00:53 <GitHub1> [artiq] sbourdeauducq commented on issue #931: That's a conda bug, iirc worked around in artiq 2.4. https://github.com/m-labs/artiq/issues/931#issuecomment-368181107

00:56 <GitHub46> [artiq] sbourdeauducq commented on issue #908: It's not just a change in the timing, it seems that jitter also increases significantly. Noise/jitter issues (on-chip or on-board), e.g. crosstalk, MMCM failure. https://github.com/m-labs/artiq/issues/908#issuecomment-368181491

02:04 <GitHub88> [artiq] jonaskeller commented on issue #902: So far, it's looking good. I left it idle for 6h and didn't see it crash, then set `net_trace` to 1 and left an experiment running in a loop to stress the core device. It's been running in that mode for over 2h now, which is a lot considering that I never saw it survive for more than a minute before. I'll leave it running over the weekend and let you know Monday how it went

07:16 <GitHub143> [artiq] jordens commented on issue #932: You'll need to make ually downgrade openocd to a version that was current when 2.4 was released https://github.com/m-labs/artiq/issues/932#issuecomment-368207047

08:41 <rjo> cjbe: what do you use lab.m-labs for?

08:41 <rjo> cjbe: i haven't done drtio but i would use either the DAC or the 10G BiDis that I mentioned a couple of weeks ago.

08:55 <GitHub86> [artiq] jordens pushed 1 new commit to master: https://github.com/m-labs/artiq/commit/b466a569bf49e50559234209f9f42032df50e89b

08:55 <GitHub86> artiq/master b466a56 Robert Jordens: coredevice: export spi2

08:55 <GitHub149> [artiq] jordens force-pushed spi2 from 7bd3ead to 54b5149: https://github.com/m-labs/artiq/commits/spi2

08:55 <GitHub149> artiq/spi2 3322733 Robert Jordens: hmc830: be explicit about SPI mode selection

08:55 <GitHub149> artiq/spi2 478b65d Robert Jordens: firmware, sayma: port converter_spi to spi2...

08:55 <GitHub149> artiq/spi2 e47dabe Robert Jordens: conda: bump migen (sayma_rtm spi pulls)

08:56 <rjo> bb-m-labs: force build artiq --branch=spi2

08:56 <bb-m-labs> The build has been queued, I'll give a shout when it starts

09:06 <GitHub56> [artiq] sbourdeauducq commented on issue #932: @jordens can you delete (or move to the dev channel) the buggy openocd versions? If course they are the ones that get installed as per Murphy's law... https://github.com/m-labs/artiq/issues/932#issuecomment-368213549

09:07 <bb-m-labs> build #1268 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/1268 blamelist: Robert Jordens <rj@m-labs.hk>

09:07 <bb-m-labs> build #2105 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/2105 blamelist: Robert Jordens <rj@m-labs.hk>

09:07 <bb-m-labs> build #2106 forced

09:07 <bb-m-labs> I'll give a shout when the build finishes

09:17 <bb-m-labs> build #1269 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/1269

09:17 <bb-m-labs> build #2106 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/2106

10:09 <GitHub167> [artiq] jordens commented on issue #932: @sbourdeauducq It's not buggy openocd versions but incompatibility between the proxy bitstream shipped in the old board packages and the new openocd versions. Moving openocd won't help.... https://github.com/m-labs/artiq/issues/932#issuecomment-368217430

10:47 <GitHub174> [pdq] jordens created spi2 (+1 new commit): https://github.com/m-labs/pdq/commit/65f2a2c10b07

10:47 <GitHub174> pdq/spi2 65f2a2c Robert Jordens: port to new artiq/misoc spi2 core

12:26 <GitHub0> [smoltcp] whitequark commented on pull request #167 6cf2c03: No, don't add an `NdiscPacket`, just move the `impl` blocks into a separate file. https://github.com/m-labs/smoltcp/pull/167#discussion_r170420091

12:26 <GitHub18> [smoltcp] whitequark commented on pull request #167 6cf2c03: No, don't add an `NdiscPacket`, just move the `impl` blocks and enums into a separate file. https://github.com/m-labs/smoltcp/pull/167#discussion_r170420091

12:44 <GitHub127> [artiq] jordens deleted urukul-ashafir at ca3d829: https://github.com/m-labs/artiq/commit/ca3d829

13:25 <GitHub77> [smoltcp] batonius commented on issue #166: > what with the many ManagedSlice::Owned instances in your PoC code.... https://github.com/m-labs/smoltcp/issues/166#issuecomment-368228359

14:13 <GitHub156> [smoltcp] whitequark commented on issue #166: I'm not happy about this. You're making smoltcp perform significantly worse for the use case it is primarily intended for, which is small embedded devices. I don't think this is acceptable. https://github.com/m-labs/smoltcp/issues/166#issuecomment-368231079

16:46 <GitHub52> [smoltcp] batonius commented on issue #166: Then I'll try to rethink the design to keep it this way, maybe along the lines of the old `Device` trait, with handler types owning packet buffers and returning them in destructors. The counterexample of loopback based on `RingBuffer` would still be impossible tho. https://github.com/m-labs/smoltcp/issues/166#issuecomment-368241377

17:07 <GitHub78> [artiq] jordens pushed 1 new commit to release-2: https://github.com/m-labs/artiq/commit/3aafcc621a691ab4a101c72090d7d9f707ad2095

17:07 <GitHub78> artiq/release-2 3aafcc6 Robert Jordens: conda: pin openocd (artiq, artiq-dev) #932

17:12 <bb-m-labs> build #2107 of artiq is complete: Failure [failed python_unittest_1] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/2107 blamelist: Robert Jordens <rj@m-labs.hk>

17:14 <GitHub56> [smoltcp] dlrobertson opened issue #170: Improve socket lookup time. https://github.com/m-labs/smoltcp/issues/170

17:42 <GitHub175> [smoltcp] batonius commented on issue #170: I've tried to do that in #19, and even had a working prototype, but eventually lost interest, the `SocketRef` type we have is a remnant of that attempt. https://github.com/m-labs/smoltcp/issues/170#issuecomment-368245570

17:46 <GitHub113> [smoltcp] dlrobertson closed issue #170: Improve socket lookup time. https://github.com/m-labs/smoltcp/issues/170

17:46 <GitHub97> [smoltcp] dlrobertson commented on issue #170: Gotcha. Didn't look back that far. Post IPv6 I might look into that. Closing this as a duplicate of #19. https://github.com/m-labs/smoltcp/issues/170#issuecomment-368245860

21:04 <GitHub158> [smoltcp] whitequark commented on issue #166: > with handler types owning packet buffers and returning them in destructors... https://github.com/m-labs/smoltcp/issues/166#issuecomment-368260125