1. Implementing Rule Checking Early in the Design Cycle to Reduce Design Iterations and Verification Time Kent Moffat DesignAnalyst Product Manager Mentor Graphics

2. Moffat S204/MAPLD 2004 2 Design Checking: Evaluating RTL code to determine if any violations exist based on a set of coding rules

3. Moffat S204/MAPLD 2004 3 What Problems Do Design Checkers Solve? n Verification consumes 50-70% of FPGA design cycles — Design checking finds and fixes errors earlier in the design process to reduce costly verification and other downstream tool iterations later n Design reuse is essential to meet FPGA design schedule constraints — Design checking helps ensure adherence to HDL coding guidelines essential for efficient reuse

4. Moffat S204/MAPLD 2004 4 Early Checking Design Flow SimulationSynthesisP & R HDL Design Corp Rule Policies Specific Vendor Policies Group Rule Policies Design Checker

5. Moffat S204/MAPLD 2004 5 Encapsulate Knowledge Best Practices Reuse Techniques Known Issues Checking Tool Apply Knowledge Understand Design Issues Detect & Fix Issues Share Knowledge Team Access Issue Reports Design Checking Tenets HDL Checking Tool

6. Moffat S204/MAPLD 2004 6 Today’s Focus Checker Evolution C Verilog VHDL Predictive Analysis Push as much checking as possible to the specification phase Syntax Semantics Style Rules Static Dynamic The beginning ? SystemC PSL

7. Moffat S204/MAPLD 2004 7 Why Focus on Static Checks? Syntax Style Language Static Rules Gate-Level Engine Driven High Low Static Checkers Interfaced Engines Ease of Use Performance # of Users Use Frequency Checks

8. Moffat S204/MAPLD 2004 8 How Fast is Fast Enough? DesignLinesSecRulesChecks/Sec Ethernet9,8239160174,631 Leon uP14,86010160237,769 PicoJava60,75180160121,502 MicroSparc139,08029016076,734   Tests run on a PC: 2 GHz Pentium 4, RAM 1 G, Windows 2000   Used same set of rules in each test including   HDL syntax and semantics   Reuse Methodology Manual ruleset   Checks/Sec = (Lines * Rules) /Sec (Lines include comments)

9. Moffat S204/MAPLD 2004 9 What Types of Checks can be Performed? n n Standard language & syntax checks n n Good coding practices n n Format & readability n n Downstream tool checks n n Portability/reuse checks n n Cross-language compatibility

10. Moffat S204/MAPLD 2004 10 Example Rule Categories n n Allow n n Assignments n n Clocks & Resets n n Complexity n n Conditions n n Configurations n n Declarations n n Directives n n FSM n n Gates n n HDL Syntax & Semantics n n Instances n n Labels n n Naming n n Order n n Partitioning n n Race Conditions n n Ranges n n Registers n n Sensitivity n n Style n n Subprograms n n VITAL

11. Moffat S204/MAPLD 2004 11 Reuse Methodology Manual Examples

12. Moffat S204/MAPLD 2004 12 Value of Early Detection Specification Simulation Synthesis P & R $$$$ Cost = [(T fix + T reiterate ) * Cost Engineer ] + T market Reuse Cost n n Cost of error detection increases further downstream n n Cost is multiplied if code must be edited and re-verified for subsequent designs Problem $ Time = $ …

13. Moffat S204/MAPLD 2004 13 Understanding the Process Rule Set Policy Parameterized Base Rules Define Ruleset Add Rulesets to Policy Results Table Graphics Source Checker n n Text Editor n n Graphical Editors n n Design browser n n Shell level Run from: 1 2 3 4 Reports

14. Moffat S204/MAPLD 2004 14 Building Custom Rulesets n n Create ruleset(s) n n Use search to find a base rule n n Access online help n n Drag & drop base rulesets & rules into your own rulesets n n Change rule parameters n n Create policies that contain rulesets n n Disable rules n n Lock down & share rules with the team

15. Moffat S204/MAPLD 2004 15 Examining Results   High-level summary   Totals rolled up through groups   Expand, collapse, & cross-ref   Group, filter, & sort results   View only what you want   Save the views   See the code fragment & Msg.   Cross reference   View message & rule help   Disable rules

16. Moffat S204/MAPLD 2004 16 Analyze Results within Text Editor n n Code Browser indicates errors & warnings n n Violation report provides navigation n n Code lines highlighted n n Hover help for each violation n n Step through errors n n Show rule/rule help n n Rerun analysis

17. Moffat S204/MAPLD 2004 17 View Reports n n Export Summary report as CSV, TSV, or HTML n n Export Result Table as CSV, TSV, or HTML n n Export rules used in ASCII

18. Moffat S204/MAPLD 2004 18 Assisted Violation Correction n n If a tool knows exactly what the problem is & how to fix it, then …. n n A good percentage of rule violations fit this scenario n n The tool should have modes of correction: “do it”, step through & change, etc.

19. Moffat S204/MAPLD 2004 19 Challenges and Limitations n Mapping desired design rules into checking tool — Not always obvious match — Custom rule creation may be necessary n Tradeoff between run-time tool performance and depth of analysis — Must be fast to be used frequently — Deep analysis takes longer n Some rule violations are not detectable by static analysis techniques

20. Moffat S204/MAPLD 2004 20 Summary n n HDL design checking tools save design time by identifying errors earlier to avoid costly iterations downstream n n Design checkers help ensure that HDL code is reusable from the start n n Key to frequent usage is high runtime performance and interactivity