09:33 <sb0> whitequark: should warnings like "DeprecationWarning: instead of `self.submodules +=`, use `m.submodules +=`" be turned on when using nmigen compatibility mode?

13:43 <whitequark> sb: they're on by default

14:42 <_whitenotifier-3> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±8] https://git.io/fjghd

14:43 <_whitenotifier-3> [m-labs/nmigen] whitequark efb2d77 - build.{dsl,res,plat}: add PinsN and DiffPairsN.

14:46 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/544778342?utm_source=github_status&utm_medium=notification

14:46 <_whitenotifier-3> [nmigen] Success. 80.74% (+0.03%) compared to ad1a40c - https://codecov.io/gh/m-labs/nmigen/commit/efb2d773c3fa34f7805c034a828bdfe23fcf71f9

14:46 <_whitenotifier-3> [nmigen] Failure. 62.79% of diff hit (target 80.7%) - https://codecov.io/gh/m-labs/nmigen/commit/efb2d773c3fa34f7805c034a828bdfe23fcf71f9

14:53 <_whitenotifier-3> [nmigen] jfng synchronize pull request #99: vendor.xilinx_7series: implement DDR I/O buffers. - https://git.io/fjgXa

14:53 <_whitenotifier-3> [nmigen] Failure. 80.7% (-0.04%) compared to efb2d77 - https://codecov.io/gh/m-labs/nmigen/compare/efb2d773c3fa34f7805c034a828bdfe23fcf71f9...2dcf51638e3de2ba3f6d40f3d73299bd997a761a

14:53 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing efb2d77...2dcf516 - https://codecov.io/gh/m-labs/nmigen/compare/efb2d773c3fa34f7805c034a828bdfe23fcf71f9...2dcf51638e3de2ba3f6d40f3d73299bd997a761a

14:57 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/544783325?utm_source=github_status&utm_medium=notification

14:57 <_whitenotifier-3> [nmigen] Success. 80.74% remains the same compared to efb2d77 - https://codecov.io/gh/m-labs/nmigen/compare/efb2d773c3fa34f7805c034a828bdfe23fcf71f9...67bb0007445847809b8efa93695acc4f0dd750fe

14:57 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing efb2d77...67bb000 - https://codecov.io/gh/m-labs/nmigen/compare/efb2d773c3fa34f7805c034a828bdfe23fcf71f9...67bb0007445847809b8efa93695acc4f0dd750fe

15:06 <_whitenotifier-3> [m-labs/nmigen-boards] whitequark pushed 1 commit to master [+0/-0/±5] https://git.io/fjgjz

15:06 <_whitenotifier-3> [m-labs/nmigen-boards] whitequark 755ae93 - Use the new active-low PinsN platform feature.

15:47 <sb0> whitequark: but I don't see the point of those warnings. when you're in compatibility mode, you should be writing old-style migen code, no?

15:48 <sb0> is this something to teach the new syntax to existing migen users?

15:52 <key2> sb0: yeah

15:52 <key2> sb0: I had this convo with him, and it is

16:11 <whitequark> with who?

16:44 <_whitenotifier-3> [nmigen] sbourdeauducq opened issue #100: nmigen.compat Case.makedefault() broken - https://git.io/fj2vQ

16:47 <_whitenotifier-3> [nmigen] sbourdeauducq commented on issue #100: nmigen.compat Case.makedefault() broken - https://git.io/fj2vb

17:38 <_whitenotifier-3> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/fj2fd

17:38 <_whitenotifier-3> [m-labs/nmigen] whitequark d3ed390 - vendor.lattice_ice40: fix typo.

17:40 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/544854513?utm_source=github_status&utm_medium=notification

17:40 <_whitenotifier-3> [nmigen] Success. 80.74% remains the same compared to efb2d77 - https://codecov.io/gh/m-labs/nmigen/commit/d3ed390b9d087ea63ad03901aaf3cb1d51eafa77

17:40 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing efb2d77...d3ed390 - https://codecov.io/gh/m-labs/nmigen/commit/d3ed390b9d087ea63ad03901aaf3cb1d51eafa77

19:24 <whitequark> sb0: the idea is that you can convert old-style code to new-style code piece by piece, perhaps even in the same file

19:24 <whitequark> so the warnings show you which places are to be converted

19:25 <whitequark> now, the thing is, i wrote nmigen, so i'm by definition unable to see how easy or not it is to convert code from omigen to nmigen

19:25 <whitequark> so, of course, i'm open to feedback on that front

19:25 <whitequark> maybe we should get rid of the warnings or something

19:55 <_whitenotifier-3> [nmigen] whitequark closed pull request #99: vendor.xilinx_7series: implement DDR I/O buffers. - https://git.io/fjgXa

19:55 <_whitenotifier-3> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/fj2Tm

19:55 <_whitenotifier-3> [m-labs/nmigen] jfng 3b303c3 - vendor.xilinx_7series: implement DDR I/O buffers.

19:56 <_whitenotifier-3> [nmigen] whitequark commented on pull request #40: WIP: Expand and document lib.cdc - https://git.io/fj2TO

19:57 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/544909570?utm_source=github_status&utm_medium=notification

19:57 <_whitenotifier-3> [nmigen] Success. 80.74% remains the same compared to d3ed390 - https://codecov.io/gh/m-labs/nmigen/commit/3b303c333445d1878c08a07ad174d10be4f0cba5

19:57 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing d3ed390...3b303c3 - https://codecov.io/gh/m-labs/nmigen/commit/3b303c333445d1878c08a07ad174d10be4f0cba5

20:02 <cr1901_modern> PinsN is a good idea. Though AIUI, the main reason icebreaker has active low and hi is b/c esden wanted the board to be used in an educational setting. So, students learning nmigen have that lesson abstracted away lol

20:36 <whitequark> hah

22:29 <_whitenotifier-3> [m-labs/nmigen] whitequark pushed 2 commits to master [+0/-0/±3] https://git.io/fj2I0

22:29 <_whitenotifier-3> [m-labs/nmigen] whitequark e52b15d - hdl.ast: add name_suffix=".." option to Signal.like().

22:29 <_whitenotifier-3> [m-labs/nmigen] whitequark 6beba3a - Simplify code by using Signal.like(name_suffix="..") appropriately.

22:31 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/544974637?utm_source=github_status&utm_medium=notification

22:31 <_whitenotifier-3> [nmigen] Success. 80.75% (+0.01%) compared to d3ed390 - https://codecov.io/gh/m-labs/nmigen/commit/6beba3a48b71e191930bd2aeed453c2e8047bb48

22:31 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing d3ed390...6beba3a - https://codecov.io/gh/m-labs/nmigen/commit/6beba3a48b71e191930bd2aeed453c2e8047bb48

22:51 <_whitenotifier-3> [nmigen] whitequark commented on issue #100: nmigen.compat Case.makedefault() broken - https://git.io/fj2IK

23:37 <sb0> whitequark: the contents of the warnings seems ok, but I think the main use case for the compat mode is deal with a large migen codebase that will simply never get converted; and for that, the warning shouldn't be there. so they should not be turned on by default.

23:40 <whitequark> sb0: should there be migen codebases that never get converted?

23:41 <whitequark> that sounds like a failure on the part of nmigen

23:46 <TD-Linux> I like the warnings. I don't see a use case for dual compat migen/nmigen codebases

23:47 <TD-Linux> well I guess one case might be to pull e.g. litedram into your nmigen design

23:48 <TD-Linux> (which heavyx already does)