03:20 <Xark> Hello. I thought I would play with SymbiFlow, but following the "Getting Started" doc on Ubuntu 18.04, it fails for me on first "make" with "/xsltproc: error while loading shared libraries: libiconv.so.2: cannot open shared object file: No such file or directory". Is this a known issue (libiconv is apparently included as part of libc6 on Ubuntu)?

06:26 <sf-slack3> <kgugala> hi Xarc

06:27 <sf-slack3> <kgugala> Xarc: try running make all_conda first

06:27 <sf-slack3> <kgugala> and then the rest

06:27 <sf-slack3> <kgugala> conda should get the missing lib

15:09 <Xark> sf-slack3: Thanks! Okay, I will try that later. Appreciate the help. :)

18:18 <hackerfoo> A simple design with DRAM, PCIe, and GPIO connected through AXI takes ~20% of an xc7a200t with Vivado. That's more than I expected.

18:19 <hackerfoo> https://github.com/RHSResearchLLC/NiteFury/tree/master/Sample-Projects/Project-0/FPGA-A200T-3

18:19 <tpb> Title: NiteFury/Sample-Projects/Project-0/FPGA-A200T-3 at master · RHSResearchLLC/NiteFury · GitHub (at github.com)

18:40 <daveshah> Vivado's IP cores are pretty big, I wouldn't be surprised if it was less than half that with LiteX and its cores

18:40 <daveshah> https://www.bunniestudios.com/blog/?p=5018 has a comparison

18:41 <daveshah> > LiteX produces a design that uses about 20% of an XC7A50 FPGA with a runtime of about 10 minutes, whereas Vivado produces a design that consumes 85% of the same FPGA with a runtime of about 30-45 minutes.

19:46 <hackerfoo> Still, 10% of a 200T is a lot for VtR to handle when we're currently only working with 1/5th of a 35T.

19:47 <litghost> Plan is to use all of the 35T for LiteX

19:52 <hackerfoo> daveshah: Reading that article, skipping DRAM calibration seems iffy. Seems like companies providing IP can't cut corners like that.

19:52 <hackerfoo> Maybe you could do it with partial reconfiguration.

19:52 <daveshah> LiteX doesn't skip calibration

19:53 <hackerfoo> I'm just reading the article, not referring to LiteX.

19:53 <daveshah> Ah right

19:56 <hackerfoo> I guess I should try making a LiteX version of Project-0.

20:00 <hackerfoo> litghost: Is there a LiteX design yet?

20:06 <litghost> In prjxray? Yes, https://github.com/SymbiFlow/prjxray/tree/master/minitests/litex . For Symbiflow-arch-defs? Not yet, https://github.com/SymbiFlow/symbiflow-arch-defs/issues/911

20:06 <tpb> Title: prjxray/minitests/litex at master · SymbiFlow/prjxray · GitHub (at github.com)

20:11 <hackerfoo> litghost: Where is arty.py?

20:12 <hackerfoo> Wow. Don't try to look at top.v online.

20:12 <litghost> hackerfoo: mkurc made the README, might make an issue and assign to him about clarifying that

20:13 <litghost> hackerfoo: I don't know off hand

20:13 <hackerfoo> Thanks

20:26 <sf-slack3> <mkurc> @hackerfoo The prjxray repo contains only generated files. the 'arty.py' is part of one of LiteX repos.

20:27 <hackerfoo> mkurc: How do I know which version of arty.py generated the files? Also, which version of LiteX?

20:28 <sf-slack3> <mkurc> @hackerfoo Well, that was just a "minitest" so I didn't push that info to the repo.

20:29 <sf-slack3> <mkurc> @hackerfoo But I can find it out

20:29 <hackerfoo> I see. I still think documenting this is important.