01:25 <GitHub51> [artiq] sbourdeauducq commented on commit 708c25b: Why? https://git.io/v1ysn

01:28 <GitHub152> [artiq] sbourdeauducq pushed 1 new commit to master: https://git.io/v1ys2

01:28 <GitHub152> artiq/master 6b99858 Sebastien Bourdeauducq: rtio: use same reset for counter_rtio whatever the interface delay is

02:07 <bb-m-labs> build #271 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/271

02:30 <bb-m-labs> build #1170 of artiq is complete: Failure [failed python_unittest_1] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1170 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>

03:50 <GitHub126> [artiq] sbourdeauducq commented on issue #637: > just getting the emulator to cooperate and to export basic MiSoC peripherals like the timer and mailbox is a major work.... https://git.io/v1y8a

04:05 <sb0> rjo, how far are you into the new pdq code?

04:19 <sb0> shall we get this over soon?

04:20 <sb0> whitequark, how do you plan on getting rid of the runtime/ and runtime.rs/ folders? we need to reorganize the firmware to support different targets (runtime, drtio satellite manager)

04:22 <sb0> those will share code like i2c bitbanging stuff and clock chip drivers

05:08 <whitequark> sb0: by killing lwip, and moving the remainig glue C file into the rust stuff, and building it through cargo likely

05:08 <whitequark> as for reorganize: there are several ways

05:09 <whitequark> I think an elegant one is putting the HAL in its own crate, and then masking features in it using #[cfg(feature = "i2c")] and such

05:41 <GitHub175> [smoltcp] whitequark pushed 2 new commits to master: https://git.io/v1yEd

05:41 <GitHub175> smoltcp/master 3377b91 whitequark: Implement network part of communication through UDP sockets.

05:41 <GitHub175> smoltcp/master 659f6ed whitequark: Implement UDP sockets.

05:41 <whitequark> ok, that's almost all of the buffering story done. let me finish them this evening...

07:39 <GitHub84> [artiq] jordens commented on commit 708c25b: There is no rtio clock yet. And dac_setup will do it later anyway. https://git.io/v1ywX

08:41 <mithro> Is there a good example of a test which writes some values to CSR registers and then reads output signals? I think https://github.com/m-labs/misoc/blob/master/misoc/test/test_spi.py is an okay style to copy?

08:43 <sb0> mithro, yes or you can even not map the csrs on a bus, look at the artiq drtio tests

08:43 <sb0> there are some rough edges with that, e.g. write_from_dev doesn't work

08:44 <sb0> mithro, how are the docs?

08:45 <mithro> sb0: On my todo list for this week, as soon as we (IE _florent_) gets the encoder stuff working I'll be looking at it

08:45 <mithro> sb0: Just sent a pull request with some minor docs :-P

08:46 <GitHub147> [misoc] sbourdeauducq pushed 1 new commit to master: https://git.io/v1yP2

08:46 <GitHub147> misoc/master 25d9e5d Tim 'mithro' Ansell: Adding some documentation to code_8b10b.py

08:48 <bb-m-labs> build #185 of misoc is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/misoc/builds/185

09:25 <rjo> sb0: pdq needs kernel support still.

09:33 <GitHub149> [artiq] jordens opened issue #641: startup_kernels hang https://git.io/v1yyv

09:35 <sb0> rjo, are you sure it's the cri/dma changes that broke the startup kernels?

09:36 <sb0> if it contains an empty run() it shouldn't touch that part... hm

09:36 <rjo> it worked before i merged master.

09:39 <bb-m-labs> build #272 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/272

09:43 <rjo> i.e. from that perspective it happened betwee 68de724 and 4b61020

09:46 <mithro> My simtest seems to be stuck in some way on the following statement -> http://paste.debian.net/902280/

09:49 <mithro> Hrm, it seems to be actually something to do with the fsm that I'm using

09:51 <rjo> sb0: maybe the fact that phaser has no rtio clock at that point triggers the bug. but it worked fine before.

09:53 <mithro> Ahh, it's because I'm not using NextValue...

10:03 <bb-m-labs> build #1171 of artiq is complete: Failure [failed python_unittest_1] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1171

19:43 <GitHub166> [artiq] dleibrandt opened issue #642: HasEnvironment prepare method https://git.io/v19GN