flow: Move pre-place buffering from abc to OR floorplan

[openroad-ci]

No description provided.

[povik]

Secure run at secure-floorplan-buffering

[oharboe]

Just a heads up, used in global_route.tcl too. Should be fine near as I can see.

$ grep -r repair_design_helper scripts/*
scripts/floorplan.tcl:  repair_design_helper
scripts/floorplan.tcl:  repair_design_helper
scripts/global_route.tcl:    repair_design_helper
scripts/resize.tcl:repair_design_helper
scripts/util.tcl:proc repair_design_helper {} {

[povik]

Non-metric failures from public CI:

------------------------------
       Failing designs
------------------------------
asap7 mock-array (base)
  Last log file 5_1_grt.log
  Found 6 errors in the logs.
      [ERROR DRT-0073] No access point for ces_3_3/io_ins_down[10] (Element).
      [ERROR DRT-0073] No access point for ces_3_0/io_ins_down[10] (Element).
      [ERROR DRT-0073] No access point for ces_0_0/io_ins_down[10] (Element).
      [ERROR DRT-0073] No access point for ces_1_2/io_ins_down[10] (Element).
      [ERROR DRT-0073] No access point for ces_0_6/io_ins_left[14] (Element).
      [ERROR DRT-0073] No access point for ces_1_7/io_ins_left[14] (Element).

this has the symptoms of The-OpenROAD-Project/OpenROAD#6991

ihp-sg13g2 i2c-gpio-expander (base)
  Last log file 3_3_place_gp.log
  Found 1 errors in the logs.
      [ERROR GRT-0042] Pin io_address_0_PAD does not have geometries in a valid routing layer.

sky130hd ibex (base)
  Last log file 5_2_route.log
  Found 1 errors in the logs.
      [ERROR GRT-0232] Routing congestion too high. Check the congestion heatmap in the GUI.

sky130hd microwatt (base)
  Last log file 5_1_grt.log
  Found 1 errors in the logs.
      [ERROR GRT-0116] Global routing finished with congestion. Check the congestion regions in the DRC Viewer.

[povik]

Curious failure mode from macro placer on gf12/bp_single. @AcKoucher is there anything I should do to triage?

[AcKoucher]

@povik I think this might be related with the approach we're using to deal with PADs in MPL - regarding how we map PADs with the internal objects of MPL. There is a PR that changes the current approach. Could you try using those changes? I just updated it.