<xobs>
cr1901_modern: I'm using the embedded version -- the other install prevents you from installing more than one version, and naturally can't be distributed.
<cr1901_modern>
embedded version?
<xobs>
cr1901_modern: zipimport.ZipImportError: can't decompress data; zlib not available
<Dar1us>
whitequark: as in, too big to distribute?
<sb0>
or port/rewrite nix to windows :)
<Dar1us>
do a Xilinx and ship it in a Linux VM ;)
<sb0>
something like the "anaconda navigator" GUI but using a nix or nix-like layer would both work well and be easy to use for people who don't want to mess with package management
<xobs>
Right now what I do is ship binaries and have people add it to their PATH. Then I add things like migen, litex, and such to PYTHONPATH via a script that re-executes itself. This also sets PYTHONHASHSEED.
<sb0>
this PYTHONHASHSEED thing is a hack
<xobs>
This is nice for a few reasons: (1) it's completely offline, and so I can give it to people at a workshop where the Internet is usually terrible, (2) it works cross-platform, and (3) the uninstall process is "delete the toolchain directory".
<sb0>
newer migen/misoc/nmigen should not need it, otherwise it's a bug
lkcl has joined #m-labs
harryho has quit [Remote host closed the connection]
harryho has joined #m-labs
rohitksingh has quit [Ping timeout: 276 seconds]
kjh-m has quit [Ping timeout: 245 seconds]
kjh-m has joined #m-labs
jaeckel has quit [Ping timeout: 245 seconds]
<_whitenotifier>
[m-labs/nmigen] whitequark pushed 2 commits to master [+0/-0/±2] https://git.io/JenNR
<_whitenotifier>
[m-labs/nmigen] whitequark 3a1dae5 - back.rtlil: it is not necessary to match binop operand width.
<_whitenotifier>
[m-labs/nmigen] whitequark d3f7cc8 - back.rtlil: sign of rhs and lhs of ${sshr,sshl,pow} don't need to match.
<Dar1us>
hmm is verilator supposed to be slower doing FST than VCD dumping?
<Dar1us>
oops this VM is set for only 1 CPU, that probably does help
<Dar1us>
*doesn't
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<Dar1us>
hmm even so, very very slow
<Dar1us>
TBH I am pretty surprised - I would expect compression to have fairly minimal overhead and would be a net win due to greatly reduced disk IO (even onto an SSD)
<whitequark>
depends on compression algo. if it's bz2...
<Dar1us>
hmm I guess so
<Dar1us>
I only linked with -lz so I assume it's just gzip
<_whitenotifier>
[m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/Jenx7