技有限公司
电子发烧友 (电路图) 湘公网安备 43011202000918 号 电信与信息服务业务经营许可证:合字B2-20210191 工商网监 湘ICP备2023018690号