03:38 <GitHub109> [scripts] xiangfu pushed 1 new commit to master: http://git.io/dFCYtQ

03:38 <GitHub109> [scripts/master] reflash_m1.sh : update help message - Xiangfu Liu

16:15 <Thihi> m

16:32 <qwebirc39462> hi

16:33 <qwebirc39462> what doe the milky mist do

16:53 <wpwrak> it's quiz time - you have 10 seconds to answer ! ;-)

16:56 <lekernel> "ERROR:PhysDesignRules:797 - Illegal placement. Component dummy must be placed in a bonded site." grmbl. obviously, if I figured out the undocumented UCF constraint to use unbonded I/O sites, I do want to use an unbonded I/O site ...

16:58 <lekernel> the whole P&R runs correctly on those sites, except it won't let me generate a programming file

16:59 <lekernel> I wonder if there is a hidden "skip DRC" option in bitgen. it would be convenient, for that and for reverse engineering purposes ...

17:59 <lekernel> http://www.youtube.com/watch?v=0h-RhyopUmc

17:59 <lekernel> flickernoise 2.0 patch editor should be like that :-)

18:07 <mwalle> lekernel: what can you do with unbonded io sites?

18:09 <lekernel> use the I/O delay elements internally

18:09 <lekernel> without wasting package pins

18:10 <lekernel> also, the Spartan-6 soft DDR PHY uses a dummy I/O site to help calibrate against PVT drift

18:11 <lekernel> I'm sure I saw unbonded I/O sites used in some designs... but maybe not in S6

18:11 <lekernel> there could be a good reason for that, e.g. if they no longer test unbonded I/Os

19:20 <lekernel> https://github.com/milkymist/llvm-lm32/commit/b1e6f6805b0115255f9f38bb65edb30b1cf8a845

23:33 <qi-bot> The Firmware build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-11062011-2242/