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>
嗨,
我有一个越来越难以在Vivado 2017.4中路由的设计。 我的大多数运行都是在Linux上完成的。 我也尝试过使用Windows机器,它可以产生更好的效果。 我切换到Vivado 2018.2,它在Linux中提供了更好的结果(非常低的TNS)和Windows中的良好时序(没有时序违规)。 我比较了Linux和Windows运行之间的综合日志,发现没有相关的差异。 但是比较实施日志显示,在Linux中,该工具发现设计难以路由(高拥塞),而在Windows中,该工具没有报告拥塞问题,并且能够关闭时间。 如何在Linux中复制Windows结果(我需要在Linux机器上提交构建版本)? 谢谢, 米歇尔 以上来自于谷歌翻译 以下为原文 Hi, I have a design which is increasingly difficult to route in Vivado 2017.4. Most of my runs have been done on Linux. I have also tried with a Windows machine which yielded much better results. I switched to Vivado 2018.2 which provided better results in Linux (very low TNS) and good timing in Windows (no timing violations). I have compared the synthesis logs between the Linux and Windows runs and found no relevant differences. But comparing the implementation logs showed that in Linux the tool found the design difficult to route (high congestion) while in Windows the tool did not report congestion issues and was able to close timing. How can I replicate the Windows results in Linux (I need to submit builds on Linux machines)? Thanks, Michel |
|
相关推荐
6个回答
|
|
好问题
两台机器都有相同数量的内存和相同的处理器吗? 两个vivados都设置相同吗? 以上来自于谷歌翻译 以下为原文 good question do both machines have the same amount of ram and same processor ? are both vivados set up the same ? |
|
|
|
嗨@ michverc.I会尝试比较日志,看看他们分歧的地方,以及有什么不同。
在每个阶段使用report_timing_summary命令来查看时序的变化也是有帮助的。 -------------------------------------------------- -----------------------不要忘记回答,kudo,并接受为解决方案.------------- -------------------------------------------------- ---------- 以上来自于谷歌翻译 以下为原文 Hi @michverc. I would try comparing the logs to see where they diverge, and what is different. It also would be helpful to use the report_timing_summary command at each stage to see where the timing changes as well. ------------------------------------------------------------------------- Don’t forget to reply, kudo, and accept as solution. ------------------------------------------------------------------------- |
|
|
|
我在两个设置上使用相同的xpr文件和源文件。
我比较了synth_1和impl_1日志。 synth_1 runme.log几乎完全相同。 在impl_1运行中,拥塞报告与报告高拥塞的Linux版本不同,而Windows版本则没有。 在此步骤之前,两个日志非常相似。 Linux和Windows设置均为Intel Xeon@2.4GHz,具有16个内核和128 GB RAM 我的设计是使用XCVU9P。 利用率相当低(30%),但拥堵区域较高 以上来自于谷歌翻译 以下为原文 I use the same xpr file and source files on both setups. I have compared both synth_1 and impl_1 logs. The synth_1 runme.log are almost identical. In the impl_1 run the congestion reports are different with the Linux version reporting high congestion while the Windows one does not. Prior to this step both logs are very similar. Both Linux and Windows setups are Intel Xeon @2.4GHz with 16 cores and 128 GB RAM My design is using XCVU9P. Utilization is rather low (30%) but with high congestion regions |
|
|
|
Hi@michverc。有可用的日志吗?
可能存在更多差异,可能有助于提出行动方案。 此外,时间分歧在哪里? -------------------------------------------------- -----------------------不要忘记回答,kudo,并接受为解决方案.------------- -------------------------------------------------- ---------- 以上来自于谷歌翻译 以下为原文 Hi @michverc. Are the logs available? There might be more differences included that could help suggest a course of action. Also, where does the timing diverge? ------------------------------------------------------------------------- Don’t forget to reply, kudo, and accept as solution. ------------------------------------------------------------------------- |
|
|
|
使用相同的输入文件(RTL,IP,约束)和相同版本的Vivado(2017.4或2018.2),我们是否应该期望Linux和Windows之间存在不同的行为?
我在项目模式下运行,其中中间步骤更有限。 以上来自于谷歌翻译 以下为原文 With the same input files (RTL, IPs, constraints) and same version of Vivado (2017.4 or 2018.2) should we expect different behavior between Linux and Windows? I am running in Project mode where intermediate steps are more limited. |
|
|
|
你好@ michverc。
虽然尽一切努力保持不同操作系统的可重复性,但不同的操作系统可能会导致差异。 以下答案记录中提供了更多信息。 https://www.xilinx.com/support/answers/61599.html -------------------------------------------------- -----------------------不要忘记回答,kudo,并接受为解决方案.------------- -------------------------------------------------- ---------- 以上来自于谷歌翻译 以下为原文 Hi @michverc. While every effort is made to maintain repeatability even different operating systems, a different operating system could cause the difference. More information is provided in the below answer record. https://www.xilinx.com/support/answers/61599.html ------------------------------------------------------------------------- Don’t forget to reply, kudo, and accept as solution. ------------------------------------------------------------------------- |
|
|
|
只有小组成员才能发言,加入小组>>
2424 浏览 7 评论
2825 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?
2294 浏览 9 评论
3374 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?
2465 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?
1228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
590浏览 1评论
求一块XILINX开发板KC705,VC707,KC105和KCU1500
452浏览 1评论
2006浏览 0评论
731浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2024-12-25 10:07 , Processed in 1.320340 second(s), Total 88, Slave 72 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