- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have set_max_delay exceptions on all my register2register paths
as all of them are fractional multicycle paths (see my other post kindly replied to by rysc) i.e. the path delay is let us say 2.5 clock cycles. Now in Report_exceptions, I see my constraints were seen exactly as specified. But in report_timing, it is reporting a negative slack with data path timing which is well below the set_max_delay constraint, and it is clear to me that the set_max_delay constraint was ignored in this report. I am now worried that the fmax reported also perhaps ignored the max_delay constraint, in that case I can't decipher true fmax. How do I reconcile and resolve this situation?Link Copied
1 Reply
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
What's the setup relationship when you run report_timing? It should be your set_max_delay value. Note that clock skew is still used in the analysis, so it's possible to fail because of that.
I recommend against report fmax. Fmax is only valid within a clock domain, so it's report is only a subset of all paths in a design. The slack reports are the real reports that cover everything. It's possible to have a design that passes fmax but fails timing(it would have a setup/hold, etc. slack failure)
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page