@serebit @JoshStrobl It's fine, no need to be sorry.
I was expecting this, it was more of a last ditch effort in case there was anything that could be done in the interim until I have the budget for at least something 2nd hand that would would work properly, even if basically, with OpenCL (in this case, an example would be a Sapphire Radeon HD 7700 Ghz Edition (GCN1.0) that my friend has at his repair shop, which unless someone else will need specifically that card (there are lots of old NVidia cards there that are more suited for his customers, which only need enough graphics power for 1080p (some only need 720p) on YouTube) my friend is keeping for me for when I have the budget (around the equivalent of ~70$). My budget is first aimed towards a new motherboard, however, before my current one breathes it's last breath.

I'm sorry I hadn't realized that I hadn't answered the question.
Currently, my main usecase is cRARk, partly for testing it's reliability, however once I have OpenCL running reliably, I intend to expand that to whatever tasks I have at hand that require OpenCL (lack of a proper environment is one of the main reasons why I cannot start utilizing anything that requires compute (another reason is lack of a budget for a full upgrade, especially with something that is not 2nd hand)).

As mentioned, I have a Haswell CPU, which isn't supported by intel-compute-runtime (with clinfo not even showing any platforms whatsoever with it).

I'm really sorry about my utterly long disappearance. My hands have long since recovered (thank you for your comment @serebit), however I had no access to my desktop (and I still have not gotten the heatsink I was supposed to get together with my laptop's new motherboard (iGPU to iGPU+dGPU), so still no working laptop) until now.

I'm sorry about how late I am returning to this issue, I got my hands injured last week carrying bricks.

My sister's laptop:
Removing TLP had absolutely no effect on performance, still as sluggish as ever.
However I had remembered schedutil, after following it's progress for the past year or two (I forget how long), so I reinstalled TLP and set schedutil, and the performance issues are gone.

Aye, on both. However as of yet, it has only aided performance' which is why it has not crossed my mind.
I'll test without it, and return with results.

Additional testing shows a ~50-66% graphical performance hit on my desktop with 5.10, depending on usecase.
My desktop is simply powerful enough, with me mainly relying on nongraphical performance, that I did not notice the eventual hit there.

In T9379#178728, @Jacalz wrote:

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:
In all other situations, it works just fine.

! In T9379#178637, @JoshStrobl wrote:
Random locations

Thanks, will it be feasible to add an option to reorganize the icons, without allowing them to be placed wherever?

It's not a perfect calculation, it's an estimate. I'm using static values for the items since getting the calculated heights and widths of items added (which is something I did at the beginning) introduced inconsistencies due to how delayed Gtk can be at allocating and the allocation being inconsistent itself.

Ah, then I guess that this will take time, and is low priority.
In any case, it could be argued that this is much better in some ways than the previous implementation (though I do feel it to be a bit lacking, since I make heavy usage of the desktop).

No it isn't. The icon heights are different and your second column first item has a label that wraps onto the second line, and the label height is variable depending on if its one or two lines. The icons center themselves in whatever internal container GtkImage uses for it.

Thanks for the explanation. In my opinion, it is still not as it should, however understand that this is due to GTK limitations.
Any plans for icon size standardization? That may help out.

There are multiple reports of crashes and SIGABRT for gthumb. See their issue tracker.

Thanks, I shall look into it.

@JoshStrobl By "arbitrary", do you mean random locations on the desktop, that are not aligned to a grid, or anywhere on the desktop, but aligned to a grid, or perhaps even reorganizing the icons?

My freezes certainly have not returned, though now I have issues of a different sort (the PC freezing, outright crashing, and then rebooting), which I suspect are down to failing hardware (probably my dGPU, as no matter what I do, including cleaning the card, and replacing the thermal paste, it idles at ~70 degrees celsius, and rises to 80+ under extended load, though it could also be the board, since Asrock's 9-series boards are known to be unstable).

This could be an issue, since from my experience speaking with the Lollypop developer (on completely unrelated bugs), I saw that Lollypop now requires libhandy, which is against policy on Solus, so it will need to be possible to compile Lollypop without libhandy.

Thanks, when I had attempted to test updating it myself, I got an error that I was not the owner (even though all I had done after modifying the package.yml was run "make", so I thought that @JoshStrobl needs to be the one updating it.

In T9252#176857, @hyphens wrote:

maybe try the -lts kernel? that worked for me: https://discuss.getsol.us/d/5518-can-t-run-away-from-freezes-forever/8

In T9252#176731, @JoshStrobl wrote:

Update: I am now seeing these issues also on 5.7.10, though a lot less often than on 5.6.19.

We don't have the 5.7 kernel as our current kernel, so I have no idea what you're talking about. That was reverted clear back in July. Your issue is not related. The first issue by @endumiuz is a CPU soft lock, there isn't really any information on it, we'd need to see if it's resolved with newer kernels when we finally upgrade to that (blocked by NVMe issues). The second issue presented by @endumiuz is, at first glance, related to a bug in Dash to Dock extension for GNOME Shell, so I would suggest they ensure they're using the default settings since they shouldn't be getting any sort if "infinite recursion" issues.

For whatever reason, I can set my laptop monitor's refresh rate to 120Hz when set below the native resolution, even though it's EDID does not list 120Hz as an option.
Could this perhaps be related?

An update on this.

Update: I am now seeing these issues also on 5.7.10, though a lot less often than on 5.6.19.

I can confirm that this is also happening to me, with a slight variation.
For me, it is only the graphics and keyboard input actions that freeze (I can toggle Num/Caps lock), on both my desktop and my laptop (recently upgraded with a new Barracuda Pro).
Additionally, the screen keeps flickering on Intel graphics.

Certainly interesting, I wonder how this stacks up to Clear Sans, especially with RTL languages.

In T8904#171584, @juliocormor wrote:

In T8904#171295, @moriel5 wrote:

@juliocormor With pleasure. I hope your issue will get resolved, though.

I'm home and trying to test your solution. I'm not sure what do you mean with "refresh XRandr's information via radeon-profile". I have radeon-profile installed but i can't find any options to do that?
Also, I think on the distros the second screen worked were using the modesetting driver for the iGPU, while in Solus the system it's using Intel driver. I have read about how to change it but I'm not sure how. If someone helps me change drivers I think the issue is solved.

That is a problem. Noted.

@juliocormor With pleasure. I hope your issue will get resolved, though.

@Darius I'm not ruling out the drivers being out of date, that happens all too often with new AMD hardware, however I said that the ISO may be corrupt, which means that no matter how many times you reflash it, you will have problems, and the only recourse is to delete it, and download it again (that was what I had to do in my case).

Thanks.

Rather late, however I had a similar issue on my desktop (not at home for the time being, and will need to repair the hard drive when I return, so no way to test anything), however by simply refreshing XRandr's information (via radeon-profile), my TV (old, low-end, with bloated capacitors, 42" 1080p non-smart TV that I got from a friend that was throwing it out after replacing it with a 4K smart TV) was instantly picked up, though I had to due this each time I connected or disconnected it to my Radeon HD6950.

I remember when I had issues installing Solus in the past, and I forget who suggested this to me (probably via IRC, since I cannot find this on the Phabricator) (I think it was either Ikey, Josh, or Drake), however perhaps your installation media got corrupted (that's what happened to me)?

This issue can be closed for now, since the current drivers lack support for UVP (more reverse engineering will need to be done with Microsoft's proprietary protocols).

Update: I have managed to get networking to work via the chroot, and update the system, however there is no change unfortunately.

I seem to have gotten myself into this predicament now, however after attempting to fix it via the means outlined in the Solus Help Center (except for updating the system, as it was already mostly up-to-date (just plasma-desktop-branding not being in the index, and I do not have that installed in any case), plus I was unable to get networking to work inside the chroot), I had managed to fix the Systemd errors, however not the fact that the system is unable to switch-root.

Now, ever since I had updated my desktop, the same thing is happening there.

This is weird, but at least things appear to be working.

Thanks, I'll take a look at them.
Update: I do not see any firmware updates for the 9260, unfortunately.

@aquasp Thanks, it appears I also had that installed.
Back when I had initially set up Solus, I installed this to provide additional codec coverage, however looking at the Arch Wiki page, it is pretty much irrelevant unless you are using Nouveau on NVidia's 900-series or newer cards, and even then, it only appears to support decoding (plus, no support by either Firefox or Chromium).
Otherwise, everything it supports seems to be supported by VA-API and/or NVidia's proprietary driver.

Sure thing:

I intend to try maintaining this package.
I'll reopen this issue if I see that I could reasonably do so.

Thanks, I was concerned whether Srain would be added due to some if it's shortcomings (which will probably be fixed), however I see we were of the same mind on this.

Added abi_symbols

@joebonrichie From what I can see from the patch notes, it looks to be pre-Gen4 graphics.

This is weird, I had noticed that lately my laptop's time and date isn't syncing for whatever reason, and I am getting the issue from time-admin brings me the same message, while gnome-control-center simply does nothing at all.

In D8295#133667, @DataDrake wrote:

In D8295#133050, @moriel5 wrote:

That's odd, abi_symbols wasn't added.

It did when I built it locally just now. Did you forget to git add?

@aquasp Yes, I do (both actually, the 32-bit package for Wine support), as well as the others (though these are more for cases where I need to boot from another PC).

Extremely late update, since I had forgotten all about it, and I'm usually not at home.
My fake "7570" has apparently gotten damaged at the same time as my DH55HC motherboard (thankfully it was just the RTC chip (on the DH55HC), and I got a new one of the same model, however I still need to find someone that will be able to help me replace it).
It still works, but not very well, with VGA having limited resolutions (up to something like 1366x768p), and ignoring the actual monitor's settings, and DVI getting a black screen, no idea about HDMI.

@DataDrake Then in that case, I'll check whether this solves the "blank screen on wake" issue that's been plaguing every PC using Intel graphics (everywhere from i3-540 to i5-6200U, including a BayTrail Celeron and an i3-3220) I had installed Solus on since kernel ~4.19 (granted, it happens much less often, and is a minor issue since I just need to switch TTYs, but it still happens,and the less bugs, the better).

@DataDrake Thanks, I see I had mixed the KMS driver with the DDX driver, and the Mesa driver with the KMS driver.
So in short, anything relying on Gen7 or older?

I'm curious, what does X-Org rely upon if you remove the driver, CPU-based graphics?

That's odd, abi_symbols wasn't added.

Updated.
Now development headers are also created.

1. Updating D8295: Update FlashROM to v1.2

Accepted.
At least we can get some ideas for a greeter in the future.

Sure thing, I'd prefer that as well.

Thanks, I did not realize that since I was using the CLI when accidentally sending this differential.

Please ignore this, I had accidentally sent this while reviewing how I need send the differential (to refresh my memory).
A proper differential will be submitted soon.

That is certainly odd.
I'm on the unstable branch, and I had no such issues, not on my 9260, nor on my AX200 (the AX200 does heat up to ~50-55 degrees occasionally, however I expect that to be due to how new it is), perhaps it is due to the fact that I was busy using my computer at the time, and did not have to time to reboot it until the issue was fixed.

It looks like at least some of the modules are licensed under the GPLv2, from what the LICENSE files in the package show, however the legal language makes it confusing as to what is open source, what is open source under an exception, and what is closed-source.

Apparently tss2-esys from tpm2-tss is now required to compile xmlb, one of fwupd's dependencies.

Thanks, that would be great.

Thanks, though I do not think that it is relevant, since my model is neither.
From dmesg, it looks like the kernel thinks it's a sonixj camera for some reason.

Hello everyone.
First, I need to apologize, since I wasn't at home half the time, and the other half I was trying to catch up with other tasks.
Second, I wish to maintain this package, however I'm not sure I'll be able to, since I shall be joining a new learning program soon, and as such, I'll be online even less. So it depends on the release cadence.

Sorry, I was on a family trip and did not have time to do anything.
I just got back ~20 minutes ago.

Sorry, I thought that this was necessary for it be accepted.
Thanks for educating me, this wont happen again from me.

It took some time, due to issues with the package.yml (I guess I cannot use more than one line, so spaces it is), but with the help of some of the veterans (such as @Girtablulu amongst others), I present my package.yml, currently on version 0.4.2 of drawing.

It appears that gcc-ada is required for Intel graphics to work on most boards when compiling CoreBoot.
So I had found when attempting to compile CoreBoot (for the first time) for my grandmother's old computer (apparently it is almost completely supported, with just a few things left which are merely untested), after building for her a new computer (AMD B450-based).

I just checked, and all is well now.

Fixed Markdown to use "-" instead of "*".

Thanks, though I am always on Unstable (on my own PCs), so I just installed it by running sudo eopkg update.
grub-customizer is working perfectly now.

Markdown corrections.