pnutzh4x0r @ pnutzh4x0r @lemmy.ndlug.org

Posts

245

Comments

167

Joined

2 yr. ago

I wish they had a mastodon account... they have https://mozilla.social, but they don't have an account there... which is bizarre.

They do have an account for Firefox Nightly and Firefox Dev Tools account though.

Oh, just to be clear... I'm not the author of the blog post. I just shared the link :]

Pop already uses zstd for its initramfs. You can check by looking at /etc/initramfs-tools/initramfs.conf.

There was an issue where they were considering switching to xz to save space, but when I tried that, it actually made things slower, so I'd avoid that if possible:

https://github.com/pop-os/default-settings/issues/168

They just pushed out 117.0.1 in PR 277

I think this is the author being humble. jmmv is a long time NetBSD and FreeBSD contributor (tmpfs, ATF, pkg_comp), has worked as a SRE at Google, and has been a developer on projects such as Bazel (build infrastructure). They probably know a thing or two about performance.

Regarding the overall point of the blog, I agree with jmmv. Big O is a measure of efficiency at scale, not a measure of performance.

As someone who teaches Data Structures and Systems Programming courses, I demonstrate this to students early on by showing them multiple solutions to a problem such as how to detect duplicates in a stream of input. After analyzing the time and space complexities of the different solutions, we run it the programs and measure the time. It turns out that the O(nlogn) version using sorting can beat out the O(n) version due to cache locality and how memory actually works.

Big O is a useful tool, but it doesn't directly translate to performance. Understanding how systems work is a lot more useful and important if you really care about optimization and performance.

Yeah, this is what I do... I host a couple of ergo IRC servers and an instance of thelounge for those that want that interface (also offer gamja). Personally, I use weechat to connect to the server.

I think this is part of the recommended (external) nginx configuration for lemmy:

    
limit_req_zone $binary_remote_addr zone={{domain}}_ratelimit:10m rate=1r/s;

  

Which can be found here

I've been using GPaste as my clipboard manager for a while. It can save multiple items in the clipboard and you can switch between which items you want to paste.

It doesn't have a keybinding for the "last second thing", but if you are OK with using your mouse to switch to another item, it would work for what you are asking.

Pretty sad state of affairs :|

Interesting... that has nothing to do with compositing, but rather TLS/SSL (ie. encryption). I'm not sure what would be the cause of that though :|

I'm not sure if it is a known issue, but in general, Pop does not support NVIDIA with Wayland. In fact, that is one of the primary reasons why Pop still uses Xorg by default.

It might be inconvenient, but perhaps try to switch to Xorg whenever you do an upgrade that involves GPU drivers and see if that avoids the freeze. Not an ideal workaround, but it could tell you if the issue is Wayland specific or not.

You can try running gnome-control-center with the verbose (-v) flag to get more debugging information:

    
gnome-control-center -v

  

Do you happen to have NVIDIA? If so, it seems like this is an issue with WebKit and certain GPUs:

https://bugs.webkit.org/show_bug.cgi?id=176852

Pop supports automatic updates since 22.04, but it is not enabled by default. You can check the settings under "OS Upgrades and Recovery" panel in the settings application.

Note that the dry-run flag means don't actually upgrade, just show what you would do, which could be used by another program to see what would happen if an upgrade were to happen. I'm guessing packagekit is using this output to determine what packages are available for upgrading.

Looks like Pop just pushed out linux-firmware 20230815 which contains this revert:

https://github.com/pop-os/linux-firmware/commit/ae80f1f10609ee6b81f96b98b97aecdb52b5ff5f

It causes some systems to fail to resume from S3 and some systems to fail to boot. Our priority should be making things work with kernels that we can actively support, not EOL kernels. Stable kernels, the LTS kernel, and development kernels all work with the latest firmware.

It seems like this should fix your issue.

I'll keep this in mind. Thanks.

My son is gonna be so happy... better not tell him yet though.

Here is a list of all the recent changes from yesterday:

https://github.com/pop-os/repo-release/pull/270

The only thing I can think of is the amd64-microcode update causing issues, though perhaps the linux-firmware could also be a culprit.

Both have updates to AMD CPU and GPUs. Perhaps you can try updating everything but those two components when you have time to mess around.

Can you be more specific about what you mean by it won't boot? Where or how is the computer getting stuck? How far into the boot sequence?

There is a matrix chat: https://matrix.to/#/#lemmy-admin-support-topics:discuss.online

POSTs are how federation works (ActivityPub is a Push-based protocol). When you "subscribe" to a community on say lemmy.ml, you are telling it to periodically send you updates about that community. This comes in the form of POSTS.

As to the frequency of the POSTs, I can imagine something like lemmy.ml having a lot of activity that it needs to inform your instance of (new votes, new comments, new posts, etc)... but I'm not sure if one request per second is reasonable or not.