Open oharboe opened 2 months ago
Yosys 0.43 (git sha1 ead4718e5, g++ 13.2.0-23ubuntu4 -fPIC -Os)
Linux
I realize this is not the shortest possible test-case for this problem, but it is a standalone test-case with everything needed to reproduce it. Thanks to @hovind for figuring this out. This was discovered while working with OpenROAD on ASAP7.
To reproduce:
untar https://drive.google.com/file/d/1x1WbP7VeAsBBjP4zuXUaUmnka_AikNCp/view?usp=sharing
Then run:
./run-me-adder-asap7-base.sh
Output:
26. Executing Verilog backend. Dumping module `\adder'. End of script. Logfile hash: f8c4509b61, CPU: user 1.16s system 0.05s, MEM: 132.97 MB peak Yosys 0.43 (git sha1 ead4718e5, g++ 13.2.0-23ubuntu4 -fPIC -Os) Time spent: 24% 6x read_liberty (0 sec), 23% 2x stat (0 sec), ... + eqy -d ./logs/asap7/adder/base/4_eqy_output --force ./objects/asap7/adder/base/1_2_test.eqy + tee ./logs/asap7/adder/base/1_2_test.txt EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] read_gold: starting process "yosys -ql ./logs/asap7/adder/base/4_eqy_output/gold.log ./logs/asap7/adder/base/4_eqy_output/gold.ys" EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] read_gold: finished (returncode=0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] read_gate: starting process "yosys -ql ./logs/asap7/adder/base/4_eqy_output/gate.log ./logs/asap7/adder/base/4_eqy_output/gate.ys" EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] read_gate: finished (returncode=0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] combine: starting process "yosys -ql ./logs/asap7/adder/base/4_eqy_output/combine.log ./logs/asap7/adder/base/4_eqy_output/combine.ys" EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] combine: finished (returncode=0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] partition: starting process "cd ./logs/asap7/adder/base/4_eqy_output; yosys -ql partition.log partition.ys" EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] partition: finished (returncode=0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: starting process "make -C ./logs/asap7/adder/base/4_eqy_output -f strategies.mk" EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: make: Entering directory '/home/oyvind/OpenROAD-flow-scripts/flow/foo/logs/asap7/adder/base/4_eqy_output' EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: Running strategy 'basic' on 'adder.io_dst'.. EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: Could not prove equivalence of partition 'adder.io_dst' using strategy 'basic' EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: make -f strategies.mk summary EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: make[1]: Entering directory '/home/oyvind/OpenROAD-flow-scripts/flow/foo/logs/asap7/adder/base/4_eqy_output' EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: make[1]: Leaving directory '/home/oyvind/OpenROAD-flow-scripts/flow/foo/logs/asap7/adder/base/4_eqy_output' EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: make: Leaving directory '/home/oyvind/OpenROAD-flow-scripts/flow/foo/logs/asap7/adder/base/4_eqy_output' EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] run: finished (returncode=0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] Warning: Failed to prove equivalence for 1/1 partitions: EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] Failed to prove equivalence of partition adder.io_dst EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] summary: Elapsed clock time [H:MM:SS (secs)]: 0:00:00 (0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] summary: Elapsed process time [H:MM:SS (secs)]: 0:00:00 (0) EQY 2:05:25 [./logs/asap7/adder/base/4_eqy_output] DONE (FAIL, rc=2)
no failure in eqy
failure
Is this another manifestation of https://github.com/YosysHQ/yosys/issues/3879 ?
@QuantamHD @maliberty FYI
Version
Yosys 0.43 (git sha1 ead4718e5, g++ 13.2.0-23ubuntu4 -fPIC -Os)
On which OS did this happen?
Linux
Reproduction Steps
I realize this is not the shortest possible test-case for this problem, but it is a standalone test-case with everything needed to reproduce it. Thanks to @hovind for figuring this out. This was discovered while working with OpenROAD on ASAP7.
To reproduce:
untar https://drive.google.com/file/d/1x1WbP7VeAsBBjP4zuXUaUmnka_AikNCp/view?usp=sharing
Then run:
Output:
Expected Behavior
no failure in eqy
Actual Behavior
failure