<GitHub-m-labs>
[artiq] sbourdeauducq commented on issue #1249: This is correct. 2 is a 32-bit integer which you are raising to the 48th power modulo 2**32, which yields 0. Same problem with shifting 1.... https://github.com/m-labs/artiq/issues/1249#issuecomment-454607387
<sb0>
whitequark: nothing is using the replace feature anymore though...
<whitequark>
sb0: what was using it?
<whitequark>
it has somewhat strange behavior when combined with FWFT, too
rohitksingh has quit [Ping timeout: 244 seconds]
rohitksingh has joined #m-labs
<GitHub-m-labs>
[artiq] whitequark commented on issue #1249: Thanks. If constant folding is implemented correctly, it never changes semantics of the code being compiled. As such, your assertions about constant folding happening or not are irrelevant (unless you are trying to highlight a bug in constant folding passes, that is.) https://github.com/m-labs/artiq/issues/1249#issuecomment-454854096
<_whitenotifier-c>
[m-labs/nmigen] whitequark pushed 3 commits to master [+2/-0/±3] https://git.io/fhllf