06:59 <larsc> it's evidence ;)

08:38 <ysionneau> I'm probably not the right person to list the cool features of NetBSD since I only chose based on those anecdotes

08:39 <ysionneau> but some facts are that among the goals of the projects are : do things the right way (who don't ? I don't know) *and* be very portable

20:00 <ysionneau> First try at pushing NetBSD kernel binary to the real Milkymist One hardware through flterm

20:00 <ysionneau> obviously nothing shows up

20:00 <ysionneau> retrying just in case :)

20:03 <ysionneau> same thing for 2nd try :(

20:04 <ysionneau> Let's try to use gdb on the M1 :) (never tried gdb on lm32 on anything else than qemu)

20:13 <sb0> you should use netboot

20:13 <sb0> flterm takes forever

20:13 <ysionneau> yes it's slow

20:14 <davidc__> sb0: you see what I said about Lattice stuff?

20:18 <bvernoux> hi

20:19 <bvernoux> does anyone here have backed and received a parallella board ?

20:30 <ysionneau> sb0: is it possible to debug via GDB *and* to use flterm (both connected to the same /dev/ttyUSB0 then) to type something in the bios ?

20:30 <ysionneau> can both read-write ttyUS0?

20:30 <ysionneau> ah ok, gdb-passthrough :)

20:37 <ysionneau> seems to work, cool

20:43 <ysionneau> humm is it normal that if I do set $pc=0x40000000 then "si" it executes at pc= 0x10000020 ? ^^"

20:43 <ysionneau> does it only report PC>>2 (since PC inside lm32 core is only 30 bits)

20:50 <ysionneau> ah, I guess it means I'm debugging the debugger...

20:52 <sb0> davidc__, yes, I saw it

20:55 <davidc__> sb0: I'll get it to ya at some point, but probably not what you were looking for

21:12 <ysionneau> what's monitor.rom ?

21:13 <ysionneau> ah, I guess it's for jtag

21:13 <ysionneau> I don't think I care then

21:26 <mwalle> larsc: (localparam some long time ago) i opened a case for that problem some time ago, but xilinx only said "use vivadi (or whatever the name for the new tools is)

21:30 <ysionneau> gn8

21:31 <mwalle> hi ysionneau

21:31 <mwalle> :)

21:32 <mwalle> and yeah, the you already found the passthrough parameter ;)

21:34 <mwalle> ysionneau: you could try software breakpoints with single stepping, which might or might not work better

21:35 <mwalle> i dont know if you can make gdb use soft/hardware breakpoints for singlestepping