ChanServ changed the topic of #yosys to: Yosys Open SYnthesis Suite: https://github.com/YosysHQ/yosys/ | Channel logs: https://libera.irclog.whitequark.org/yosys/
tpb has quit [Remote host closed the connection]
tpb has joined #yosys
<lofty> so-offish: TRELLIS_COMB is a LUT4, or half a CCU2C, or a quarter of a TRELLIS_DPR16X4
<so-offish> lofty: Yeah but I need fixed routing; I need to be able to tell the tool "No really: This TRELLIS_COMB, right here"
<so-offish> lofty: I was kinda using TRELLIS_SLICE to get around the packer.
<so-offish> lofty: tnt: Thank you both for your help today. Gotta sign off; I'll bbl.
<lofty> so-offish: the tool for that has always been the BEL attribute in nextpnr.
bl0x_ has joined #yosys
bl0x has quit [Ping timeout: 268 seconds]
skipwich_ has joined #yosys
skipwich has quit [Ping timeout: 248 seconds]
skipwich_ has quit [Read error: Connection reset by peer]
skipwich has joined #yosys
skipwich_ has joined #yosys
skipwich has quit [Read error: Connection reset by peer]
AdamHorden has quit [Ping timeout: 260 seconds]
skipwich_ has quit [Quit: DISCONNECT]
skipwich has joined #yosys
pbsds has quit [Quit: The Lounge - https://thelounge.chat]
pbsds has joined #yosys
indy_ is now known as indy
AdamHorden has joined #yosys
lofty has quit [Ping timeout: 276 seconds]
lofty has joined #yosys
<datenwolf> ... *ooff*, I'm getting major Qt related PTSD flashbacks with regard to this exception in the UI code (one of my previous projects sufferd from similar issues after a Qt version bump)
tux3 has quit [Remote host closed the connection]
tux3 has joined #yosys
strobo has quit [Read error: Connection reset by peer]
strobo has joined #yosys
tlwoerner has quit [Quit: Leaving]
tlwoerner has joined #yosys
so-offish has quit [Ping timeout: 250 seconds]
so-offish has joined #yosys
<so-offish> lofty: saw your answer yesterday about the BEL attribute (rather than using TRELLIS_SLICE directly); thank you. :)
<datenwolf> identified the bug. Locales do strike once again... and here NextPNR is yet another example why stringly typed values are harmful. gui/worker.cc uses std::to_string to convert numerics into property strings (to_string does conversion with locale), timing.cc assign_budget will then retrieve those settings through Context::setting<T> which internally uses boost::lexical_cast<T> that for some reason
<datenwolf> won't use the same locale, as std::to_string. Recommended workaround: On GUI startup nuke the locale and force it to C
* datenwolf chuckles on the thought, that he's haunted by locale induced numeric<->string roundtrip conversion bugs. first_time_questionmark.gif
DiffieHellman has quit [Remote host closed the connection]
DiffieHellman has joined #yosys
sugarbeet has quit [Ping timeout: 260 seconds]
sugarbeet has joined #yosys
strobo has quit [Read error: Connection reset by peer]
<so-offish> ok everyone: who is super pumped to help me get setup to help debug this qt issue in nextpnr
<so-offish> That came out more mangled than I intended.
strobo has joined #yosys
dormito has quit [Ping timeout: 265 seconds]
dormito has joined #yosys
nonchip has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
nonchip has joined #yosys
<so-offish> datenwolf: which bug?
<so-offish> datenwolf: The one I reported earlier?