Hi As you can see in the project I've already implemented DEF and LEF parsers using BISON/FLEX. I mainly use them to read DEF and LEF files in order to load the layouts into the layout editor, but the grammar is there. It's a general data structur which also can be used otherwise. A new start for place and route would be pretty cool. Graywolf is full of old baggage (30 year old, unused appendix code) and ugly coding style in some places. I've cleaned it up as good as I was able to, but it still hase some pretty nasty behaviour. And QRouter has some performance issues when it comes to multi threading, I had to implement the multithreading feature myself and it's still buggy.

Also maybe you wanna take the problem on in a more robust programming language like Rust or Go?

Cheers David

On Saturday, 28 April 2018 12:01:08 PM HKT p4w5 wrote:

Hej everybody!

Sorry for not joining mumble sessions, I feel a little of on communications, but anyway, hope my question finds some positive echo ;)

As I tried to understand project structures of the whole design flow - correct me if I am wrong - each part of the program uses its own file formats or parsers as data structures for macros and layers, from what I skimmed:

• Graywolf uses a YACC parser for some file formats (don't know which)
• in the qflow repo is some form of lefdef parser using YACC, where I am

not sure if it is just for some UI stuff or supposed to be a actual parser

• in qroute 20% of the LOCs are for lefdef parsing with a hand written

parser

I am not sure if there are some more forms of lefdef readers? So my general plan would be to write a more generic router (being more algorithmic agnostic providing a general data structure, to experiment with some routing algorithms, splitting in global and detailed routing and so on).

Before dealing with this I would opt to have some common core dealing with lefdef reading and writing. Is there maybe already some tool doing this? In the last week I started going through the lefdef ref manual 5.8 and to manually parse lef files into some c++ structs, see[1] to have a look yourself. From quite some big lef files it already can parse >90% the mayor chunck. Being quite uncommented and not tougth through to much, but before wasting more time on this I wanted to get some feedback.

Is the general lefdef-lib thing a idea worth spending more energy on? Is there already another attempt to do so in progress?

Thanks for information and feedback! P.S. Is there already something up for the hackathon? Would be definitely up for it ;) Paul

[1] https://github.com/paulwuertz/siliconverter _______________________________________________ Libre-silicon-devel mailing list Libre-silicon-devel@list.libresilicon.com http://list.libresilicon.com/mailman/listinfo/libre-silicon-devel