August 2023 - Arch-general - lists.archlinux.org

What laptop's do you guys like?
by Polarian 15 Aug '23

15 Aug '23

Hello, Although this is a little off topic because this is hardware relating to Arch Linux and not Arch Linux itself, I am curious to see what others use. ArchWiki has laptop pages, but that doesn't show what people prefer to use with Arch Linux. So I have a few questions: - What Laptop do you use currently for Arch Linux? - What do you use the Laptop for? - If you could buy any Laptop for Arch Linux right now, what would you buy? I got a feeling there will be a range of responses, some standing behind old Lenovo Thinkpads, and some vouching for modern laptops. One final question, is there any Laptop company you feel will be a good asset to the Linux community as a whole? Or better, Arch Linux directly :P If you respond, thank you for your time :) Have a good night, -- Polarian GPG signature: 0770E5312238C760 Website: https://polarian.dev JID/XMPP: polarian(a)polarian.dev

21 24

Fwd: Something going wrong with Firefox 116.0.2
by Andy Pieters 15 Aug '23

15 Aug '23

On Tue, 15 Aug 2023 at 13:22, pete <petegn(a)mail.com> wrote: > Makes me wonder what desktops asre being used KDE /Plasma here KDE for > years > from the early days mian problems i had were when i had a look at Gnome > the FF > started being unruley runnin 116.0.2 on Arch here fully updated no > issues > > try using KDE .. > > As far as I'm concerned, I haven't used gnome since RHEL4 circa 2004. Qt/KDE/Plasma all the way baby

1 0

Re: Something going wrong with Firefox 116.0.2
by Andy Pieters 15 Aug '23

15 Aug '23

On Mon, 14 Aug 2023 16:01:58 +0100 > Polarian <polarian(a)polarian.dev> wrote: > As a quick note, since the UX update, Firefox has become one of the > heaviest browsers, overtaking chromium/google chrome. This doesn't help [you] and I apologise, but this is literally the reason why I stepped away from Firefox years ago. That and the memory leaks it was constantly suffering from drove me away to Chrome. If Firefox can behave I'd welcome it back in a heartbeat... On Mon, 14 Aug 2023 at 19:10, Andy Pieters <Andy(a)andypieters.me.uk> wrote: > On Mon, 14 Aug 2023 at 18:27, <mirto(a)riseup.net> wrote: > >> On Mon, 14 Aug 2023 16:01:58 +0100 >> Polarian <polarian(a)polarian.dev> wrote: >> As a quick note, since the UX update, Firefox has become one of the > > heaviest browsers, overtaking chromium/google chrome. > > > This doesn't help and I apologise, but this is literally the reason why I > stepped away from firefox years ago. > That and the memory leaks it was constantly suffering from drove me away > to Chrome. > > If firefox can behave I'd welcome it back in a heartbeat... > >

2 1

Xorg 3D acceleration question (Alder lake)
by Polarian 11 Aug '23

11 Aug '23

Hello, So the ArchWiki can be very confusing about this all, and I am sure a lot of people will just install every package and will just hope that it does something, but I would like to actually understand it, so sorry for the ML noise, I would like some clarifications from those who are knowledgeable about this topic. So reason I got a new Laptop with alder lake, and I am trying my best to optimise it, I dropped 70% battery in just 2 hours when its a 73wh battery, that is absurd, so I checked the two following which is recommended to improve performance and battery life: 1. 3D acceleration is in use 2. Hardware video acceleration is in use As for 1, it was by default, however the ArchWiki on the xorg page (https://wiki.archlinux.org/title/Xorg#Driver_installation) lists the following driver to be installed for xorg: xf86-video-intel Now running glxinfo before the installing of said package, direct rendering is enabled (DRI), which would indicate that 3D acceleration is working as needed. So I am a little confused on why it works without having the xorg driver installed? I have the xorg metapackage installed (I am lazy, and the xorg utilities are useful). is xf86-video-intel used to enable DRI3? Is DRI2 possible without said driver? The wiki is really not clear about this, and I know a lot of people probably wonder why it matters if 3D acceleration is already enabled, but I would like to know what is going on :) Also, saying "Just use wayland" is not a valid answer, already had this happen before so just putting it out there. I would appreciate a detailed explanation on what is going on to clear the confusion. Now back to point 2, this was not enabled, VAAPI is the Intel API for hardware video acceleration and after installing the required driver (intel-media-driver) vainfo (from libva-utils) successfully shows support for Hardware video acceleration. If anyone else has any suggestions or experience with optimising alder lake mobile chips on Arch Linux, I would also appreciate the pointers. As a small note, I used AMD (ryzen 5 5500U) with my old laptop, I can not remember if I could be bothered to setup hardware video acceleration or going through optimising the integrated graphics, I ran that Arch Install for 2 years and typical me doesn't document any of my configuration or dotfiles, so I am trying to piece distant memories back together, lucky the ArchWiki is a good refresher :P Thank you for your time, -- Polarian GPG signature: 0770E5312238C760 Website: https://polarian.dev JID/XMPP: polarian(a)polarian.dev

4 9

Is it necessary for CONFIG_PAGE_POOL_STATS to be enabled by default in Arch Linux 2023.08.01?
by sunying＠isrc.iscas.ac.cn 10 Aug '23

10 Aug '23

Hello, I recently noticed that in Arch Linux 2023.08.01, CONFIG_PAGE_POOL_STATS=y. As we know, this option incurs additional CPU and memory cost. In addition to enabling this option, these statistics are only available if the driver using the page pool supports exporting this data. Does enabling this configuration by default have any performance impact? Is it necessary to enable it by default? Thank you for your time. ------------------------------ Institute of Software Chinese Academy of Sciences Intelligent Software Research Center Mail:sunying@isrc.iscas.ac.cn Web: https://isrc.iscas.ac.cn/

2 1

Packaging wireguard re-resolve DNS
by LuKaRo 06 Aug '23

06 Aug '23

Hi everyone, wireguard users need to manually re-resolve the DNS when using endpoints with changing IPs. The wiki page lists an example solution using a systemd timer[0]. I like that solution a lot and use it heavily on multiple machines, and as I prefer to have system files managed by pacman instead of manually creating them, I created an AUR package for that. Today I set up another machine using that package, and I thought, why not push the package to the public, so others can use it as well. Of course, you can as well just copy-paste the example on the wiki page, but I guess there are more people out there who prefer having system files managed by pacman instead of manually creating them. Its easier, after all. However, less than an hour after pushing my package to the AUR and mentioning it in the wiki, the change was reverted and a deletion request was filed with the comment > useless package which literally just copy-pastes the units from the wiki Yes, that's right, it doesn't need to do more than that, because the example on the wiki works great, no need to change that. However, I still think it adheres to the AUR submission guidelines, as I can imagine I'm not the only one who prefers installing an AUR package over copying file contents from the wiki. What do you think? Is this really a useless package? Then go ahead, delete it. But I'm not sure if that's the prevalent opinion, or if there are others like me that like seeing things like that on the AUR, that make your life easier. So at least I wanted to raise the discussion. Would love to hear your input, lukaro [0]: https://wiki.archlinux.org/title/WireGuard#Endpoint_with_changing_IP

5 4

systemd 254 - How will we disable soft-reboot?
by David C. Rankin 04 Aug '23

04 Aug '23

Arch devs, I read that systemd 254 will enable a soft-reboot similar to how windows does fast-boot does it. (The Register: https://go.reg.cx/tdml/dfd67/64f3cc7f/8aaa3ef6/49jh) Mentioned in the article is this soft-reboot will prevent a full reboot - preventing kernel updates in that mode. But the article is silent, and indeed hints at the open question of if or how this will be disabled to allow kernel updates, dual-booting, etc.. Personally, just like on windows, I want to permanently disable it. I don't know if that has been settled yet on Arch, but wanted to know whether we will have the ability to just disable the systemd soft-boot. I have a 12 second cold-start to full desktop -- I have no need for a soft-reboot that will cause issues with dual-booting and kernel updates. Anybody know if, and how we will be able to do it? -- David C. Rankin, J.D.,P.E.

8 10