Introduce verilog -> cocotb simulation fud2 path
#1997
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Sketch an example fud2 op * Fix typo --------- Co-authored-by: Adrian Sampson <adrian@radbox.org>
Testing locally on vec add, the hard-coded cocotb tests pass, meaning there is parity with the previous bash scripts
Makefile expects a DATA_PATH and VERILOG_SOURCE
cocotb expects datapath and relatviepath to be relative to the actual Makefile, not from the invocation
nathanielnrn
force-pushed
the
yxi-cocotb
branch
from
80c5644
to
5ba23e1
Compare
nathanielnrn
force-pushed
the
yxi-cocotb
branch
from
078a015
to
f2ba9c3
Compare
This reverts commit 09ac9c2.
|
I think this is finally ready for review if someone gets the chance. I know #2084 has some implications on |
|
Yeah definitely merge this first. We're not quite ready to migrate things yet. |
sampsyo
approved these changes
Jun 5, 2024
| @@ -174,7 +174,7 @@ jobs: | |||
| uses: actions-rs/cargo@v1 | |||
| with: | |||
| command: build | |||
| args: --all-features --manifest-path /home/calyx/Cargo.toml | |||
| args: --all --all-features --manifest-path /home/calyx/Cargo.toml | |||
There was a problem hiding this comment.
Two tiny things: cargo build --help suggests that --all is deprecated in favor of --workspace, and just wanted to make sure that we actually do want to build everything in the workspace.
| //Get yxi file from main compute program. | ||
| //TODO(nate): Can this use the `yxi` operation instead of hardcoding the build cmd calyx rule with arguments? | ||
| // Get yxi file from main compute program. | ||
| // TODO(nate): Can this use the `yxi` operation instead of hardcoding the build cmd calyx rule with arguments? |
There was a problem hiding this comment.
This will probably have to wait for #1958.
| // Cocotb is wants files relative to the location of the makefile. | ||
| // This is annoying to calculate on the fly, so we just copy necessary files to the build directory | ||
| e.rule("copy", "cp $in $out")?; | ||
| e.rule("make", "make DATA_PATH=$sim_data VERILOG_SOURCE=$in COCOTB_LOG_LEVEL=CRITICAL > $out")?; |
There was a problem hiding this comment.
Maybe this rule should be called make-cocotb or something (instead of the more generic make)?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Introduces a path accepting
.vfiles, presumably axi-wrapped via afud2 foo.futil --through axi-wrappedand outputting the final contents of AXI rams simulated viacocotb.Invocation currently has to be manually specified with
--throughdue to the behavior ofguess_stateAn invocation looks like
fud2 <path to axi wrapped verilog> --from verilog-noverify --to cocotb-axi --set sim.data=<path to .data/json file>This PR also introduces some
cocotbpython runners that largely looks like the xilinx cocotb harness. Those will hopefully be removed once we get the wrapper to properly interface with XRT.Adds some
runttests as well that check the state of some AXI generation intermediate steps (including simulation output with cocotb). I think that the main changes of note besides that is thiscocotbpython runner mentioned above and thefud2changes