| persia | Nice job holstein! Great coordination with all the teams. | 00:07 |
|---|---|---|
| holstein | yeah, it really came together there nicely :) | 00:07 |
| ScottL | good job holstein ! thank you much :) | 00:31 |
| astraljava | re: bug 622583, this is very confusing. At times there is talk about including -rt under the support of UKT, and now they're being removed completely. Huh? | 09:01 |
| ubottu | Launchpad bug 622583 in linux-meta-rt (Ubuntu) "Remove the linux-meta-rt packages" [Low,Triaged] https://launchpad.net/bugs/622583 | 09:01 |
| abogani | . | 12:30 |
| ScottL | this may seem like a bad situation but i think this may have an advantage for us | 13:27 |
| ScottL | since we cannot control when the -rt kernel are available (meaning when a new one will come out) | 13:28 |
| ScottL | and there is pressure even from within the studio team to keep the kernel (studio vs desktop) aligned | 13:29 |
| ScottL | and i believe UKT hasn't really been directly support the -rt kernel, it's really been abogani | 13:29 |
| ScottL | it makes sense to remove it from the archives | 13:29 |
| ScottL | BUT that means we can host it in ppa without messing around with UKT | 13:30 |
| ScottL | that means we have *complete* control over and without having to interface with someone else about it :) | 13:30 |
| ScottL | we just have to let users know where it is | 13:30 |
| abogani | https://launchpad.net/~abogani/+archive/ppa | 13:30 |
| ScottL | like with abogani's ppa | 13:31 |
| ScottL | touche abogani | 13:31 |
| abogani | :-) | 13:31 |
| * abogani sorry for have removed -rt from official archives but there isn't a better way to handle *this* situation. | 13:32 | |
| * abogani hopes in you understanding | 13:32 | |
| abogani | *your | 13:35 |
| abogani | My English is getting worse rapidly... :-( | 13:37 |
| abogani | ScottL: I agreed with you expect that I have preferred have -lowlatency kernel into official archives and used as default kernel in Studio. That because it is a PREEMPT (not PREEMPT-RT) kernel but is always better than -generic. | 13:39 |
| * ScottL was taking daughter out to school bus | 13:39 | |
| ScottL | abogani, i absolutely understand removing -rt kernel from the archives and completely support it | 13:39 |
| ScottL | i believe that including it in the first place was probably a misstep given that we can't control when it would be released and the UKT reasonable constraint of aligning kernels between studio and desktop | 13:40 |
| ScottL | furthermore i agree with you abogani that we should probably move towards including the -lowlatency kernel in the archives as our official kernel for both i386 and amd64 | 13:41 |
| ScottL | i've delayed (and quite honest forgotten about) it because work has been sooo busy and intense for the past couple of day, but i hope today to talk to JFo more about that later | 13:42 |
| * ScottL is off this Friday workday :) | 13:43 | |
| abogani | ScottL, :-) | 13:43 |
| abogani | ScottL, FYI: The -lowlatency kernel maintenance is very *trivial* and *everyone* could do it but benefit are interesting. | 13:45 |
| ScottL | abogani, why do you say "interesting" | 14:06 |
| abogani | ScottL: Effort is minimal (because for very low cost of -lowlatency maintenance) but improving respect -generic is really interesting (so with a very good results from a latency/jitter technical view). The -lowlatency is a full preempatble kernel after all! | 14:09 |
| abogani | In short we can improve a lot with a minimal effort. | 14:09 |
| * abogani think that when -lowlatency will be into official archives a lot of people will start to use if. All applications seems more fast and all system is more reactive. | 14:11 | |
| abogani | s/if/it | 14:11 |
| ScottL | hi quadrispro | 14:13 |
| quadrispro | ciao ScottL ! | 14:13 |
| ScottL | abogani, i thought by "interesting" you meant it had some strange or unexpected benefits | 14:19 |
| abogani | ScottL, ... some really good (expected) benefits. | 14:20 |
| ScottL | :) | 14:25 |
| ScottL | hi JFo , you got a minute ? | 14:29 |
| JFo | ScottL, sure | 14:29 |
| ScottL | sorry, for not getting back to you a couple of days ago | 14:30 |
| ScottL | but what i would like to talk about is the possibility of getting the -lowlatency kernel into the archives | 14:30 |
| ScottL | since the -rt kernel has been removed from the archives, this would be a very good kernel to have in a ubuntu studio installation by default | 14:31 |
| JFo | no problem, let me ask the team if that is possible. | 14:32 |
| JFo | I think the consensus was that if there was an -rt kernel, then a -preempt one would be extra work | 14:32 |
| abogani | JFo, I would want let you notice that -lowlatency offers good latency/jitter performance meanwhile it have a very low maintenance costs (and It could be version aligned with -generic one). | 14:35 |
| ScottL | JFo, if the -lowlatency kernel was included in the archives would the UKT support it directly by creating and uploading? | 14:37 |
| ScottL | or would it be community (i.e. studio and abogani ) maintained? if this is the case it would be nice to designate a point of contact for abogani for uploading | 14:38 |
| JFo | ScottL, I think that is a conversation that needs to happen between abogani and the team. The -rt stuff is something that is before my time here. | 14:38 |
| JFo | as all I have been told | 14:38 |
| JFo | is that there would be duplication of effort to have a -rt and a -preempt | 14:39 |
| JFo | but I will ask | 14:39 |
| abogani | JFo, I understand. In any case please take a look of simple configuration differences between -generic and -lowlatency at http://kernel.ubuntu.com/git?p=abogani/ubuntu-lowlatency-maverick.git;a=blob;f=debian.lowlatency/README.Debian;h=102b656fb3f26e30b3094fc91705ee5f3e0ccac8;hb=refs/heads/lowlatency | 14:39 |
| abogani | JFo, Thanks for all. | 14:40 |
| JFo | abogani, no problem :) | 14:40 |
| ScottL | JFo, definite thanks for helping coordinate this :) | 14:40 |
| JFo | ScottL, my pleasure | 14:41 |
| JFo | I just don't know how much help I will be :) | 14:41 |
| JFo | lots of things flying around at the moment :) | 14:41 |
| ScottL | JFo, completely understandable, and this isn't something that needs to be resolved right at this moment | 14:41 |
| JFo | ok | 14:42 |
| ScottL | but i would like to find a well thought out path forward where everyone understand the expectations which functions well for our uses :) | 14:42 |
| JFo | ScottL, maybe it is worthwhile to get you together with some of the team to discuss | 14:43 |
| JFo | I'll see what I can do about that | 14:43 |
| JFo | looks like the people I would ask are off today | 14:43 |
| JFo | so may be Monday before I have anything solid | 14:43 |
| ScottL | JFo, that would be outstanding, i'll be off (US holiday) on monday so that looks very promising :) | 14:47 |
| JFo | same here | 14:47 |
| JFo | but I'll still see what I can do | 14:47 |
| JFo | so let's plan to discuss Tuesday | 14:47 |
| ScottL | JFo, that sound good | 14:49 |
| JFo | excellent | 14:49 |
| * JFo pencils that in :) | 14:49 | |
| astraljava | abogani1: I see your reasoning, and understand that better now. Thanks for the explanation! :) | 15:29 |
| abogani1 | astraljava: :-) | 15:30 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!