rahulk29 / sram22

A configurable SRAM generator
BSD 3-Clause "New" or "Revised" License
40 stars 3 forks source link

Fail to generate layout. "assertion failed: nx >= 0 && ny >= 0" #444

Closed lu0de closed 10 months ago

lu0de commented 1 year ago

(base) [luode@localhost sram22_64x32m4w8]$ sram22 -o .


|\ __|\ |\ |\ \ \ / __ \ / \ \ \ _|\ \ |\ \ \ |\ \ \ \__\ \ \/_/|/ ///|/ /| \ _____ \ \ \ \ \ \ \|| \ _|// / /|// / / |__|\ \ \ \ \ \ \ \ \ \ \ \ \ \ / /_/ / /_/ ___\ \ _\ \ _\ _\ _\ \ _|_\____\ |\|||||||||| || |||| |_____|

Reading configuration file...

Configuration file: "/home/luode/temp/sram22_64x32m4w8/sram22.toml" SRAM parameters: Number of words: 256 Data width: 32 Mux ratio: 4 Write size: 8 Control mode: ReplicaV2

Tasks: Done [1/4] Generate plan Done [2/4] Generate netlist ⠁ In Progress [3/4] Generate layout ⠁ Pending [4/4] Generate Verilog Warning: Importing empty port: vpb Warning: Importing empty port: vnb Warning: Importing empty port: vpb Warning: Importing empty port: vpb thread 'main' panicked at /home/luode/.cargo/git/checkouts/substrate-86fbb4e3c8030876/82a793a/substrate/src/layout/routing/auto/mod.rs:108:9: assertion failed: nx >= 0 && ny >= 0 stack backtrace: 0: rust_begin_unwind at /rustc/cc66ad468955717ab92600c770da8c1601a4ff33/library/std/src/panicking.rs:595:5 1: core::panicking::panic_fmt at /rustc/cc66ad468955717ab92600c770da8c1601a4ff33/library/core/src/panicking.rs:67:14 2: core::panicking::panic at /rustc/cc66ad468955717ab92600c770da8c1601a4ff33/library/core/src/panicking.rs:117:5 3: substrate::layout::routing::auto::GreedyRouter::with_config 4: sram22::blocks::control::layout::::route 5: ::layout 6: substrate::data::SubstrateCtx::instantiate_layout 7: sram22::blocks::sram::layout::::layout 8: substrate::data::SubstrateCtx::instantiate_layout 9: ::layout 10: substrate::data::SubstrateCtx::write_layout 11: sram22::plan::execute_plan 12: sram22::cli::run note: Some details are omitted, run with RUST_BACKTRACE=full for a verbose backtrace.

barakhoffer commented 10 months ago

Same here. any solution for this?

rohanku commented 10 months ago

Sorry for the late reply. We aren't able to reproduce this issue locally, it seems like generating a 256x32m4w8 SRAM works fine on the latest version. Could you try running cargo update and cargo clean, then rerunning the installation steps? It is possible that you are using an old version of Substrate since git dependencies are not semantically versioned.

rohanku commented 10 months ago

Actually, after a little bit more thought it likely is an issue of using the standard SKY 130 PDK. We made a couple modifications to the PDK to allow Substrate to read the standard cells correctly, so you will need to clone our version of the PDK instead: https://github.com/ucb-substrate/skywater-pdk.

I've updated the README to indicate this, let me know if it fixes your issue.

barakhoffer commented 10 months ago

Yes, it fixes the problem. thanks!