Database based timing variation analysis
A method for timing analysis of a circuit design includes, for each group of one or more instances of a cell of a cell library in the circuit design, determining timing related data for the group according to circuit context of the group in the design. The context includes at least one of a path depth, an output load, and an input slew rate. The determined timing related data are applied to analyze the circuit design.
1. A computer-implemented method for analyzing sensitivity of a cell library to process variation comprising:
accepting a first database comprising timing related data for a plurality of cells selected independently of a specific circuit design;
for each cell of the plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations predetermined independently of a specific circuit design, using a computer to determine one or more cell delay variance values including applying a numerical analysis procedure by propagating sensitivity of signal level with respect to device parameters in a time-based simulation; and
storing a second database comprising data associating each cell of the plurality of cells with the cell delay variance values for each input slew rate and output load combination of the predetermined set of input slew rate and output load combinations.
2. The method of claim 1 further computing multiple depth-based timing data for each cell of the plurality of cells according to a combination of single-depth timing data.
3. The method of claim 2 wherein computing the multiple depth-based data for a cell includes accumulating distribution of delay data from stage to stage in a series of the instances of the cell.
4. The method of claim 3 wherein computing the multiple depth-based data for a cell includes using the determined sensitivity with respect to a device parameter for a cell to predict sensitivity to the device parameter for a series arrangement of the cell, and determining the depth-based timing data using the predicted sensitivity.
5. The method of claim 1 further comprising graphically representing characteristics dependent on the one or more cell delay variance values as a two dimensional or three dimensional graph for each of a plurality of cells.
6. The method of claim 1 wherein accepting the first database of timing related data comprises accepting a database that includes data associated with a plurality of combinations of a cell, an input slew rate, and an output load.
7. The method of claim 1 further comprising forming the second database according to a numerical analysis procedure by evaluating an effect of variation of local parameters of a cell from a selection of extreme values of global parameters of the cell.
8. The method of claim 1 wherein the numerical analysis procedure comprises determining by sampling values of the local parameters a distribution over delay for a plurality of load and slew conditions.
9. The method of claim 1 wherein forming the second database further comprises computing multiple depth-based timing data according to a combination of single-depth timing data.
10. The method of claim 1 further comprising for each cell of a plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations, determining a cell delay sensitivity value for at least one of a temperature factor, a voltage factor, and a power related factor.
11. A computer-implemented method for analyzing sensitivity of a cell library to process variation comprising:
accepting a first database comprising timing related data for a plurality of cells selected independently of a specific circuit design;
for each cell of the plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations predetermined independently of a specific circuit design, using a computer to determine one or more cell delay variance values; and
determining data, including determining timing derate values from the cell delay variance values, suitable for use by a separate timing tool for analysis of a specific circuit design comprising cells of the plurality of cells of the first database, and storing a second database comprising data associating each cell of the plurality of cells with a derate value for each input slew rate and output load combination of the predetermined set of input slew rate and output load combinations.
12. The method of claim 11 wherein determining the data includes using the specific circuit design.
13. The method of claim 11 wherein determining the data is performed independently of the specific circuit design.
14. The method of claim 11 further comprising applying the timing tool to analyze the specific circuit design.
15. The method of claim 11 further computing multiple depth-based timing data for each cell of the plurality of cells according to a combination of single-depth timing data.
16. The method of claim 15 wherein computing the multiple depth-based data for a cell includes accumulating distribution of delay data from stage to stage in a series of the instances of the cell.
17. The method of claim 15 wherein computing the multiple depth-based data for a cell includes using the determined sensitivity with respect to a device parameter for a cell to predict sensitivity to the device parameter for a series arrangement of the cell, and determining the depth-based timing data using the predicted sensitivity.
18. The method of claim 11 further comprising graphically representing characteristics dependent on the one or more cell delay variance values as a two dimensional or three dimensional graph for each of a plurality of cells.
19. The method of claim 11 wherein accepting the first database of timing related data comprises accepting a database that includes data associated with a plurality of combinations of a cell, an input slew rate, and an output load.
20. The method of claim 11 further comprising forming the second database according to a numerical analysis procedure by evaluating an effect of variation of local parameters of a cell from a selection of extreme values of global parameters of the cell.
21. The method of claim 11 wherein the numerical analysis procedure comprises determining by sampling values of the local parameters a distribution over delay for a plurality of load and slew conditions.
22. The method of claim 11 wherein forming the second database further comprises computing multiple depth-based timing data according to a combination of single-depth timing data.
23. The method of claim 11 further comprising for each cell of a plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations, determining a cell delay sensitivity value for at least one of a temperature factor, a voltage factor, and a power related factor.
24. The software of claim 11 further comprising for each cell of a plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations, determining a cell delay sensitivity value for at least one of a temperature factor, a voltage factor, and a power related factor.
25. Software comprising instructions stored on a non-transitory machine-readable medium that executed on a data processing system cause said system to analyze sensitivity of a cell library to process variation, the analyzing comprising:
accessing a first database comprising timing related data for a plurality of cells selected independently of a specific circuit design;
for each cell of the plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations predetermined independently of a specific circuit design, using a computer to determine one or more cell delay variance values including applying a numerical analysis procedure by propagating sensitivity of signal level with respect to device parameters in a time-based simulation; and
storing a second database comprising data associating each cell of the plurality of cells with the cell delay variance values for each input slew rate and output load combination of the predetermined set of input slew rate and output load combinations.
26. The software of claim 25 further computing multiple depth-based timing data for each cell of the plurality of cells according to a combination of single-depth timing data.
27. The software of claim 26 wherein computing the multiple depth-based data for a cell includes accumulating distribution of delay data from stage to stage in a series of the instances of the cell.
28. The software of claim 27 wherein computing the multiple depth-based data for a cell includes using the determined sensitivity with respect to a device parameter for a cell to predict sensitivity to the device parameter for a series arrangement of the cell, and determining the depth-based timing data using the predicted sensitivity.
29. The software of claim 25 further comprising graphically representing characteristics dependent on the one or more cell delay variance values as a two dimensional or three dimensional graph for each of a plurality of cells.
30. The software of claim 25 wherein accessing the first database of timing related data comprises accessing a database that includes data associated with a plurality of combinations of a cell, an input slew rate, and an output load.
31. The software of claim 25 further comprising forming the second database according to a numerical analysis procedure by evaluating an effect of variation of local parameters of a cell from a selection of extreme values of global parameters of the cell.
32. The software of claim 25 wherein the numerical analysis procedure comprises determining by sampling values of the local parameters a distribution over delay for a plurality of load and slew conditions.
33. The software of claim 25 wherein forming the second database further comprises computing multiple depth-based timing data according to a combination of single-depth timing data.
34. The software of claim 25 further comprising for each cell of a plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations, determining a cell delay sensitivity value for at least one of a temperature factor, a voltage factor, and a power related factor.
35. Software comprising instructions stored on a non-transitory machine-readable medium that executed on a data processing system cause said system to analyze sensitivity of a cell library to process variation, the analyzing comprising:
accessing a first database comprising timing related data for a plurality of cells selected independently of a specific circuit design;
for each cell of the plurality of cells in the first database, and for each combination of a predetermined set of input slew rate and output load combinations predetermined independently of a specific circuit design, using a computer to determine one or more cell delay variance values; and
determining data, including determining timing derate values from the cell delay variance values, suitable for use by a separate timing tool for analysis of a specific circuit design comprising cells of the plurality of cells of the first database, and storing a second database comprising data associating each cell of the plurality of cells with a derate value for each input slew rate and output load combination of the predetermined set of input slew rate and output load combinations.
36. The software of claim 35 wherein determining the data includes using the specific circuit design.
37. The software of claim 35 wherein determining the data is performed independently of the specific circuit design.
38. The software of claim 35 further comprising applying the timing tool to analyze the specific circuit design.
39. The software of claim 35 further computing multiple depth-based timing data for each cell of the plurality of cells according to a combination of single-depth timing data.
40. The software of claim 39 wherein computing the multiple depth-based data for a cell includes accumulating distribution of delay data from stage to stage in a series of the instances of the cell.
41. The software of claim 39 wherein computing the multiple depth-based data for a cell includes using the determined sensitivity with respect to a device parameter for a cell to predict sensitivity to the device parameter for a series arrangement of the cell, and determining the depth-based timing data using the predicted sensitivity.
42. The software of claim 35 further comprising graphically representing characteristics dependent on the one or more cell delay variance values as a two dimensional or three dimensional graph for each of a plurality of cells.
43. The software of claim 35 wherein accessing the first database of timing related data comprises accessing a database that includes data associated with a plurality of combinations of a cell, an input slew rate, and an output load.
44. The software of claim 35 further comprising forming the second database according to a numerical analysis procedure by evaluating an effect of variation of local parameters of a cell from a selection of extreme values of global parameters of the cell.
45. The software of claim 35 wherein the numerical analysis procedure comprises determining by sampling values of the local parameters a distribution over delay for a plurality of load and slew conditions.
46. The software of claim 35 wherein forming the second database further comprises computing multiple depth-based timing data according to a combination of single-depth timing data.