*** tpb has joined #yosys | 00:00 | |
*** knielsen has quit IRC | 01:20 | |
*** kraiskil has joined #yosys | 01:20 | |
*** somlo has joined #yosys | 01:26 | |
*** rohitksingh has joined #yosys | 02:01 | |
*** knielsen has joined #yosys | 03:38 | |
*** citypw has joined #yosys | 03:43 | |
*** PyroPeter has quit IRC | 03:58 | |
*** PyroPeter has joined #yosys | 04:12 | |
*** _whitelogger has quit IRC | 04:15 | |
*** _whitelogger has joined #yosys | 04:17 | |
emeb_mac | So I spent some time fiddling around with building latest yosys on rpi4. No idea why clang wouldn't build it, but it always crashed on the very first file - even when I #ifdefed the entire contents of the file. | 04:18 |
---|---|---|
emeb_mac | I tweaked the makefile to use gcc instead of clang and it appears to be building just fine though | 04:18 |
*** craigo has quit IRC | 04:40 | |
*** _whitelogger has quit IRC | 05:42 | |
*** _whitelogger has joined #yosys | 05:44 | |
*** emeb_mac has quit IRC | 06:42 | |
*** Jybz has joined #yosys | 06:45 | |
*** Cerpin has quit IRC | 06:47 | |
*** Cerpin has joined #yosys | 06:49 | |
*** kraiskil has quit IRC | 06:56 | |
*** _whitelogger has quit IRC | 07:36 | |
*** _whitelogger has joined #yosys | 07:38 | |
corecode | clang crashed? | 08:16 |
*** rohitksingh has quit IRC | 08:27 | |
*** rohitksingh has joined #yosys | 08:33 | |
*** fsasm has joined #yosys | 09:08 | |
*** adjtm_ has quit IRC | 10:09 | |
*** _whitelogger has quit IRC | 10:24 | |
*** _whitelogger has joined #yosys | 10:26 | |
*** vidbina_ has joined #yosys | 10:31 | |
*** vidbina_ has quit IRC | 10:53 | |
*** cr1901_modern has quit IRC | 11:49 | |
*** cr1901_modern has joined #yosys | 12:04 | |
*** Jybz has quit IRC | 12:27 | |
*** proteusguy has joined #yosys | 12:27 | |
*** Jybz has joined #yosys | 12:28 | |
ZirconiumX | daveshah: For $mul mapping, is it better to map $macc or use mul2dsp? | 12:32 |
daveshah | ZirconiumX: definitely mul2dsp | 12:33 |
daveshah | $macc is a pretty nasty cell for internal mapping only imo | 12:33 |
ZirconiumX | So, the Cyclone V has variable-width multipliers; you can set them to be 9x9, 18x18 or 27x27 | 12:34 |
ZirconiumX | But mul2dsp seems to assume only a single size of multiplier. | 12:35 |
daveshah | That would need changing | 12:41 |
daveshah | You could probably map to an intermediate cell of the largest size and then down map as appropriate to start with | 12:42 |
daveshah | But the decomposition might need work to optimally use the smaller multipliers | 12:42 |
ZirconiumX | Worth filling a bug as a feature request? | 12:48 |
daveshah | Sure | 12:50 |
*** flaviusb has quit IRC | 13:05 | |
*** rohitksingh has quit IRC | 14:09 | |
*** kraiskil has joined #yosys | 14:17 | |
*** Laksen has joined #yosys | 14:48 | |
*** craigo has joined #yosys | 15:02 | |
*** emeb has joined #yosys | 15:21 | |
*** kraiskil has quit IRC | 15:52 | |
*** X-Scale` has joined #yosys | 16:32 | |
*** X-Scale has quit IRC | 16:33 | |
*** X-Scale` is now known as X-Scale | 16:33 | |
*** Laksen has quit IRC | 16:38 | |
*** citypw has quit IRC | 16:48 | |
*** emeb has quit IRC | 17:27 | |
*** emeb_mac has joined #yosys | 17:28 | |
*** emeb_mac has quit IRC | 17:33 | |
*** X-Scale` has joined #yosys | 18:32 | |
*** X-Scale has quit IRC | 18:33 | |
*** X-Scale` is now known as X-Scale | 18:33 | |
*** zkms has quit IRC | 18:48 | |
*** zkms has joined #yosys | 18:49 | |
*** craigo has quit IRC | 19:01 | |
*** emeb has joined #yosys | 19:37 | |
*** emeb has quit IRC | 19:49 | |
*** emeb has joined #yosys | 19:51 | |
*** emeb_mac has joined #yosys | 20:48 | |
*** rohitksingh has joined #yosys | 21:50 | |
*** adjtm has joined #yosys | 22:13 | |
*** Cerpin has quit IRC | 22:25 | |
*** fsasm has quit IRC | 22:29 | |
*** Jybz has quit IRC | 22:32 | |
*** pie_ has quit IRC | 22:33 | |
*** craigo has joined #yosys | 22:34 | |
*** Cerpin has joined #yosys | 23:01 | |
*** pie_ has joined #yosys | 23:06 | |
*** emeb has quit IRC | 23:57 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!