Wednesday, 2019-05-08

*** tpb has joined #vtr-dev00:00
tpbTitle: Comparing master...c_internal · verilog-to-routing/vtr-verilog-to-routing · GitHub (at
acomodikem_: a question about `t_type_descriptor`: shouldn't the indices be `StrongId`instead of plain `int`?
tpbTitle: vtr-verilog-to-routing/physical_types.h at f6d5c7e2857f0aaa412a0d8165a41312c17ebcb6 · verilog-to-routing/vtr-verilog-to-routing · GitHub (at
kem_acomodi: VPR has historically used plain int as indicies/handles14:50
kem_acomodi: We've been slowly transitioning various data structures to use StrongId's instead14:50
kem_acomodi: Which give better error checking (compilation errors if you pass the wrong type of ID, instead of silent acceptance)14:52
kem_acomodi: However no one's transitioned the type descriptors to use StrongIds yet14:52
acomodikem_: all right, I am about to open a PR with an initial working change to address the equivalent tiles feature. Probably, alongside with this change, if it does not require too much effort, I could also start transitioning to the usage of StrongId also for type descriptors15:09
acomodikem_, litghost, mithro: the Equivalent Tile PR is now available here:
tpbTitle: WIP [DNM]: Equivalent Tiles placement by acomodi · Pull Request #559 · verilog-to-routing/vtr-verilog-to-routing · GitHub (at
litghostacomodi: Can you look into ?15:44
tpbTitle: [WIP]: Update VTR with upstream patches. by litghost · Pull Request #700 · SymbiFlow/symbiflow-arch-defs · GitHub (at
acomodilitghost: sure, kem_ any ideas on what are the possible causes of an infinite routing? overused RR nodes get down to 1-10 and VPR cannot get to a feasible solution.15:47
acomodikem_: ^15:48
litghostacomodi: I found that the high fanout stuff caused it on xc7, and disabled it15:48
tpbTitle: symbiflow-arch-defs/devices.cmake at master · SymbiFlow/symbiflow-arch-defs · GitHub (at
acomodilitghost: ok, I'll try to disable that as well for ice40 and run it locally to see if that solves it15:49
litghostacomodi: However it is disabled on all arches for us15:49
litghostacomodi: I used a git bisect last time this happened15:49
litghostacomodi: Probably the right solution here too15:49
kem_acomodi: When you hit congestion which doesn't resolve that can be tough to debug15:50
kem_acomodi: Its not always obvious whether it is an algorithm issue or more fundamental to the design15:51
litghostkem_: The only thing that changed was VTR15:52
kem_acomodi: Some optimizations like high fanout routing do restrict the router search space to save CPU time, so it is possible it could cause issues as litghost mentioned.15:52
litghostkem_: The input eblif and arch.xml and rrgraph remained the same15:52
kem_litghost: What was changed in VPR then?15:53
litghostkem_: We merged upstream into our fork15:53
litghostkem_: So "a lot"15:53
kem_litghost: OK15:53
tpbTitle: Merge upstream by acomodi · Pull Request #49 · SymbiFlow/vtr-verilog-to-routing · GitHub (at
kem_litghost: Thats a tough one to debug15:54
litghostYa, this is a case where bisect is the right answer15:54
litghostThat's what I did last time we hit this15:54
acomodikem_, litghost: I will go on and use git bisect and try to get where the issue is15:54
kem_litghost: Makes sense15:55
litghostacomodi: Once we've identified the issue, it is probably worth reducing to a test case and pushing upstream15:55
kem_litghost: I guess my point is that the thing that fails in the bisect may or may not be the fundamental issue15:55
litghostkem_: Of course.  Step 1 is identifying proximate cause, what change induced the symptom15:56
kem_litghost: Yep agreed15:56

Generated by 2.13.1 by Marius Gedminas - find it at!