#m-labs on 2019-01-01 — irc logs at freenode.irclog.whitequark.org

03:08 <_whitenotifier-6> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±2] https://git.io/fhtg7

03:08 <_whitenotifier-6> [m-labs/nmigen] whitequark d78e6c1 - hdl.mem: add DummyPort, for testing and verification.

03:17 <_whitenotifier-6> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/474004333?utm_source=github_status&utm_medium=notification

03:17 <_whitenotifier-6> [nmigen] Success. 81.45% (+0.08%) compared to ae3c583 - https://codecov.io/gh/m-labs/nmigen/commit/d78e6c155b6de72ab56f7cb123bc2b7cdb8f6b04

03:17 <_whitenotifier-6> [nmigen] Success. 100% of diff hit (target 81.36%) - https://codecov.io/gh/m-labs/nmigen/commit/d78e6c155b6de72ab56f7cb123bc2b7cdb8f6b04

03:20 <whitequark> rjo: ping

03:21 <key2> ping timeout

03:26 <_whitenotifier-6> [nmigen] whitequark opened issue #16: Support transparent ports with RE - https://git.io/fht2e

03:35 <_whitenotifier-6> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±2] https://git.io/fht2L

03:35 <_whitenotifier-6> [m-labs/nmigen] whitequark 031a9e2 - hdl.rec: use a helpful error on unknown field reference.

03:37 <_whitenotifier-6> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/474007256?utm_source=github_status&utm_medium=notification

03:37 <_whitenotifier-6> [nmigen] Success. 81.46% (+0.01%) compared to d78e6c1 - https://codecov.io/gh/m-labs/nmigen/commit/031a9e26160748bc426e6d2a0b687603e1345bff

03:37 <_whitenotifier-6> [nmigen] Success. 100% of diff hit (target 81.45%) - https://codecov.io/gh/m-labs/nmigen/commit/031a9e26160748bc426e6d2a0b687603e1345bff

03:39 <_whitenotifier-6> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±2] https://git.io/fht2s

03:39 <_whitenotifier-6> [m-labs/nmigen] whitequark 3c07d8d - hdl.rec: include record name in error message.

03:41 <_whitenotifier-6> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/474007631?utm_source=github_status&utm_medium=notification

03:41 <_whitenotifier-6> [nmigen] Success. 81.48% (+0.01%) compared to 031a9e2 - https://codecov.io/gh/m-labs/nmigen/commit/3c07d8d52cef639dad501e86522db804796e0d6e

03:41 <_whitenotifier-6> [nmigen] Success. 100% of diff hit (target 81.46%) - https://codecov.io/gh/m-labs/nmigen/commit/3c07d8d52cef639dad501e86522db804796e0d6e

03:48 <_whitenotifier-6> [nmigen] The6P4C opened issue #17: Add a more user-friendly error message when the yosys binary can't be found - https://git.io/fht2R

04:01 <sb0> whitequark, rjo: OK to wipe the HDD on build.lab.m-labs.hk?

04:02 <whitequark> sb0: the old buildserver?

04:24 <sb0> whitequark: the new computer that was supposed to be used as a faster build machine, but which you never finished setting up

04:25 <whitequark> why do you want to wipe the HDD?

04:25 <whitequark> well, SSD

04:25 <whitequark> it doesn't have a HDD

04:26 <whitequark> you should at least back up /etc, because it has useful kernel and xen parameters

04:27 <whitequark> everything else is disposable i think

04:27 <sb0> whitequark: to use as an experimental nix build machine. if you have a better use for this computer then please do it. right now it's only collecting dust.

04:27 <whitequark> i'm pretty sure i configured it as a buildslave

04:28 <whitequark> if not, sure, i can do that

04:28 <whitequark> though you'll need to turn it on

04:28 <sb0> whitequark: can't we just replace the old machine completely?

04:28 <whitequark> sure

04:29 <sb0> iirc the buildslave didn't work

04:29 <whitequark> hm

04:29 <whitequark> i'm fairly sure i verified it, but if it didn't work, i can fix it

04:29 <whitequark> and/or migrate the old machine completely, yes

04:29 <whitequark> turn it on and ping me.

04:29 <sb0> there was just this artiq-quick builder, which was never finished, and CI didn't use it

04:30 <whitequark> oh, i remember that

04:30 <whitequark> but it's unrelated to the build machines

04:30 <whitequark> artiq-quick was not "quick" in the sense of fast

04:30 <whitequark> it is supposed to build people's pull requests when poked on IRC

04:30 <whitequark> i have provided instructions for this and iirc someone (hartytp?) used it

04:31 <whitequark> successfully, that is

04:33 <sb0> whitequark: it's on

04:34 <whitequark> ok, i'll take a look soon

04:34 <whitequark> do you have anything running on lab. that needs migrating?

04:35 <whitequark> oh. i remember now what was the problem with migrating the old machine.

04:35 <whitequark> the new one did not run xen, or rather, xen killed overclocking.

04:35 <whitequark> so i'll need to migrate the VMs to use kvm.

04:36 <whitequark> i think i started looking into kvm, but the tooling around it was kind of crap, so it never went anywhere.

04:36 <whitequark> but i can just run qemu -kvm instead of all the xml junk they added around

04:39 <sb0> whitequark: it's not stable with overclocking anyway

04:39 <sb0> I just disabled it

04:39 <sb0> running a few vivado instances concurrently systematically crashed the whole machine

04:42 <whitequark> sb0: why disable and not bump it down by 0.1 GHz?

04:48 <sb0> is there really a significant performance gain from this overclocking?

04:48 <sb0> and it conflicts with the VMs as you said

04:49 <whitequark> yes, actually, there is

04:50 <whitequark> i've spent a few days measuring it and it's nontrivial

04:50 <whitequark> iirc it was set up at 5 GHz when i put it in the lab

04:50 <whitequark> and 4.9 is about as far as it should go according to intel

04:50 <whitequark> so you only really need to knock it down one step

04:50 <whitequark> 5 GHz was stable in my experiments with one vivado process but i guess two pushed it enough

04:51 <whitequark> as for xen, there's no real difference between xen and kvm for our purposes. actually kvm might be slightly nicer.

04:52 <whitequark> it's mostly that i'm not used to libvirt but know xl

04:54 <whitequark> another problem with xen is that iirc it not only inhibits overclocking but it also inhibits turboboost

04:54 <whitequark> so if we put xen there it'll be even slower than it is right now

05:45 <GitHub168> [smoltcp] jhwgh1968 commented on issue #266: I reviewed my code, and I think the case I was thinking of is already covered.... https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711078

05:48 <GitHub43> [smoltcp] jhwgh1968 commented on issue #256: Since it didn't get linked to this issue, I have a phase 1 implementation in #266.... https://github.com/m-labs/smoltcp/issues/256#issuecomment-450711148

05:48 <GitHub43> [smoltcp] whitequark commented on pull request #266 2552d0b: This seems like a weird optimization, but since the RFC spells it out explicitly, I guess it's fine. https://github.com/m-labs/smoltcp/pull/266#discussion_r244628623

05:49 <GitHub31> [smoltcp] whitequark commented on issue #266: @m-labs-homu r+ https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711180

05:51 <GitHub40> [smoltcp] m-labs-homu commented on issue #266: :pushpin: Commit 2552d0b has been approved by `whitequark`

05:51 <GitHub32> [smoltcp] whitequark commented on issue #266: @m-labs-homu r+ https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711229

05:51 <GitHub67> [smoltcp] whitequark commented on issue #266: @m-labs-homu r+ https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711180

05:51 <GitHub188> [smoltcp] m-labs-homu commented on issue #266: :bulb: This pull request was already approved, no need to approve it again. https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711231

05:51 <GitHub182> [smoltcp] m-labs-homu commented on issue #266: :pushpin: Commit 2552d0b has been approved by `whitequark`

05:51 <GitHub139> [smoltcp] m-labs-homu pushed 1 new commit to auto: https://github.com/m-labs/smoltcp/commit/ed8dce015c45ee6e37c8d671f290a6d9b7fb49ac

05:51 <GitHub139> smoltcp/auto ed8dce0 jhwgh1968: Minimal Implementation of TCP Selective Acknowledgement...

05:51 <GitHub29> [smoltcp] m-labs-homu commented on issue #266: :hourglass: Testing commit 2552d0b414dc88d23db32150f02d84b807fdbe85 with merge ed8dce015c45ee6e37c8d671f290a6d9b7fb49ac... https://github.com/m-labs/smoltcp/pull/266#issuecomment-450711239

06:02 <travis-ci> m-labs/smoltcp#1191 (auto - ed8dce0 : jhwgh1968): The build passed.

06:02 <travis-ci> Change view : https://github.com/m-labs/smoltcp/compare/ef014b118b19...ed8dce015c45

06:02 <travis-ci> Build details : https://travis-ci.org/m-labs/smoltcp/builds/474022451

06:02 <GitHub2> [smoltcp] m-labs-homu merged auto into master: https://github.com/m-labs/smoltcp/compare/556672f79811...ed8dce015c45

06:02 <GitHub140> [smoltcp] m-labs-homu commented on issue #266: :sunny: Test successful - [status-travis](https://travis-ci.org/m-labs/smoltcp/builds/474022451?utm_source=github_status&utm_medium=notification)

06:02 <GitHub131> [smoltcp] m-labs-homu closed pull request #266: Minimal Implementation of TCP Selective Acknowledgement (master...selective-ack-phase-1) https://github.com/m-labs/smoltcp/pull/266

15:41 <GitHub-m-labs> [artiq] sbourdeauducq pushed 3 new commits to master: https://github.com/m-labs/artiq/compare/421ad9c9168e...48793b7ecf70

15:41 <GitHub-m-labs> artiq/master 1e7ba32 Sebastien Bourdeauducq: nix: add development environment

15:41 <GitHub-m-labs> artiq/master e279980 Sebastien Bourdeauducq: nix: do not install development packages in user environment

15:41 <GitHub-m-labs> artiq/master 48793b7 Sebastien Bourdeauducq: nix: reorganize .nix files