Arch Linux Forums forum

Hi, I have a problem with my USB headset on Linux. It produces a tick/pop/electrical-discharge-like sound approximately once every second whenever audio is playing. I've reproduced it on Arch with the regular, LTS, RT and Zen kernels, as well as on Fedora, CachyOS and Ubuntu LTS, so it does not appear to be specific to my installed Arch setup or kernel. The same headset does **not** produce the periodic tick when used with native Windows on the same hardware and also on another laptop. I also tested Windows in a VirtualBox VM with USB passthrough, and the tick occurs there too, so the Windows test only works when Windows has direct/native access to the USB device. I've already tested this directly through ALSA (`speaker-test -D hw:2,0 -c 2 -r 48000 -F S16_LE -t sine -f 1000`), rather than only through PipeWire, and the behavior remains. I'm trying to determine whether this is a Linux USB/audio driver issue, or something else. MmdrzaArch% uname -a Linux MmdrzaArch 7.2.6-arch2-1 #1 SMP PREEMPT_DYNAMIC Mon, 14 Sep 2026 22:41:30 +0000 x86_64 GNU/Linux MmdrzaArch% lsusb -v -d 0c76:1720 Bus 002 Device 007: ID 0c76:1720 JMTek, LLC. USBFH11 Couldn't open device, some information will be missing Negotiated speed: Full Speed (12Mbps) Device Descriptor: bLength 18 bDescriptorType 1 bcdUSB 1.10 bDeviceClass 0 [unknown] bDeviceSubClass 0 [unknown] bDeviceProtocol 0 bMaxPacketSize0 64 idVendor 0x0c76 JMTek, LLC. idProduct 0x1720 USBFH11 bcdDevice 1.00 iManufacturer 1 Solid State System Co.

pacman.log: [2026-09-22T06:55:03+0200] [PACMAN] Running 'pacman --config=./pacman.conf -Syu linux-lts linux-lts-headers' [2026-09-22T06:55:03+0200] [PACMAN] synchronizing package lists [2026-09-22T06:55:30+0200] [PACMAN] starting full system upgrade [2026-09-22T06:55:59+0200] [ALPM] running '70-dkms-upgrade.hook'... [2026-09-22T06:55:59+0200] [ALPM-SCRIPTLET] ==> dkms remove --no-depmod vboxhost/7.2.18_OSE -k 6.18.53-1-lts [2026-09-22T06:56:03+0200] [ALPM] transaction started [2026-09-22T06:56:04+0200] [ALPM] reinstalled linux-lts (6.18.53-1) [2026-09-22T06:56:07+0200] [ALPM] reinstalled linux-lts-headers (6.18.53-1) [2026-09-22T06:56:07+0200] [ALPM] upgraded fribidi (1.0.16-2 -> 1.0.17-1) [2026-09-22T06:56:07+0200] [ALPM] transaction completed [2026-09-22T06:56:09+0200] [ALPM] running '35-systemd-update.hook'... [2026-09-22T06:56:09+0200] [ALPM] running '60-depmod.hook'... [2026-09-22T06:56:13+0200] [ALPM] running '70-dkms-install.hook'... [2026-09-22T06:56:13+0200] [ALPM-SCRIPTLET] ==> dkms install --no-depmod bcachefs/v1.39.6 -k 6.18.53-1-lts [2026-09-22T06:57:43+0200] [ALPM-SCRIPTLET] [2026-09-22T06:57:43+0200] [ALPM-SCRIPTLET] Error! Bad return status for module build on kernel: 6.18.53-1-lts (x86_64) [2026-09-22T06:57:43+0200] [ALPM-SCRIPTLET] Consult /var/lib/dkms/bcachefs/v1.39.6/build/make.log for more information. [2026-09-22T06:57:43+0200] [ALPM-SCRIPTLET] ==> WARNING: `dkms install --no-depmod bcachefs/v1.39.6 -k 6.18.53-1-lts' exited 10 [2026-09-22T06:57:43+0200]

hey all. recently tried to play Elden Ring: Nightreign and Sea of Thieves. none of these games work. the EAC splash screen displays for a few frames before disappearing. no eac process runs in the background of my system and games will not let me connect. i've tried installing glibc-eac to no avail. set as many launch options as i could think of but i'm stumped. proton logs yield nothing, running steam through a terminal isn't useful either. any and all help appreciated regardless. thanks for the work u guys do

When logging in with a new user in kde, if mariadb is installed, akonadi creates a user database with mariadb. If mariadb is later removed the kde panels hang on login for a minute or so as akonadi tries and fails to open the user database. The workaround I've come up with was to remove the database files. According to a KDE developer akonadi should depend on mariadb (it's currently an optional dependency in arch). Link to the discussion: https://discuss.kde.org/t/when-replacin … -bug/50325

couldnt restart hyprland as switching tty doesnt transfer the hyprland session id stored in env vars pkill/killall ? pgrep? ps aux ?

lib32-pcap has been removed from the official repositories does your system still require it?

ewaller wrote: Look at the output of pacman -Qsq gimp and see if it lists a rational set of packages. If it looks correct, then... sudo pacman -Rs $(pacman -Qsq gimp) For those that stumble upon this. In the case that there are packages in the list you don't want, you can redirect to a temp file, remove that with your editor. Then just use sudo pacman -Rs $(cat /path/to/tmpfile) Not the most elegant, but it's quick and gets it done. Alternatively, you can just eyeball the list once, then pipe within the shell expanded part to sed with a list of lines/package names to blank out. Figure this was probably implicit, but worth stating (was just removing vlc, but not playerctl and needed to look this up).

Hi, I built Textile, an open source desktop intelligence engine and voice companion for Linux. It hooks the Gemini Live API up to Model Context Protocol tool execution, with a small Quickshell face UI so you get some visual feedback while it's working. Most of the desktop AI assistants are either web wrappers or are just extremely slow. I wanted a solution that was fast, native to Linux, and capable of real-time interaction. I chose to follow open closed principle. That's the whole goal behind the plugin system. You write a Yarn which is a class that groups related capabilities and then mark methods as Strands with a @strand decorator. Textile handles the schema validation using Pydantic v2, wires it into both the MCP server and the voice agent, and isolates anything risky in its own subprocess so a bad tool call don't crash the rest of the assistant. Then Weave and any MCP client and execute these strands seamlessly without any further tweaking. The naming "Yarn, Strand, Loom, Warp, Weft, Tapestry..." is a whole textile metaphor, I just wanted a good naming theme so I chose this, I like it so much though. Here a some components of the Textile, with their use: Twill — the MCP server (system stats, process management) Weave — the voice agent, LiveKit + Gemini Live(you can choose any Gemini model that LiveKit supports, default is Gemini 3.8 Live). Loom — dispatches tool calls and enforces the tier/permission model Canvas — the Quickshell UI that shows what the agent's doing, it

Sep 21 20:13:25 PC kernel: ata5: failed to resume link (SControl 0) Sep 21 20:13:25 PC kernel: ata5: SATA link down (SStatus 0 SControl 0) … Sep 21 20:13:29 PC kernel: ata1: link is slow to respond, please be patient (ready=0) Sep 21 20:13:29 PC vivaldi-stable[1620]: [1794:1799:0921/201329.521481:ERROR:chromium/net/socket/ssl_client_socket_impl.cc:962] handshake failed; returned -1, SSL error code 1, net_error -202 Sep 21 20:13:29 PC kernel: nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing. Sep 21 20:13:31 PC kernel: ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) … Sep 21 20:13:37 PC systemd[1]: Failed to start Samba NMB Daemon. Sep 21 20:14:03 PC kernel: r8169 0000:04:00.0 eno1: Link is Down Sep 21 20:14:13 PC kernel: r8169 0000:04:00.0 eno1: Link is Up - 1Gbps/Full - flow control rx/tx Sep 21 20:14:19 PC kernel: r8169 0000:04:00.0 eno1: Link is Down Sep 21 20:14:22 PC kernel: r8169 0000:04:00.0 eno1: Link is Up - 1Gbps/Full - flow control rx/tx Sep 21 20:14:26 PC kernel: r8169 0000:04:00.0 eno1: Link is Down … So some ata device struggles w/ the wakeup as well and your ethernet device starts to flicker - looks more like a wider PCI issue => "pcie_aspm=off"?

42_ftdDEEZY wrote: so does it implement ai? (if so i will not download it) Please do not place your question in code tags. What do you mean by implement AI? What do you consider the AI; the weights, the LLM engine, the agent framework, the prompt, the context?

Yeah I should have done that. I never cleared pacman cache.

I know maintainers are busy and they do the best they can. Let me start off with that before I get flamed. Been using Arch for a decade, I know the ins and outs. At least I hope I do. I use an Intel Arc A750 and B580 on two of my machines. The B580 can get updated through fwupd, which is nice. But the A750 can't. So I need to use igsc to update the firmware on it. Unfortunately, it's still stuck on the 0.9.5 release. v1.0.0 released exactly one year ago today, 2025-09-21. Since then, many more releases have been made and it is at v1.3.2 now. Bumpbuddy is just relabeling the issue again and again for months now. The packages is just being rebuilt with 0.9.5 as the base for new intel-metee releases. Which confuses me even more. lt's already the latest version on the repos. Also, it's the same maintainer for the other two related packages, intel-compute-runtime and intel-metee. They're the latest version and updated regularly, yet igsc has been stuck at 0.9.5 for exactly a year now. Is there a particular. specific reason? Is it blocked by something? Is it forgotten? I reported a bug upstream and it got fixed, but I can still use the old version just fine with no issues. So it doesn't really affect me. But I'm just curious because the only other time I saw an update take this long was with the Python stuff.