My JACK installation was broke at some point and I was pretty sure that it was because of https://dev.getsol.us/T9378. I had pretty much given up and accepted that it was broken after all the time spent trying to get it working. Seems strange that the limits.conf file would be the issue, given that I know that I tested that in what feels like 100 times, but if it is, I will happily accept that :)

@TeenCorn The best way of testing it is by opening Ardour. It is going to error and won't be usable if realtime isn't working.

Can this be reconsidered now that we have 4.14 as the LTS kernel?

I see, thanks for looking into the issue. Also, fyi, we probably can't deprecate fuse-exfat as it would mean that users of exfat on the LTS kernels.

I don't think we can deprecate exfat-utils just yet. This one is not yet supported by Gnome disks (you can't partition using exfat).

In D10305#169149, @aleksvor wrote:

@Jacalz, that's because how bugfix releases of bash and readline work. They don't make self-contained tarballs for minor releases - they release patchsets that need to be applied on top of major release tarball.

Just out of curiosity, why are there always multiple patches back-ported from upstream every time bash is updated. Wouldn't it be enough to just use the releases?
Sorry if this is a stupid question, was just wandering as we don't really do so with any other packages unless there are very critical bugs that need fixed.

Use pkgconfig instead

Sorry for editing the policy. It was the only way it would let me unassign myself.

The version looks to be a buggy mess. Freezes and stuff. I am not the maintainer of the package. You can probably unassign me. I don't have the time to look into why it doesn't work.

Sorry, it looks like I have stopped getting notifications for new releases. Will get this updated.

Oh gosh, sorry. I had forgotten that I was the maintainer of this :/

No worries. I hadn’t realised that it was a kernel config and opened the task a bit too late, that’s all :)

No rush. Thanks for adding it. Much appreciated :)

@JoshStrobl I saw that you just added a new config to the kernel. Any chance that you might be able to squeeze this in before the rebuilds? 🙃

@silke Thanks for posting the differential. I have much less time over nowadays, so feel free to keep yourself as the maintainer please :)

Alright. Probably shouldn't have jumped to that conclusion that soon. Looks like it might need to be enabled in the kernel settings first and Gnome Disks does not support these userspace utilities just yet.

Yay. This should now be unblocked. Will have a look at possibly packaging this in the near future :)

Nice. Thank you for doing this. I used to do this myself, but with the limited time I have nowadays, I am very happy to see someone continuing the tradition 🙂

I think all it does is tell LMMS to fail more gracefully if wine isn't installed at runtime. Last time I had a go at it, it was broken and wouldn't build. You have a version that now manages to build. I figured you could test how it behaves with and without wine installed ;)

We used -DWANT_VST_NOWINE=1 before to get a round it, see https://dev.getsol.us/R1974:1d4f6e6654dc05c4f086d4c51d4f05cefe3e3d16.

This is a nice change and thanks for fixing wine build, however I don't think adding wine-32bit as a rundep is a good way to move forward. Wine is a massive piece of code and I don't think every LLMS user wants to download wine when they install LMMS.

Sorry for not getting brave updated for some time now. I've had massive amounts of work to do IRL and haven't gotten any time to do packaging stuff.

Will go ahead and add it to fluidsynth when this lands 🙂

I updated to latest unstable and almost everything is working fine, except for one issue with the budgie application menu (sorry if that's not the correct name).
If I press it just after the desktop has just appeared after a computer restart, it shows up too high up and does not respond to clicks:

@YakoYakoYokuYoku Do you think this could be something you would be willing to work on, given the work you are putting into updating and testing D10026? 🙂

Alright then. Will do so :)

@silke Thank you. Looks like I am remembering incorrectly then. Looking forward to seeing one posted :)

It is going to be very exciting to hopefully see those patches landing @silke. I think I remember seeing a differential revision opened here for integrating it, but I can't seem to find it any more. Am I remembering incorrectly or is it just very hard to find?

Sure thing. Will go ahead later this week when I get some time :)

Sorry. I thought that I already had updated to 3.0. Will get that updated and fixed soon 🙂

This is an issue with JACK sadly. The kernel config CONFIG_RT_GROUP_SCHED breaks realtime jack support. See https://discuss.getsol.us/d/608-jack-audio-refuses-to-start-in-realtime-mode/8 for more information.

Thanks for the kind words. Here are the resulting files for how for I managed to get:

Sorry people. I haven’t had any luck with getting this thing set up for testing due to the poor amount of documentation for doing so. I probably won’t have the time to put in more effort into this due to having a lot going on in life at the moment. Someone else will have to continue this.

This brings up the topic that we probably should drop FeedReader from the repositories.

Alright. Looking into this, it is very clear to me that I have absolutely no idea how to test this. Documentation is scarce or non existing :I

Looks like the pw-jack and pw-pulse got built automatically. Will try do some testing and hopefully get a revision out in a few days.

@ermo Perhaps. I guess I can give it a try :)

It was turned off when we had old JACK, but should now be feasible now that we have JACK2 :)

Sorry @JoshStrobl. I can't use the release tarballs on GitHub because the checks are failing for some reason when using those. Would you mind re-reviewing?

Build with opus (thanks @aleksvor)

Thinking about this, I think that I would rather update when the next release is out. A lot has happened since the latest release and maintaining multiple patches is a bit too much work.

• Update to 0.20

Same here. Haven’t seen this for a long time.

Thanks for the hard work. Greatly appreciated :)

Oh gosh, sorry @serebit. I did look though the changes, but it looks like I missed that one. Will rely om ctrl + f in the future :I

@joebonrichie Mentioned earlier in the thread that it might be worth adding --enable-fma`along with AVX2. Any luck with that?

Sure @JoshStrobl. I will give it a go now that upstream work on better building of shared libraries has landed. Might need to be updated to master, but I will keep you posted with the results :)

Thanks for your hard work with testing @serebit. Looks really promising 🙂
We’re you testing with -ffast-math and the other flags as well?

From the release announcement, it looks like the release relies on stuff in the 0.20 release of libraw. Would you mind updating that at the same time?