<cr1901>
whitequark[cis]: Just as a ping, I'll try to work on minisat and boolector today after I sleep. Am still trying to get stuff off my "things to do stack (incl yowasp-boolector)". We'll see if I mux it or fail.
notgull has quit [Ping timeout: 264 seconds]
notgull has joined #amaranth-lang
frgo has joined #amaranth-lang
frgo has quit [Quit: Leaving...]
Chips4MakersakaS has quit [Quit: Idle timeout reached: 172800s]
<_whitenotifier-5>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] df3968d - Deploying to main from @ amaranth-lang/amaranth@7d295b040a347e99383f4ba157382ab49b250529 🚀
<_whitenotifier-5>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] b377a9a - Deploying to main from @ amaranth-lang/amaranth@fc9369b8e19dd57e88a1b09396ff9a4a771e0170 🚀
<_whitenotifier-7>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] c8175ef - Deploying to main from @ amaranth-lang/amaranth@ea561378acb99554b52a57ac27f49a2f23761a2f 🚀
<tpw_rules>
i'm not on the latest commit but RTLIL is not reproducibile in certain circumstances it seems. is that a goal?
<tpw_rules>
hm, it may be something in amaranth-soc
<Wanda[cis]>
you mean we have nondeterministic output?
<tpw_rules>
yes
<Wanda[cis]>
I just squashed a bug that did exactly this
<Wanda[cis]>
another piece of _xfrm.py bites the dust
<tpw_rules>
no, the particular behavior i'm observing is not fixed. i think there's something in the amaranth-soc machinery hooking stuff up non-deterministically
<Wanda[cis]>
mhm
<tpw_rules>
rather than something inherent to rtlil or amaranth itself
<Wanda[cis]>
could be, I can't really comment on soc stuff
<_whitenotifier-5>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] 8ef6d6b - Deploying to main from @ amaranth-lang/amaranth@127fe1fd2e33f94520ceff8e42f04383f5f10591 🚀
<tpw_rules>
whitequark[cis]: you wrote it five years ago, but do you remember how the promise of iteration "in ascending order of their address" is guaranteed for various functions here: https://github.com/amaranth-lang/amaranth-soc/blob/main/amaranth_soc/memory.py ? as i see it they are iterated in order they are added
<whitequark[cis]>
i do not
<tpw_rules>
i think as documented is a valuable property to have, but the code doesn't currently do that. would you like an issue and possible PR to follow up and fix that?
<_whitenotifier-5>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] 88f6e6c - Deploying to main from @ amaranth-lang/amaranth@c2001fe9356174d12cab022f0bbc3261dbf47f5a 🚀
<_whitenotifier-7>
[amaranth-lang/amaranth-lang.github.io] github-merge-queue[bot] 5f1049a - Deploying to main from @ amaranth-lang/amaranth@31a12c03d19fef601b6515c431c5b686d2ba3702 🚀