the $() bit is basically interpolating in the value of the whoami command, which just spits our your username
also, has anyone had any issues with video playback occasionally freezing and stuttering (getting stuck on one frame for a few seconds) while the audio keeps playing? seems to happen with youtube for me occassionally
Awesome! This works for me as well on Bluefin. Thanks for finding this!!
1 Like
I’ll still see it not come up immediately on boot regardless of whether or not the GPU is installed. I’m just suspecting that the wlan0 interface renaming to wlp5s0 or wlp2s0 is impacted by the hardware configuration, and isn’t really a part of my issues not seeing the interface come up immediately on boot. I’m not an expert at reading those logs though.
I wonder why it’s getting renamed; mine has always shown wlan0.
I’m running Manjaro, not just vanilla Arch, so I’m suspecting that’s part of it, but I had the same behavior on the Fedora live installer (though the KDE Spin).
Is it practical to put a mechanical switch to disconnect the power to dGPU like the switches to mic and webcam?
1 Like
Ah, so there is something in my logs, something with FLR.
3/9/24 8:41 PM kernel mt7921e 0000:05:00.0: not ready 1023ms after FLR; waiting
3/9/24 8:41 PM kernel mt7921e 0000:05:00.0: not ready 2047ms after FLR; waiting
3/9/24 8:42 PM kernel mt7921e 0000:05:00.0: not ready 4095ms after FLR; waiting
3/9/24 8:42 PM kernel mt7921e 0000:05:00.0: not ready 8191ms after FLR; waiting
3/9/24 8:42 PM kernel mt7921e 0000:05:00.0: not ready 16383ms after FLR; waiting
3/9/24 8:42 PM kernel mt7921e 0000:05:00.0: not ready 32767ms after FLR; waiting
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0: not ready 65535ms after FLR; giving up
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0: enabling device (0000 -> 0002)
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0: ASIC revision: 79220010
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0: HW/SW Version: 0x8a108a10, Build Time: 20231120183400a
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0: WM Firmware Version: ____000000, Build Time: 20231120183441
3/9/24 8:43 PM kernel mt7921e 0000:05:00.0 wlp5s0: renamed from wlan0
Anyone got any ideas of how I can skip the 65s of waiting?
I don’t think that would be all that practical. If you shut off the power to the GPU while the system is on, I would think it would crash the system because PCIe is not hot-swappable. If the GPU were connected over USB 4.0/thunderbolt, it wouldn’t have the same issue because the driver is designed for hot swapping.
Ah, OK. Just I have been exposed to using $user to assign dialout rights through other places, but I’m not an experienced *nix user.
Has anyone considered creating a AUR packages to include various tweaks and dependencies for FW laptops? Like FW13-Intel, FW13-amd and FW16-AMD including udev rules, audio EQ tweaks, ryzen-PPD, etc?
It might be a stretch, but there may be a use case for this. If you’re in extremely bright outdoor light (like the sun is shining directly on the screen), turning off the backlight might actually make the screen more readable. I mean it will look terrible, but the contrast may improve without backlighting when front lighting is so intense. So this could be an argument for why the screen should remain enabled even with backlight at zero.
But if the backlight is bright enough to overcome even the brightest sunlight, then my theory is moot.
Anyone got an easyeffects profile? The one for the 13 gave me a bunch of errors and it still sounds bad.
Oops I was in windows audio mode after a bios reset. Still not great though.
I believe I tested this once once, it didn’t work, unfortunately.
This is correct. wl (wireless lan), p5/p2 (pci bus location), s0 (probably slot 0). Makes sense the pci bus location might change with/without gpu installed. With my gpu installed I have wlp4s0 though.
Yes, mostly when using it heavily. Downloading a steam game for example. Interestingly, I had my Steam Deck right next to it last night when the wireless cut out…and the steam deck ALSO lost internet connection, but still connected to wifi. I’m wondering if it is something on my UDM Pro (router) or with my AC Pros (Access points, I have 2). Haven’t had the time to really investigate yet. But this never happened before, until the Framework was added to the mix.
Looking at lspci output with dGPU and only one NVMe SSD:
01:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 XL Upstream Port of PCI Express Switch (rev 12)
02:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 XL Downstream Port of PCI Express Switch (rev 12)
03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 33 [Radeon RX 7700S/7600/7600S/7600M XT/PRO W7600] (rev c1)
03:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Navi 31 HDMI/DP Audio
04:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter
05:00.0 Non-Volatile memory controller: Sandisk Corp WD Black SN770 / PC SN740 256GB / PC SN560 (DRAM-less) NVMe SSD (rev 01)
and my interface is also wlp4s0.
I guess without dGPU and with two NVMe SSDs the Wifi card could end up on address 05:00.0 and thus being renamed to wlp5s0 ?
I also have a Unifi setup (using UDM Pro, USW-Aggregation, USW 16-PoE, U6 Lite AP) and also didn’t have time to investigate further yet.
But I didn’t notice dropouts, only kinda low throughput when Steam loads from another PC in the same network. It only reached 11n speeds ~20-35 MiB/s (160-300 mbps) although the network path is capable of >=120 MiB/s (1 Gbps, with some 10gig links).
Perhaps the phenomena is related to this recent Kernel addition: AMD WBRF Ready For Linux 6.8 To Mitigate WiFi Radio Interference - Phoronix
Note: Linux 6.8 came out this week and isn’t available (pre-packaged) on Arch yet.