01:48 <GitHub63> [smoltcp] dlrobertson commented on issue #236: > ManagedMap means you need alloc at least, so it won't work anymore on bare-metal systems.... https://github.com/m-labs/smoltcp/pull/236#issuecomment-406762344

01:50 <sb0> Samuel, mismatch between openocd version and flash proxy bitstream version, probably

03:16 <sb0> Samuel, it's loading the one that comes with artiq. iirc either it should load the one that comes in the proxy bitstream package, or be an older openocd version.

05:41 <GitHub-m-labs> [artiq] sbourdeauducq pushed 5 new commits to master: https://github.com/m-labs/artiq/compare/766d87f626ad...7b75026391fe

05:41 <GitHub-m-labs> artiq/master 031de58 Sebastien Bourdeauducq: grabber: complete RTIO PHY, untested

05:41 <GitHub-m-labs> artiq/master 3638a96 Sebastien Bourdeauducq: kasli: add false path between RTIO and CL clocks

05:41 <GitHub-m-labs> artiq/master e3ba4b9 Sebastien Bourdeauducq: grabber: minor ROI engine cleanup, export count_len, cap count width to 31

06:25 <bb-m-labs> build #1733 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/1733

06:36 <bb-m-labs> build #914 of artiq-win64-test is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/914

06:36 <bb-m-labs> build #2526 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/2526

07:47 <sb0> bb-m-labs: force build --props=package=jesd204b conda-lin64

07:47 <bb-m-labs> build #412 forced

07:47 <bb-m-labs> I'll give a shout when the build finishes

07:49 <bb-m-labs> build #412 of conda-lin64 is complete: Failure [failed conda_build] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/412

07:50 <sb0> bb-m-labs: force build --props=package=jesd204b conda-lin64

07:50 <bb-m-labs> build #413 forced

07:50 <bb-m-labs> I'll give a shout when the build finishes

07:51 <bb-m-labs> build #413 of conda-lin64 is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/conda-lin64/builds/413

07:52 <GitHub-m-labs> [artiq] sbourdeauducq commented on issue #1080: @enjoy-digital Thanks for updating the GTH with all the wizard parameters and testing on KCU105. I tested the new code on the HK board and didn't see regressions, so I committed it.... https://github.com/m-labs/artiq/issues/1080#issuecomment-406778764

07:53 <GitHub-m-labs> [artiq] sbourdeauducq commented on issue #1080: @enjoy-digital Thanks for updating the GTH with all the wizard parameters and testing on KCU105. I tested the new code on the HK board and didn't see regressions, so I committed it.... https://github.com/m-labs/artiq/issues/1080#issuecomment-406778764

08:27 <bb-m-labs> build #1734 of artiq-board is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-board/builds/1734

08:37 <bb-m-labs> build #915 of artiq-win64-test is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq-win64-test/builds/915

08:38 <bb-m-labs> build #2527 of artiq is complete: Success [build successful] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/2527

13:04 <GitHub-m-labs> [artiq] klickverbot commented on issue #1113: Oh, the problem here is much more pedestrian: DMA of course doesn't advance `TTLInOut.i_previous_timestamp`, so `count()` still uses the timeout from when the sequence was recorded (i.e. `gate_*` was actually called). Same would go for output events and `.sync()`, although that is much less frequently used.... https://github.com/m-labs/artiq/issues/1113#issuecom

13:18 <GitHub-m-labs> [artiq] jordens commented on issue #1113: Maybe invalidate `i_previous_timestamp` in `count()` (and elsewhere) to prevent it being used wrong in the DMA case and describe how to manually compute it and set it when doing `count()` ing a DMA gate. Or change the TTL api to enforce returning the gate closure time from `gate()` and require using that in `count()` (and elsewhere). https://github.com/m-labs/artiq/

14:20 <GitHub-m-labs> [artiq] klickverbot commented on issue #1113: Requiring a timestamp argument to `count` is perhaps the cleanest way to solve this. I was going to add this anyway to be able to write correct user code without fiddling with `i_previous_timestamp` or directly using `rtio_input_timestamp`, but we could remove the no-timestamp version with the next major release.... https://github.com/m-labs/artiq/issues/1113#is

17:35 <GitHub-m-labs> [artiq] jordens commented on issue #1113: I'd like to have `sync_mu(t_mu)` as a method of the core device though. It's a frequent pattern. https://github.com/m-labs/artiq/issues/1113#issuecomment-406811778

18:32 <GitHub-m-labs> [artiq] klickverbot commented on issue #1113: Indeed, and currently a rather clunky one (manually writing out the loop), unless I've missed something. https://github.com/m-labs/artiq/issues/1113#issuecomment-406814932

21:45 <GitHub70> [smoltcp] astro opened pull request #255: Add a few DHCP options (master...dhcp-wire) https://github.com/m-labs/smoltcp/pull/255

21:49 <GitHub108> [smoltcp] whitequark commented on pull request #255 4a95957: Either this function shouldn't be `pub`, or it should perform more validation and ensure that `self` is a CIDR subnet mask. https://github.com/m-labs/smoltcp/pull/255#discussion_r204222580