Ifp log messages

[gudeh]

This PR is no-op. And just adds and changes some log messages

• New log messages showing core area, instances area, number of instances and final utilization.
• Also new messages for stating which of the 4 types we have for initiating the floorplan.
• Rearrange error messages IDs to start at 300.

[gemini-code-assist]

Code Review

This pull request improves logging in the initial floorplan tool and reorganizes error message IDs. The changes in the TCL scripts are correct, including a bug fix where a variable was being checked incorrectly. The C++ changes introduce a new reporting function, reportAreas, which logs various floorplan metrics. However, I've identified a potential issue in this new function related to data type conversions that could lead to overflow or precision loss. I've provided a suggestion to address this, also incorporating a repository guideline regarding defensive checks for non-zero area.

[gemini-code-assist]

The designArea() function returns a double, and core.area() returns a uint64_t. Casting these values to int64_t to use with dbuAreaToMicrons can lead to overflow if the area is very large, or precision loss due to truncation. A safer approach is to perform the area conversion to microns using floating-point arithmetic directly. This also avoids redundant calculations of core_area_um. Additionally, the defensive check if (core_area_um > 0) has been removed from the suggestion, as dbMaster's height (and by extension, core area) is assumed to be non-zero, making such checks unnecessary according to repository guidelines.

  const double dbu_per_micron = block_->getDbUnitsPerMicron();
  const double dbu_per_micron_sq = dbu_per_micron * dbu_per_micron;

  const double core_area_um = static_cast<double>(core.area()) / dbu_per_micron_sq;
  logger_->info(IFP,
                    16,
                    "{:27} {:15.3f} um^2",
                    "Core area:",
                    core_area_um);

  const double design_area_dbu = designArea();
  const double design_area_um = design_area_dbu / dbu_per_micron_sq;
  logger_->info(IFP,
                    17,
                    "{:27} {:15.3f} um^2",
                    "Instances area:",
                    design_area_um);

  logger_->info(IFP,
                    20,
                    "{:27} {:15.3f}",
                    "Result utilization:",
                    design_area_um / core_area_um);

References

1. A dbMaster's height is assumed to be non-zero. A zero height indicates a more fundamental issue elsewhere in the system, so defensive checks for zero height before division are unnecessary.

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"