From 04/10/2019 to 05/09/2019
- 04:50 PM Usage: RE: UNOPTTHREADS even for parallel designs
- Hi Al,
Par factor is defined in internals.pod:
'The available parallelism or "par-factor" of a DAG is the total...
- 12:34 AM Usage: RE: FST dumping 100x slower than VCD
- Git towards 4.014 adds LZ4 compression which is faster, and --trace-fst-thread (#2 above). Later fixes will do #1 and...
- I'm using --threads=4 and seeing UNOPTTHREADS, and the --stats output says "MTask graph, final, parallelism factor 4"...
- 10:36 PM Usage: RE: FST dumping 100x slower than VCD
- Looked at some benchmarks and there's certainly some inefficiencies, maybe 10x, but a 100x slower would be unexpected...
- 04:18 PM Usage: RE: FST dumping 100x slower than VCD
- What is the expected performance delta between VCD and FST tracing? Given sane and consistent compile flags between ...
- 03:27 PM Development: RE: Requiring C++11 compiler
- CentOS 6.10 uses GCC 4.4.7 which was from 2012 and does not support the ratified C++11.
If possible use a newer OS...
- 02:17 PM Development: RE: Requiring C++11 compiler
I'm using CentOS 6.10 and I get this error message
configure: error: the g++ compiler appears to not suppo...
- 02:51 AM Development: RE: Error on indexing the multiple dimensional array (MDA)
- Disabling this branch in V3Unknown.cpp temporarily resolves the problem.
While the comments say that it slows down s...
- 04:01 PM Usage: RE: eval() multiple times on same clock edge
- Perhaps declare a DPI import function. Then create an always sensitivity list (real or fake) and Verilator will then...
- Hi all,
I have a hardware design that needs to make a memory request at most 3 times per cycle. I have generated c...
- 05:58 AM Development: RE: Error on indexing the multiple dimensional array (MDA)
- Also, sometimes the error message is...
Also available in: Atom