Routability iteration endless loops?

[oharboe]

What should I expect w.r.t. global routing and routability iterations for futile cases?

Is there something in the logs below that is telling me what to try w.r.t. tweaking the parameters?

The target below used to complete global placement in a reasonable amount of time, 1 hour(?), but now it seems to have become futile without some tweaking of parameters.

There's lots of (debug?) information, but I'm trying to find information that I as a user can use to fix the problem...

     3370 |   0.3821 |  1.945553e+10 |   -1.47% |  2.63e-07 |      
     3380 |   0.3509 |  1.930033e+10 |   -0.80% |  3.87e-07 |      
     3390 |   0.3326 |  1.905667e+10 |   -1.26% |  5.71e-07 |      
     3400 |   0.3159 |  1.876196e+10 |   -1.55% |  8.40e-07 |      
[INFO GPL-0040] Routability iteration: 29
[INFO GPL-0041] Total routing overflow: 1019.4382
[INFO GPL-0042] Number of overflowed tiles: 26614 (1.02%)
[INFO GPL-0043] Average top 0.5% routing congestion: 1.0636
[INFO GPL-0044] Average top 1.0% routing congestion: 1.0392
[INFO GPL-0045] Average top 2.0% routing congestion: 1.0101
[INFO GPL-0046] Average top 5.0% routing congestion: 0.9591
[INFO GPL-0047] Routability iteration weighted routing congestion: 1.0514
[INFO GPL-0049] Routing congestion (1.0514) higher than minimum (1.0509). Consecutive non-improvement count: 3.
[INFO GPL-0051] Inflated area:                 343.920 um^2 (+0.23%)
[INFO GPL-0052] Placement target density:       0.2590
[INFO GPL-0076] Removing fillers, count: Before: 256296, After: 256296 (+0.00%)
[INFO GPL-0077] Filler area (um^2)     : Before: 33146.762, After: 33146.762 (+0.00%)
[INFO GPL-0078] Removed fillers count: 0, area removed: 0.000 um^2. Remaining area to be compensated by modifying density: 343.920 um^2
[INFO GPL-0079] New target density: 0.25948477
[INFO GPL-0054] No improvement in routing congestion for 3 consecutive iterations (limit is 3).
[INFO GPL-0055] Reverting inflation values and target density from the iteration with minimum observed routing congestion.
[INFO GPL-0056] Minimum observed routing congestion: 1.0509
[INFO GPL-0057] Target density at minimum routing congestion: 0.2575
[INFO GPL-0080] Restoring 0 previously removed fillers.
[INFO GPL-0089] Routability finished. Reverting to minimal observed routing congestion, could not reach target.

[maliberty]

You have too much routing congestion so I would start with looking at the congestion heatmap in the result.

[oharboe]

What above in the log made you conclude this?

I kinda figured that out, but I'm curious what I should look for in the logs.

[maliberty]

"No improvement in routing congestion for 3 consecutive iterations (limit is 3)."