<sb0>
whitequark: why not use generic code and the synthesizer?
<sb0>
instead of platform-dependent code with LUT instantiations
<whitequark>
sb0: so that they don't get optimized into nothing
<whitequark>
if you're using pseudo-differential I/O that isn't registered in the IOB you want same delay on both
<sb0>
hmm
<whitequark>
it's niche, yes
<whitequark>
but I just wanted to make it work well no matter the requested settings
<whitequark>
I tested it on ice40 and the inverter delay skews the edges quite noticeably
<_whitenotifier-3>
[nmigen] whitequark commented on issue #78: Implement XDR buffer factories for all platforms - https://git.io/fjruM
<_whitenotifier-3>
[nmigen] whitequark closed issue #78: Implement XDR buffer factories for all platforms - https://git.io/fjEMr
rohitksingh has joined #m-labs
<whitequark>
sb0: should we reword the "under development" README warning in nMigen yet?
<whitequark>
I don't really anticipate any more core language changes
_whitelogger has joined #m-labs
mumptai has joined #m-labs
<_whitenotifier-3>
[nmigen] whitequark commented on issue #92: Using Yosys with Xilinx Series 7 devices (and non-standard tool flows) - https://git.io/fjrzz
rohitksingh has quit [Remote host closed the connection]
kbeckmann has quit [Ping timeout: 248 seconds]
kbeckmann has joined #m-labs
mumptai has quit [Remote host closed the connection]