Re: [openram-dev-group] Problem of spice simulation characterization

12 views
Skip to first unread message

Matthew Guthaus

unread,
Sep 18, 2019, 2:21:35 PM9/18/19
to openram-...@ucsc.edu
Rongmei,

You can debug this by looking at the debug output. This is described on slides 19-26 of the documentation. There are temporary files with the standard output and standard error. It is likely that there was a problem (a license problem?) running your simulator. This would be in the output.

Matt

On Wed, Sep 18, 2019 at 5:50 AM Rongmei CHEN <mei201...@gmail.com> wrote:
Hi Professor,

  I found an error during the characterization using spice simulation instead of analytical calculation. I setup the configuration file and the path to hspice which I believe should be correct. 

  The error massage is as below. Would you tell me what mistakes I made or how to correct?

ERROR: file stimuli.py: line 358: Spice simulation error: /imec/software/synopsys/release2013/hspice_v2013.03-SP1/hspice/bin/hspice -mt 2 -i /tmp/openram_chen51_19830_temp/stim.sp -o /tmp/openram_chen51_19830_temp/timing
Traceback (most recent call last):
  File "/imec/users/chen51/openram/compiler//openram.py", line 76, in <module>
    s.save()
  File "/imec/users/chen51/openram/compiler/sram.py", line 112, in save
    lib(out_dir=OPTS.output_path, sram=self.s, sp_file=sp_file)
  File "/imec/users/chen51/openram/compiler/characterizer/lib.py", line 34, in __init__
    self.characterize_corners()
  File "/imec/users/chen51/openram/compiler/characterizer/lib.py", line 94, in characterize_corners
    self.characterize()
  File "/imec/users/chen51/openram/compiler/characterizer/lib.py", line 101, in characterize
    self.compute_delay()
  File "/imec/users/chen51/openram/compiler/characterizer/lib.py", line 501, in compute_delay
    char_results = self.d.analyze(probe_address, probe_data, self.slews, self.loads)
  File "/imec/users/chen51/openram/compiler/characterizer/delay.py", line 742, in analyze
    feasible_delays = self.find_feasible_period()
  File "/imec/users/chen51/openram/compiler/characterizer/delay.py", line 442, in find_feasible_period
    feasible_delays[self.read_ports[0]] = self.find_feasible_period_one_port(self.read_ports[0])
  File "/imec/users/chen51/openram/compiler/characterizer/delay.py", line 410, in find_feasible_period_one_port
    (success, results)=self.run_delay_simulation()
  File "/imec/users/chen51/openram/compiler/characterizer/delay.py", line 476, in run_delay_simulation
    self.stim.run_sim()
  File "/imec/users/chen51/openram/compiler/characterizer/stimuli.py", line 358, in run_sim
    debug.error("Spice simulation error: " + cmd, -1)
  File "/imec/users/chen51/openram/compiler/debug.py", line 40, in error
    assert return_value == 0
AssertionError

--
https://openram.soe.ucsc.edu
---
You received this message because you are subscribed to the Google Groups "OpenRAM Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openram-dev-gr...@ucsc.edu.
To view this discussion on the web visit https://groups.google.com/a/ucsc.edu/d/msgid/openram-dev-group/cfb5b17e-899e-4a40-b1de-939288d9c03a%40ucsc.edu.


--
Matthew Guthaus
Associate Dean of Graduate Studies
Professor, Computer Science & Engineering
Baskin School of Engineering
University of California Santa Cruz

Rongmei CHEN

unread,
Sep 18, 2019, 2:53:25 PM9/18/19
to OpenRAM Dev
Hi Professor,

  Indeed. I solved the hspice license problem and the hspice characterization is running. However, it is pretty slow. One case study of 128*16 has taken up to 5 hours but not finished yet.

Best,
Rongmei

To unsubscribe from this group and stop receiving emails from it, send an email to openram-...@ucsc.edu.

Matthew Guthaus

unread,
Sep 18, 2019, 3:30:36 PM9/18/19
to openram-...@ucsc.edu
Yeah, characterizing is slow!


To unsubscribe from this group and stop receiving emails from it, send an email to openram-dev-gr...@ucsc.edu.
To view this discussion on the web visit https://groups.google.com/a/ucsc.edu/d/msgid/openram-dev-group/05c71110-c84f-4cbe-9d97-ae01a53c3f1a%40ucsc.edu.
Reply all
Reply to author
Forward
0 new messages