ChanServ changed the topic of #glasgow to: glasgow debug tool · code https://github.com/GlasgowEmbedded/Glasgow · forum https://glasgow.whitequark.org · logs https://freenode.irclog.whitequark.org/glasgow
_whitelogger has joined #glasgow
robbi5 has quit [Quit: Ping timeout (120 seconds)]
robbi5 has joined #glasgow
_whitelogger has joined #glasgow
<whitequark> btw i added icecube support to nmigen
<whitequark> of interest to some people... tnt perhaps?
_whitelogger has joined #glasgow
gregdavill has quit [Read error: Connection reset by peer]
gregdavill has joined #glasgow
_whitelogger has joined #glasgow
gregdavill_ has joined #glasgow
gregdavill has quit [Ping timeout: 265 seconds]
gregdavill_ has quit [Remote host closed the connection]
gregdavill_ has joined #glasgow
gregdavill_ has quit [Remote host closed the connection]
gregdavill_ has joined #glasgow
craigo has quit [Ping timeout: 252 seconds]
_whitelogger has joined #glasgow
_whitelogger has joined #glasgow
gregdavill has joined #glasgow
gregdavill_ has quit [Ping timeout: 250 seconds]
gregdavill_ has joined #glasgow
gregdavill has quit [Ping timeout: 250 seconds]
gregdavill_ has quit [Read error: No route to host]
gregdavill has joined #glasgow
gregdavill_ has joined #glasgow
gregdavill has quit [Ping timeout: 245 seconds]
gregdavill_ has quit [Read error: Connection reset by peer]
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark pushed 2 commits to master [+0/-0/±2] https://git.io/JesZ8
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark 22e8bc9 - Travis: build Yosys and nextpnr.
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark f83b952 - target.hardware: force use of abc9 and HeAP.
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark pushed 2 commits to master [+0/-0/±2] https://git.io/JesZz
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark 494e8ef - Travis: enable abc build.
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark 595837c - Travis: ignore deprecation warnings.
<_whitenotifier> [Glasgow] Error. The Travis CI build could not complete due to an error - https://travis-ci.org/GlasgowEmbedded/Glasgow/builds/588078412?utm_source=github_status&utm_medium=notification
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/JesZH
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark d2ec001 - Travis: cache the Yosys/abc/nextpnr artifacts.
<_whitenotifier> [Glasgow] Error. The Travis CI build could not complete due to an error - https://travis-ci.org/GlasgowEmbedded/Glasgow/builds/588080544?utm_source=github_status&utm_medium=notification
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/JesnU
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark fdd5a26 - Travis: set nextpnr architecture.
<_whitenotifier> [Glasgow] Error. The Travis CI build could not complete due to an error - https://travis-ci.org/GlasgowEmbedded/Glasgow/builds/588086682?utm_source=github_status&utm_medium=notification
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/Jesn0
<_whitenotifier> [GlasgowEmbedded/Glasgow] whitequark bec36fa - target.hardware: force use of abc9 and HeAP.
pie_ has quit [Read error: Connection reset by peer]
pie_ has joined #glasgow
yorick has joined #glasgow
_whitelogger has joined #glasgow
craigo has joined #glasgow
zkms has joined #glasgow
<electronic_eel> is there some policy about tool requirements for Glasgow patches?
<electronic_eel> I was showing my Glasgow to a friend this afternoon. Installation went smooth and all, just the -abc9 option for yosys didn't work
<electronic_eel> he used Fedora which has yosys 0.9 packaged. But -abc9 seems to be just in git, not in the 0.9 release
<electronic_eel> It took me a few minutes to realize what the problem was and revert the patch. But maybe someone new might have a problem with something like that
<emily> oh yeah
<emily> it should gate that on yosys version
<emily> whitequark:
<emily> 0.9 shipped without abc9 :(
<emily> electronic_eel: fwiw, having 0.9 packaged is pretty lucky, it only came out recently
<emily> so for many distros there is no packaged yosys that works with nmigen/glasgow
<electronic_eel> seems like there are active openfpga devs using fedora, it was packaged a few days after release. nextpnr is also packaged. very convenient
<emily> that is nice, don't you need rawhide for that though?
<emily> since stable releases only filter down every 6 months?
<electronic_eel> no, regular released f30
<emily> lucky timing I guess
<electronic_eel> it depends a bit on the application: some apps that are deemed important infrastructure and that have major updates only get them in a new fedora release
<emily> I see, so Fedora is semi-rolling-release for some packages, that's nice
<electronic_eel> smaller programs and smaller updates are also done on the current release
<emily> Debian refuses to do it for anything it doesn't has to (browsers, basically), I think
<electronic_eel> hmm, but debian releases are like every two or three years. No updates whatsoever in between, all bugfixes and security patches backported?
<electronic_eel> for a workstation that seems a bit unpractical for me
<electronic_eel> for a server this is necessary. I use centos on the servers, that has even longer support
<emily> it works a bit badly, yes
<emily> I don't think it's necessary or even necessarily desirable for servers either, personally
<emily> OpenSUSE and NixOS have rolling releases that are gated on test/QA suites passing; in combination with reproducible builds and referenceable repository snapshots this gives you rolling releases with a stability competing with that of traditional slow-cycle distros
<emily> especially in a world where so many servers are just running other versions of another distro in Docker containers because they need some newer dependency anyway :/
<electronic_eel> it depends on how many different servers / services you have running. Even a smaller company has hundreds of services running today. If you constantly have issues with stuff breaking because of rolling releases this is not acceptable
<electronic_eel> often stuff breaks in subtle manner and it takes a lot of time to figure it out and nail it to some rolling update
<emily> that's where the comprehensive test suites come in
<emily> especially as scale goes larger you have to start operating in this way, because manually fighting and tracking down each individual fire isn't possible
<electronic_eel> even if you develop this level of testing for the stuff you wrote yourself, most other software you get doesn't have it
<electronic_eel> in the open source projects I use, there often is some kind of test suite, but sometimes it is difficult to set up so it is only feasible for the developers themselves
<electronic_eel> so for the mission critical stuff I have good test suites, but not for the dozens of other programs and services that run
<electronic_eel> but even the not so important stuff can suck away a lot of time
<electronic_eel> so I prefer the very long support time for centos and happliy invest a few minutes to package up newer stuff if I need it
<_whitenotifier> [Glasgow] electroniceel opened pull request #158: cli: make output legible on black-on-white terminals too - https://git.io/JesRa
electronic_eel_ has joined #glasgow
electronic_eel has quit [Ping timeout: 268 seconds]
electronic_eel_ is now known as electronic_eel
gsuberland has quit [Remote host closed the connection]
electronic_eel has quit [Ping timeout: 240 seconds]
electronic_eel has joined #glasgow
electronic_eel has quit [Ping timeout: 245 seconds]
electronic_eel has joined #glasgow