Hello, I'm following the tutorials in the documentation but it gives these errors while generating PDN:

[ERROR]: during executing openroad script /openlane/scripts/openroad/pdn.tcl

[ERROR]: Log: designs/regfile_2r1w/runs/full_guide/logs/floorplan/8-pdn.log

[ERROR]: Last 10 lines:

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (73.826um, 541.280um), layer: 2.

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (86.823um, 541.280um), layer: 2.

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (5.520um, 535.840um), layer: 2.

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (99.820um, 535.840um), layer: 2.

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (13.680um, 535.840um), layer: 2.

[WARNING PSM-0038] Unconnected PDN node on net VPWR at location (34.836um, 535.840um), layer: 2.

[WARNING PSM-0038] message limit reached, this message will no longer print

[ERROR PSM-0069] Check connectivity failed.

Error: pdn.tcl, 39 PSM-0069

child process exited abnormally

What should I do to solve these errors?

can you share source RTL?

regfile_2r1w.v is my top module, and mem_1r1w.bb.v is a design used as a hard macro at the top.

I believe hardmacros need explicit connections to power. See https://open-source-silicon.slack.com/archives/C031R263TC7/p1662586346453849

I will try it. Do I need to add explicit connections to the top module too?

Is this your hierarchy?

user_project_wrapper
+ regfile_2r1w
  + <http://mem_1r1w.bb|mem_1r1w.bb>

I believe there is a default case when the names match where it’s not necessary, but I’d recommend adding

FP_PDN_MACRO_HOOKS

to any level with hard macros.

I don't have "user_project_wrapper" but the rest is true.

That

config.tcl

file looks like it’s the expanded version found under

<design_name>/runs/<timestamp>

. Do you have a shorter version at

<design_name>/config.tcl

or

<design_name>/config.json

? If it’s

config.tcl

, you could add

set ::env(FP_PDN_MACRO_HOOKS) "\
        lane0 VPWR VGND VPWR VGND, \
        lane1 VPWR VGND VPWR VGND"

and see if that makes a difference. I notice you don’t have a macro placement file. Are you letting the floorplanner do the placement?

I have a config.json and added

"FP_PDN_MACRO_HOOKS" : "lane0 VPWR VGND VPWR VGND, lane1 VPWR VGND VPWR VGND",

to config.json but still gives the same error. Yes, I didn't use a macro placement file. I'm new to OpenLane, so I'm just following the tutorials in the docs.

Could you post your full output log? What tutorial are you using?

@Kareem Farid Do you have any idea about why the tutorial’s not working?

I deleted and reinstalled OpenLane again, now the flow works without errors. However Klayout and OpenROAD GUI does not open. Klayout does not give any errors, just nothing shows up. OpenROAD crashes with this error:

[ERROR]: during executing openroad script /openlane/scripts/openroad/gui.tcl

[ERROR]: Log: ../dev/null

[ERROR]: Last 10 lines:

13# 0x00007F02EE6F23F8 in /lib64/libglib-2.0.so.0

14# g_main_context_iteration in /lib64/libglib-2.0.so.0

15# QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) in /lib64/libQt5Core.so.5

16# QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) in /lib64/libQt5Core.so.5

17# QThread::exec() in /lib64/libQt5Core.so.5

18# 0x00007F02F67F73B5 in /lib64/libQt5DBus.so.5

19# 0x00007F02F2B13E71 in /lib64/libQt5Core.so.5

20# 0x00007F02F6278EA5 in /lib64/libpthread.so.0

21# clone in /lib64/libc.so.6

child killed: SIGABRT

magic and xschem work ok?

How can I run xschem, does it come with the docker container?

I have never been able to run ANY gui in a docker container.

Magic works ok but xschem gives

command not found

error.

Maybe xschem is not included in the docker container.

I searched the folders, Xschem was not there. However, OpenROAD and Klayout GUIs must be executable.