02:53 <sb0> whitequark, where do I put my public key for ssh into ubuntu-trusty-buildbot.lab.m-labs.hk?

02:53 <sb0> and what username should I use?

02:55 <whitequark> sb0: I added your keys for the user sb

02:58 <sb0> whitequark, okay, how do I change for the buildbot user?

02:59 <whitequark> sudo -u buildbot -s ?

02:59 <sb0> sure, but I'm not in sudoers

03:06 <whitequark> oh

03:07 <whitequark> sec

03:07 <whitequark> done

03:08 <sb0> nope

03:08 <sb0> ah, yes, works now

03:08 <sb0> thanks

03:10 <sb0> bb-m-labs, force build --branch=release-2 artiq

03:10 <bb-m-labs> build forced [ETA 32m53s]

03:10 <bb-m-labs> I'll give a shout when the build finishes

03:15 <sb0> bb-m-labs, stop build artiq xxx

03:15 <bb-m-labs> build 1369 interrupted

03:16 <sb0> bb-m-labs, force build --branch=release-2 artiq

03:16 <bb-m-labs> The build has been queued, I'll give a shout when it starts

03:18 <bb-m-labs> build #1369 of artiq is complete: Exception [exception interrupted] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1369

03:18 <bb-m-labs> build forced [ETA 32m53s]

03:18 <bb-m-labs> I'll give a shout when the build finishes

03:28 <bb-m-labs> build #443 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/443

03:28 <bb-m-labs> build #1370 of artiq is complete: Failure [failed anaconda_upload anaconda_upload_1] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1370

03:29 <sb0> oh right, that vivado bug...

03:29 <sb0> the failed anaconda_upload is more mysterious

03:29 <sb0> bb-m-labs, force build --branch=release-2 artiq

03:29 <bb-m-labs> build forced [ETA 32m53s]

03:29 <bb-m-labs> I'll give a shout when the build finishes

03:32 <sb0> as usual, and especially with conda, things cannot go well. that was started by the seemingly innocuous artiq-dev package...

03:40 <GitHub> [artiq] sbourdeauducq commented on issue #670: @cjbe A warm welcome to the wonderful world of conda (and Python packaging in general). The build now fails because conda says it would put the package in ``/var/lib/buildbot/slaves/debian-stretch-amd64-1/miniconda/conda-bld/noarch``, but puts it into ``/var/lib/buildbot/slaves/debian-stretch-amd64-1/miniconda/conda-bld/linux-64`` instead. I do not know if there are other consequences to this cool new prob

03:48 <GitHub146> [buildbot-config] sbourdeauducq pushed 1 new commit to master: https://git.io/vy8tL

03:48 <GitHub146> buildbot-config/master a89ef53 Sebastien Bourdeauducq: work around new conda-build output name bug

03:48 <sb0> bb-m-labs, force build --branch=release-2 artiq

03:48 <bb-m-labs> build #1372 forced

03:48 <bb-m-labs> I'll give a shout when the build finishes

03:50 <bb-m-labs> build #445 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/445

03:53 <sb0> whitequark, should folders like /var/lib/buildbot/slaves/debian-stretch-amd64-2/miniconda/conda-bld/artiq-kc705-nist_clock_1488770826189 be deleted?

03:54 <sb0> bb-m-labs, force build --branch=release-2 artiq

03:54 <bb-m-labs> The build has been queued, I'll give a shout when it starts

03:59 <GitHub80> [smoltcp] whitequark pushed 1 new commit to master: https://git.io/vy8tM

03:59 <GitHub80> smoltcp/master b9e79d1 whitequark: Ignore Ethernet frames for which we aren't a destination.

04:01 <GitHub192> [smoltcp] whitequark pushed 1 new commit to master: https://git.io/vy8t9

04:01 <GitHub192> smoltcp/master 3dfcb11 whitequark: Follow-up to b9e79d19; always process broadcast frames.

04:01 <travis-ci> m-labs/smoltcp#86 (master - b9e79d1 : whitequark): The build passed.

04:01 <travis-ci> Change view : https://github.com/m-labs/smoltcp/compare/bb2edf4e8a48...b9e79d19a7d2

04:01 <travis-ci> Build details : https://travis-ci.org/m-labs/smoltcp/builds/208068582

04:02 <GitHub4> [smoltcp] whitequark pushed 1 new commit to master: https://git.io/vy8tF

04:02 <GitHub4> smoltcp/master 2aad62a whitequark: Update README to mention lack of multicast support.

04:02 <sb0> rjo, on kasli where is "configuring muxes to switch between backplane control and IDC control" coming from?

04:03 <travis-ci> m-labs/smoltcp#87 (master - 3dfcb11 : whitequark): The build passed.

04:03 <travis-ci> Change view : https://github.com/m-labs/smoltcp/compare/b9e79d19a7d2...3dfcb11d33e5

04:03 <travis-ci> Build details : https://travis-ci.org/m-labs/smoltcp/builds/208068913

04:04 <travis-ci> m-labs/smoltcp#88 (master - 2aad62a : whitequark): The build passed.

04:04 <travis-ci> Change view : https://github.com/m-labs/smoltcp/compare/3dfcb11d33e5...2aad62a79017

04:04 <travis-ci> Build details : https://travis-ci.org/m-labs/smoltcp/builds/208069047

04:05 <sb0> since when do we have backplane control?

04:08 <bb-m-labs> build #446 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/446

04:15 <bb-m-labs> build #431 of artiq-win64-test is complete: Failure [failed python_unittest] Build details are at http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/431

04:15 <bb-m-labs> build #1372 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1372

04:15 <bb-m-labs> build #1373 forced

04:15 <bb-m-labs> I'll give a shout when the build finishes

04:16 <sb0> of course, the conda garbage keeps using python 3.5.3 on windows

04:16 <sb0> bb-m-labs, stop build artiq

04:16 <bb-m-labs> try 'stop build WHICH <REASON>'

04:16 <sb0> bb-m-labs, stop build artiq fuck anaconda

04:16 <bb-m-labs> build 1373 interrupted

04:16 <bb-m-labs> build #1373 of artiq is complete: Exception [exception interrupted conda_remove] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1373

04:21 <GitHub> [artiq] sbourdeauducq commented on issue #670: Worked around that, but now the windows unittests broke. http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/431/steps/python_unittest/logs/stdio... https://github.com/m-labs/artiq/issues/670#issuecomment-284299547

04:23 <sb0> bb-m-labs, force build --branch=release-2 artiq

04:23 <bb-m-labs> build #1374 forced

04:23 <bb-m-labs> I'll give a shout when the build finishes

04:44 <bb-m-labs> build #447 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/447

04:50 <bb-m-labs> build #432 of artiq-win64-test is complete: Failure [failed python_unittest] Build details are at http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/432

04:50 <bb-m-labs> build #1374 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1374

11:12 <rjo> sb0: link?

12:43 <sb0> rjo, Thomas Harty's email

12:59 <rjo> well. he thinks that that would be needed or useful. i would disagree.

13:01 <rjo> at the end we agreed that we'd leave the backplane open to be used as some mlvds bus and not as transparent additional IDC extension busses.

13:03 <sb0> yes

13:20 <GitHub35> [misoc] whitequark pushed 1 new commit to master: https://git.io/vy8Nr

13:20 <GitHub35> misoc/master 5a60714 whitequark: wishbone.Arbiter: don't let one master starve others.

13:20 <whitequark> sb0: http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/432/steps/python_unittest/logs/stdio

13:20 <whitequark> these are not the fault of the DDS test anymore

13:23 <bb-m-labs> build #208 of misoc is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/misoc/builds/208

13:24 <GitHub> [artiq] whitequark pushed 1 new commit to master: https://github.com/m-labs/artiq/commit/12e22823a2be22bd27e72ed553f56555e14c08ea

13:24 <GitHub> artiq/master 12e2282 whitequark: conda: update misoc dependency.

13:34 <bb-m-labs> build #448 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/448 blamelist: whitequark <whitequark@whitequark.org>

13:34 <bb-m-labs> build #1375 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1375 blamelist: whitequark <whitequark@whitequark.org>

13:51 <sb0> yeah, I saw that, very annoying

13:53 <whitequark> >fatal: destination path '/var/lib/buildbot/slaves/debian-stretch-amd64-2/miniconda/conda-bld/artiq-kc705-nist_clock_1488807278127/work' already exists and is not an empty directory.

13:53 <whitequark> what the hell?

13:53 <sb0> the windows test failure, as well as this new bug, seem to have appeared after I updated conda-build to fix the mishandling of the artiq-dev dependency

13:54 <sb0> the new conda-build also will build noarch packages in the linux-64 folder, even though --output gives the path to the noarch package

13:54 <sb0> this constant stream of pesky fuckups is why we should avoid feature creep since those are good enough for wasting time already

13:55 <whitequark> elaborate?

13:55 <sb0> on?

13:55 <whitequark> I mean, what sort of feature creep causes this issue specifically

13:56 <sb0> oh, I'm just saying that things may take a lot more time than expected, since have to battle problems like conda breakage, weird fpga behavior, obscure proprietary chips, etc.

13:56 <whitequark> ah. sure.

13:58 <sb0> this particular issue isn't related to feature creep, but some discussions about the new artiq hardware are very optimistic...

16:03 <GitHub> [artiq] whitequark pushed 1 new commit to master: https://github.com/m-labs/artiq/commit/b0cdef3211b8a961699b301663ee3b3cf46a5148

16:03 <GitHub> artiq/master b0cdef3 whitequark: firmware: implement hotswap through artiq_coreboot tool.

16:03 <whitequark> finally artiq has a firmware replacement mechanism that isn't glitchy as hell

16:13 <bb-m-labs> build #449 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/449 blamelist: whitequark <whitequark@whitequark.org>

16:13 <bb-m-labs> build #1376 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1376 blamelist: whitequark <whitequark@whitequark.org>

16:22 <sb0> whitequark, you may want to add that to conda too

16:24 <sb0> ah nevermind, conda also uses setup.py now

16:48 <GitHub> [artiq] sbourdeauducq pushed 2 new commits to master: https://github.com/m-labs/artiq/compare/b0cdef3211b8...68952367947a

16:48 <GitHub> artiq/master 1e47e63 Sebastien Bourdeauducq: drtio: implement inputs in RTPacketSatellite, reorganize code

16:48 <GitHub> artiq/master 6895236 Sebastien Bourdeauducq: frontend: fix permissions

16:54 <sb0> bb-m-labs, force build artiq

16:54 <bb-m-labs> The build has been queued, I'll give a shout when it starts

16:58 <bb-m-labs> build #450 of artiq-board is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/450 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>

16:58 <bb-m-labs> build #1377 of artiq is complete: Failure [failed] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1377 blamelist: Sebastien Bourdeauducq <sb@m-labs.hk>

16:58 <bb-m-labs> build #1378 forced

16:58 <bb-m-labs> I'll give a shout when the build finishes

17:00 <bb-m-labs> build #1378 of artiq is complete: Failure [failed conda_create] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1378

17:14 <sb0> bb-m-labs, force build artiq

17:14 <bb-m-labs> build #1379 forced

17:14 <bb-m-labs> I'll give a shout when the build finishes

17:14 <sb0> conda is impressively shitty

17:16 <bb-m-labs> build #1379 of artiq is complete: Failure [failed conda_create] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1379