00:00
playthatbeat has joined #m-labs
01:35
fengling has joined #m-labs
01:52
<
rjo >
sb0: exactly.
02:32
xiangfu has joined #m-labs
02:42
fengling has quit [Quit: WeeChat 1.0]
02:45
fengling has joined #m-labs
03:04
nengel_ is now known as nengel
03:17
xiangfu has quit [Ping timeout: 240 seconds]
03:20
xiangfu has joined #m-labs
04:01
fengling has quit [Ping timeout: 244 seconds]
06:58
<
GitHub48 >
artiq/master 27fc19e Sebastien Bourdeauducq: pyon: add doc
06:58
<
GitHub48 >
artiq/master 934442b Sebastien Bourdeauducq: pc_rpc: document
06:58
<
GitHub48 >
artiq/master ea37274 Sebastien Bourdeauducq: test/pc_rpc: support slow server startups
07:09
sb0_ has joined #m-labs
07:18
xiangfu has quit [Ping timeout: 265 seconds]
07:19
xiangfu has joined #m-labs
08:35
FabM has joined #m-labs
08:37
<
GitHub60 >
artiq/master 4d0e5db Sebastien Bourdeauducq: doc/manual: split core/controller drivers
08:37
<
GitHub60 >
artiq/master 4eda58f Sebastien Bourdeauducq: doc/manual: tutorial -> getting started
08:38
sb0_ has quit [Quit: Leaving]
08:59
_florent_ has joined #m-labs
09:38
bentley` has quit [Remote host closed the connection]
09:54
MY123 has joined #m-labs
10:17
xiangfu has quit [Remote host closed the connection]
10:44
<
ysionneau >
sb0 : I noticed that If((row_open == 1) and (b == slicer.bank(bus.adr)), does not produce what I expected
10:44
<
ysionneau >
but this does If(Cat(row_open,slicer.bank(bus.adr)) == Cat(1, b),
10:44
<
ysionneau >
is the latter the "best way" of expression the logic "and" ?
10:45
<
ysionneau >
expressing*
10:45
<
sb0 >
"and" is not supported
10:45
<
ysionneau >
between _Operator it will act as logical and?
10:46
<
sb0 >
== returns 1 bit, so logical and boolean "and" are the same
10:46
<
ysionneau >
ah sure
10:46
<
ysionneau >
stupid question :)
10:46
<
sb0 >
the problem is Python doesn't support redefining and/or boolean ops
10:47
<
sb0 >
but maybe we can throw an error when one does try to use them
10:47
<
sb0 >
ysionneau, can you see if adding def __bool__(self): raise NotImplementedError in migen.fhdl.Value does that (and has no ill effects)?
10:47
<
ysionneau >
yep because here it was producing verilog, but just not what I expected
10:48
<
ysionneau >
ok I'll have a look
10:48
<
sb0 >
the Python interpreter will try to evaluate the arguments of and/or as bools - we can make that fail loudly
10:48
<
ysionneau >
(use & and | instead of 'and' 'or' fixes my issue)
10:51
<
ysionneau >
sb0: adding the
__bool__ prevents the wrong code to produce verilog
10:51
<
ysionneau >
I will keep it here and see if I get weird issues
10:53
<
sb0 >
build misoc with it ;)
10:53
<
ysionneau >
let's do that!
10:57
<
ysionneau >
at least it didn't raise any exception, let's see if the bitstream will boot
11:10
<
ysionneau >
bios and app boot :) (cpu is or1k)
11:11
<
sb0 >
cool. add a nice message to the exception telling the user to make sure they use & | and send a patch.
11:15
<
ysionneau >
NotImplementedError: For logic operations between expressions: use '&'/'|' instead of 'and'/'or'
11:16
<
ysionneau >
sounds OK?
11:19
<
sb0 >
boolean operations
11:22
bentley` has joined #m-labs
11:42
<
GitHub73 >
migen/master 286092b Yann Sionneau: Raise exception when not using correct boolean operators
12:40
<
GitHub44 >
artiq/master 1e5b05d Sebastien Bourdeauducq: pdq2: fix commandline arguments
12:40
<
GitHub44 >
artiq/master 2946fa5 Sebastien Bourdeauducq: pc_rpc: factor out asyncio server code
12:40
<
GitHub44 >
artiq/master d5a3f3e Sebastien Bourdeauducq: doc/manual: driver writing tutorial
17:31
<
ysionneau >
sb0 is I do a sim.run(500) , is it always supposed to finish after 500 clock ticks? What could cause the simulator to get stuck and never finish?
17:32
<
ysionneau >
cause I just changed a few lines (I don't have the exact diff but nothing fancy) and now it does not finish anymore
17:52
kilae has joined #m-labs
18:02
mumptai has joined #m-labs
18:17
<
ysionneau >
what's basically the coding style issue I could avoid?
18:17
<
ysionneau >
hi
_florent_ :) and thanks for the patch!
18:20
<
_florent_ >
I know that if in a fsm (or a combinatorial block?), if you do:
18:20
<
_florent_ >
b.eq(a)
18:20
<
_florent_ >
c.eq(b)
18:20
<
_florent_ >
it stucks the simulator
18:21
<
ysionneau >
_florent_: indeed your patch fixes my simulation issue!
18:22
<
ysionneau >
many thanks!
19:01
MY123 has quit [Quit: Connection closed for inactivity]
19:43
mithro has quit [Ping timeout: 272 seconds]
19:44
mithro has joined #m-labs
19:47
<
ysionneau >
very nice!
20:34
<
mumptai >
well done
20:35
<
mumptai >
and they don't personalize it like JAXA did with hayabusa
21:17
kilae has quit [Quit: ChatZilla 0.9.91 [Firefox 32.0.3/20140923175406]]
22:28
mumptai has quit [Ping timeout: 265 seconds]
23:31
_florent_ has quit [Quit: Leaving]