Hey, I'm running through the caravel macro tutoria...
# openlane-2c
cqb
03/04/2024, 1:53 AMHey, I'm running through the caravel macro tutorial in the ol2 docs. I'm about half way through and running into this error while attempting to harden the user project wrapper
Copy code
* Antenna
Failed π
Pin violations: 45
Net violations: 45
Check the report directory of OpenROAD.CheckAntennas.
* LVS
Passed β
* DRC
Passed β
[22:11:02] ERROR The following deferred step errors have been encountered: sequential.py:282
[22:11:02] ERROR hold violations found in the following corners: sequential.py:284
max_tt_025C_1v80
[22:11:02] ERROR OpenLane will now quit.v
Vijayan Krishnan
03/04/2024, 5:36 AMCheck antenna violations.
Hold violations in the design.
Check respective reports
d
donn
03/04/2024, 12:02 PM@cqb The tutorial guides you on fixing hold violations after this error: https://openlane2.readthedocs.io/en/latest/usage/caravel/index.html#id18
β¦unless you mean that this happens after following those particular steps, which is a bug.
c
cqb
03/05/2024, 2:58 AMI am as a section higher up in the guide, here: https://openlane2.readthedocs.io/en/latest/usage/caravel/index.html#id2
The tutorial walked me through the errors while hardening the aes macro which was successful, but now I am getting them on the user project wrapper.
I attempted running the flow with the config changes mentioned in the section you linked and I get this:
Copy code
* Antenna
Failed π
Pin violations: 29
Net violations: 29
Check the report directory of OpenROAD.CheckAntennas.
* LVS
Failed π
Total Errors: 438
Unmatched Pins: 52
Unmatched Nets: 4
Check the report directory of Netgen.LVS.
* DRC
Failed π
KLayout DRC errors: 0
Magic DRC errors: 2034
Check the report directories of KLayout.DRC and Magic.DRC.
[20:52:07] ERROR The following deferred step errors have been encountered: sequential.py:282
[20:52:07] ERROR 2034 Magic DRC errors found. sequential.py:284
[20:52:07] ERROR 438 LVS errors found. sequential.py:284
[20:52:07] ERROR OpenLane will now quit.cqb
03/05/2024, 3:30 AMon my first run for uzer_project_wrapper this was my antenna summary:
Copy code
ββββββββββββββββββββ³βββββββββββ³βββββββββββ³ββββββββββββββββ³βββββββββββββββββββββ³ββββββββ
β Partial/Required β Required β Partial β Net β Pin β Layer β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β 10.15 β 2773.88 β 28145.14 β wbs_dat_i[0] β mprj/wbs_dat_i[0] β met3 β
β 10.06 β 2773.88 β 27916.06 β wbs_dat_i[1] β mprj/wbs_dat_i[1] β met3 β
β 9.97 β 2773.88 β 27659.05 β wbs_adr_i[2] β mprj/wbs_adr_i[2] β met3 β
β 9.84 β 2773.88 β 27295.37 β wbs_adr_i[3] β mprj/wbs_adr_i[3] β met3 β
β 8.88 β 2773.88 β 24624.51 β wbs_dat_i[8] β mprj/wbs_dat_i[8] β met3 β
β 8.81 β 2773.88 β 24441.65 β wbs_cyc_i β mprj/wbs_cyc_i β met3 β
β 8.77 β 2773.88 β 24332.44 β wbs_stb_i β mprj/wbs_stb_i β met3 β
β 8.77 β 2773.88 β 24314.92 β wbs_dat_i[9] β mprj/wbs_dat_i[9] β met3 β
β 8.19 β 2773.88 β 22706.79 β wbs_dat_i[3] β mprj/wbs_dat_i[3] β met3 β
β 8.18 β 2773.88 β 22689.27 β wbs_adr_i[4] β mprj/wbs_adr_i[4] β met3 β
β 8.13 β 2773.88 β 22543.24 β wbs_adr_i[5] β mprj/wbs_adr_i[5] β met3 β
β 8.11 β 2773.88 β 22484.83 β wbs_adr_i[6] β mprj/wbs_adr_i[6] β met3 β
β 8.00 β 2773.88 β 22188.70 β wbs_dat_i[17] β mprj/wbs_dat_i[17] β met3 β
β 7.98 β 2773.88 β 22134.35 β wbs_adr_i[7] β mprj/wbs_adr_i[7] β met3 β
β 7.91 β 2773.88 β 21953.27 β wbs_adr_i[8] β mprj/wbs_adr_i[8] β met3 β
β 7.86 β 2773.88 β 21789.71 β wbs_dat_i[16] β mprj/wbs_dat_i[16] β met3 β
β 7.85 β 2773.88 β 21766.35 β wbs_adr_i[9] β mprj/wbs_adr_i[9] β met3 β
β 7.76 β 2773.88 β 21516.95 β wbs_dat_i[10] β mprj/wbs_dat_i[10] β met3 β
β 7.42 β 2773.88 β 20588.19 β wbs_dat_i[25] β mprj/wbs_dat_i[25] β met3 β
β 7.13 β 2773.88 β 19768.63 β wbs_dat_i[15] β mprj/wbs_dat_i[15] β met3 β
β 6.89 β 2773.88 β 19104.51 β wbs_dat_i[4] β mprj/wbs_dat_i[4] β met3 β
β 6.84 β 2773.88 β 18974.22 β wbs_dat_i[12] β mprj/wbs_dat_i[12] β met3 β
β 6.82 β 2773.88 β 18909.97 β wbs_dat_i[19] β mprj/wbs_dat_i[19] β met3 β
β 6.78 β 2773.88 β 18804.83 β wbs_dat_i[26] β mprj/wbs_dat_i[26] β met3 β
β 6.65 β 2773.88 β 18436.83 β wbs_dat_i[18] β mprj/wbs_dat_i[18] β met3 β
β 6.46 β 2773.88 β 17908.19 β wbs_dat_i[2] β mprj/wbs_dat_i[2] β met3 β
β 6.16 β 2773.88 β 17078.73 β wbs_dat_i[5] β mprj/wbs_dat_i[5] β met3 β
β 6.09 β 2773.88 β 16883.05 β wbs_dat_i[24] β mprj/wbs_dat_i[24] β met3 β
β 6.08 β 2773.88 β 16856.76 β wbs_dat_i[6] β mprj/wbs_dat_i[6] β met3 β
β 6.00 β 2773.88 β 16636.57 β wbs_dat_i[7] β mprj/wbs_dat_i[7] β met3 β
β 5.99 β 2773.88 β 16614.35 β wbs_dat_i[23] β mprj/wbs_dat_i[23] β met3 β
β 5.97 β 2773.88 β 16555.94 β wbs_dat_i[20] β mprj/wbs_dat_i[20] β met3 β
β 5.67 β 2773.88 β 15737.02 β wbs_dat_i[11] β mprj/wbs_dat_i[11] β met3 β
β 5.62 β 2773.88 β 15601.34 β wbs_we_i β mprj/wbs_we_i β met3 β
β 5.51 β 2773.88 β 15287.24 β wbs_dat_i[13] β mprj/wbs_dat_i[13] β met3 β
β 5.49 β 2773.88 β 15231.75 β wbs_dat_i[27] β mprj/wbs_dat_i[27] β met3 β
β 5.43 β 2773.88 β 15051.81 β wbs_dat_i[14] β mprj/wbs_dat_i[14] β met3 β
β 5.39 β 2773.88 β 14945.52 β wbs_dat_i[29] β mprj/wbs_dat_i[29] β met3 β
β 5.08 β 2773.88 β 14102.60 β wbs_dat_i[28] β mprj/wbs_dat_i[28] β met3 β
β 4.87 β 2773.88 β 13506.79 β wbs_dat_i[31] β mprj/wbs_dat_i[31] β met3 β
β 4.86 β 2773.88 β 13480.51 β wbs_dat_i[21] β mprj/wbs_dat_i[21] β met3 β
β 4.78 β 2773.88 β 13252.70 β wbs_dat_i[22] β mprj/wbs_dat_i[22] β met3 β
β 4.13 β 2773.88 β 11453.59 β wbs_dat_i[30] β mprj/wbs_dat_i[30] β met3 β
β 2.96 β 2773.88 β 8203.11 β wbs_adr_i[7] β mprj/wbs_adr_i[7] β met2 β
β 2.85 β 2773.88 β 7897.11 β wbs_adr_i[5] β mprj/wbs_adr_i[5] β met2 β
β 2.83 β 2773.88 β 7859.33 β wbs_adr_i[6] β mprj/wbs_adr_i[6] β met2 β
β 2.82 β 2773.88 β 7821.56 β wbs_adr_i[8] β mprj/wbs_adr_i[8] β met2 β
β 2.81 β 2773.88 β 7783.78 β wbs_adr_i[9] β mprj/wbs_adr_i[9] β met2 β
β 2.79 β 2773.88 β 7746.00 β wbs_dat_i[12] β mprj/wbs_dat_i[12] β met2 β
β 2.78 β 2773.88 β 7708.22 β wbs_dat_i[18] β mprj/wbs_dat_i[18] β met2 β
β 2.76 β 2773.88 β 7666.67 β wbs_dat_i[19] β mprj/wbs_dat_i[19] β met2 β
β 2.75 β 2773.88 β 7628.89 β wbs_dat_i[20] β mprj/wbs_dat_i[20] β met2 β
β 2.74 β 2773.88 β 7591.11 β wbs_dat_i[23] β mprj/wbs_dat_i[23] β met2 β
β 2.71 β 2773.88 β 7515.56 β wbs_dat_i[4] β mprj/wbs_dat_i[4] β met2 β
β 2.17 β 2773.88 β 6019.56 β wbs_cyc_i β mprj/wbs_cyc_i β met2 β
β 2.05 β 2773.88 β 5675.78 β wbs_dat_i[24] β mprj/wbs_dat_i[24] β met2 β
β 2.03 β 2773.88 β 5638.00 β wbs_dat_i[28] β mprj/wbs_dat_i[28] β met2 β
β 2.02 β 2773.88 β 5596.44 β wbs_dat_i[31] β mprj/wbs_dat_i[31] β met2 β
β 1.91 β 2773.88 β 5286.66 β wbs_we_i β mprj/wbs_we_i β met2 β
β 1.79 β 2773.88 β 4971.51 β wb_rst_i β mprj/wb_rst_i β met3 β
β 1.27 β 2773.88 β 3526.22 β wbs_adr_i[4] β mprj/wbs_adr_i[4] β met2 β
β 1.04 β 2773.88 β 2885.52 β wb_clk_i β mprj/wb_clk_i β met3 β
ββββββββββββββββββββ΄βββββββββββ΄βββββββββββ΄ββββββββββββββββ΄βββββββββββββββββββββ΄ββββββββcqb
03/05/2024, 3:36 AMand stapostpnr from the first run:
d
donn
03/06/2024, 8:07 AMAre you referring to this section? https://openlane2.readthedocs.io/en/latest/usage/caravel/index.html#user-project-wrapper-hardening
If so, the recommended configuration is here: https://openlane2.readthedocs.io/en/latest/usage/caravel/index.html#id10
c
cqb
03/07/2024, 1:54 AMYes, that's the section, thanks. I've reran open lane successfully with the modifications described. The manufacturability report looks fine, but the
stapostpnr Copy code
ββββββββββββ³βββββββββββ³ββββββββββ³βββββββββββ³ββββββββββ³βββββββββββ³
β β β β β β β
β β Hold β Reg to β β β of which β
β β Worst β Reg β β Hold β Reg to β
β Corner/β¦ β Slack β Paths β Hold TNS β Violatβ¦ β Reg β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Overall β -0.0061 β -0.0061 β -0.0061 β 3 β 3 β
β nom_tt_β¦ β 0.1391 β 0.1391 β 0.0000 β 0 β 0 β
β nom_ss_β¦ β 0.5329 β 0.5329 β 0.0000 β 0 β 0 β
β nom_ff_β¦ β -0.0045 β -0.0045 β -0.0045 β 1 β 1 β
β min_tt_β¦ β 0.1367 β 0.1367 β 0.0000 β 0 β 0 β
β min_ss_β¦ β 0.5299 β 0.5299 β 0.0000 β 0 β 0 β
β min_ff_β¦ β -0.0061 β -0.0061 β -0.0061 β 1 β 1 β
β max_tt_β¦ β 0.1414 β 0.1414 β 0.0000 β 0 β 0 β
β max_ss_β¦ β 0.5368 β 0.5368 β 0.0000 β 0 β 0 β
β max_ff_β¦ β -0.0029 β -0.0029 β -0.0029 β 1 β 1 β
ββββββββββββ΄βββββββββββ΄ββββββββββ΄βββββββββββ΄ββββββββββ΄βββββββββββ΄d
donn
03/07/2024, 10:17 AM@cqb It's normal for there to be some variance in those reports- even small things like placing a cell a few microns away can have a new violations and the PnR tools are highly heuristic.
I'll try my best to explain this table, but contributions/corrections from others are welcome.
To know a little bit more of what you're supposed to be looking for in these reports, this guide may be helpful: https://openlane2.readthedocs.io/en/latest/usage/corners_and_sta.html
tl;dr, you have 1 "hold violation" in 3 "corners" (definitions in the document.) The corners appear to be:
β’ nom_ff_n40C_1v95
β’ min_ff_n40C_1v95
β’ max_ff_n40C_1v95
You'll notice "ff" is common to all three. FF indicates here that you "won the silicon lottery" if you ever heard this term- your transistors are very fast. Unfortunately, there is such a thing as too fast, and that means the data changes too quickly at times- causing the hold violation. For most purposes, hold violations indicate a dead chip at those corners.
Solving hold violations is not entirely a hard science- you can mess with some variables in OpenLane to maybe achieve timing closure, chiefly in the steps running what's known as "resizing". Here's one such step: https://openlane2.readthedocs.io/en/latest/reference/step_config_vars.html#resizer-timing-optimizations-post-global-routing
You'll notice there's a variable called
GRT_RESIZER_HOLD_SLACK_MARGIN0.050.1c
cqb
03/08/2024, 8:57 PM@donn Thanks for all your help! i still haven't figured out how to get rid of the hold violations but I'll spend some time going through the docs to see if I can tinker with some config variables. Really appreciate your guidance!
π 1
26 Views