You are not logged in.
The crucial part is etc/shadow and whether that still has 'root::::::::'
You could lsinitpio -x the image and look at the file
strangely, etc/shadow in "initramfs-linux-zen.img" has
root::::::::
however, "initramfs-linux.img" had
root:*:::::::
this doesn't make much sense because I removed linux and the image is still there...
pretty sure it boots with linux-zen though.
is it safe to remove "initramfs-linux.img"?
Offline
actually i removed the images from boot, re-installed linux-zen with pacman, and edited the systemd file (line 144), and i'm back in the emergency shell!
it must have been using the "linux" image instead of "linux-zen"
Offline
also, lsblk returns nothing, so the problem must be that it doesen't see the nvme ssd
if i plug in a usb stick, lsblk shows sda, so lsblk works fine...
Last edited by liaminventions (2023-08-28 15:11:44)
Offline
lsmod | grep nvme
lsmod | grep rtsx
Online
both returned nothing at all
Offline
Can you "modprobe nvme" or "insmod nvme"?
Online
modprobe nvme made it appear in lsmod
Last edited by liaminventions (2023-08-28 15:28:39)
Offline
however insmod nvme returned "ERROR: could not load module nvme: No such file or directory"
Offline
insmod requires the full path, doesn't matter - nvme loads so it's present.
Does the nvme show up in "lspci"?
Online
i have to add lspci to the binaries in the initrd - one sec
Offline
Does the nvme show up in "lspci"?
it doesn't look like it..? https://i.imgur.com/3mPO2t9.jpg
Last edited by liaminventions (2023-08-28 16:09:28)
Offline
Loading the module won't extend the pci bus devices.
There's a lot of intel stuff and a mediatek chip, I'd correlate that w/ lspci from the iso.
This would however suggest that the nvme doesn't show up on the bus - are there UEFI/BIOS updates available for the system?
Can you post the system journal from the iso boot (in case the thing produces non-fatal but interesting errors there)?
Can you disable (other PCI) devices in the BIOS, eg. the wifi chip?
Online
Loading the module won't extend the pci bus devices.
There's a lot of intel stuff and a mediatek chip, I'd correlate that w/ lspci from the iso.
Well, here is the lspci in archiso, it has nvme in it as well.
This would however suggest that the nvme doesn't show up on the bus - are there UEFI/BIOS updates available for the system?
Yes, apparently. I have version F.26 and F.31 is available. The computer is a HP 15-dy2024nr btw. I'll update that now.
Can you post the system journal from the iso boot (in case the thing produces non-fatal but interesting errors there)?
Can you disable (other PCI) devices in the BIOS, eg. the wifi chip?
I couldn't find any options in there for that...
Offline
The entire pci bridge is missing
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: Primary bus is hard wired to 0
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: BAR 14: assigned [mem 0x82000000-0x820fffff]
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: BAR 13: no space for [io size 0x1000]
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: BAR 13: failed to assign [io size 0x1000]
Aug 28 20:30:07 archiso kernel: pci 10000:e1:00.0: BAR 0: assigned [mem 0x82000000-0x82003fff 64bit]
Aug 28 20:30:07 archiso kernel: pci 10000:e1:00.0: BAR 4: assigned [mem 0x82004000-0x820040ff 64bit]
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: PCI bridge to [bus e1]
Aug 28 20:30:07 archiso kernel: pci 10000:e0:1d.0: bridge window [mem 0x82000000-0x820fffff]
Aug 28 20:30:07 archiso kernel: pci 10000:e1:00.0: can't override BIOS ASPM; OS doesn't have ASPM control
Aug 28 20:30:07 archiso kernel: pci 10000:e1:00.0: VMD: Default LTR value set by driver
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: can't derive routing for PCI INT A
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: PCI INT A: no GSI
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: PME: Signaling with IRQ 143
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: AER: enabled with IRQ 143
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: DPC: enabled with IRQ 143
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: DPC: error containment capabilities: Int Msg #0, RPExt+ PoisonedTLP+ SwTrigger+ RP PIO Log 4, DL_ActiveErr+
Aug 28 20:30:07 archiso kernel: nvme nvme0: pci function 10000:e1:00.0
Aug 28 20:30:07 archiso kernel: pcieport 10000:e0:1d.0: can't derive routing for PCI INT A
Aug 28 20:30:07 archiso kernel: nvme 10000:e1:00.0: PCI INT A: no GSI
The bus shows up a bit on google, you could try https://bugzilla.kernel.org/show_bug.cgi?id=211879#c32
Online
It seems like my laptop only supports VMD: https://i.imgur.com/smkgPEO.jpg
The bug that I was trying to fix to begin with is that "S0ix state" (it doesn't work, i cant suspend/hybrid-sleep)
Also I just updated my bios.
The disk info: https://i.imgur.com/q19ZsoP.jpg
I gotta be honest though, the bios on this computer is very minimal and seems to be missing some settings compared to a bios on another computer
Last edited by liaminventions (2023-08-28 22:08:19)
Offline
it looks as if this issue has been looked at before https://bbs.archlinux.org/viewtopic.php?id=263567
Offline
Does booting with the fallback initrd have any effect?
Offline
Does booting with the fallback initrd have any effect?
It didn't seem to change anything.
Offline
Is vmd in the initramfs, what if you add it to MODULES?
Online
Is vmd in the initramfs, what if you add it to MODULES?
added it to modules, still didn't boot.
is there any way i could send the journal from the emergency terminal?
Offline
actually i did it, here's the journal from the emergency terminal
it looks like the issue is
Aug 29 13:40:13 archlinux kernel: vmd: probe of 0000:00:0e.0 failed with error -22
Last edited by liaminventions (2023-08-29 14:03:52)
Offline
What if you downgrade the linux package to 6.4.7-arch1-1 which is used on the ISO?
Offline
What if you downgrade the linux package to 6.4.7-arch1-1 which is used on the ISO?
strangely, it didn't seem to fix it. here are the new logs
it must not be the kernel, since i downgraded it to 6.4.7 (as well as the headers)
Offline
What's up w/ "pci=nomsi clocksource=acpi_pm"?
b/c of the "tgl_tc_phy_init" warnings… what if you remove i915 from the initramfs (kms hook and in doubt MODULES entry)?
Online
What's up w/ "pci=nomsi clocksource=acpi_pm"?
b/c of the "tgl_tc_phy_init" warnings… what if you remove i915 from the initramfs (kms hook and in doubt MODULES entry)?
no way, i removed "i915" from the initrd and removed "pci=nomsi clocksource=acpi_pm" from grub, and now it boots!
iirc, "clocksource=acpi_pci" was there because the clock was not working correctly, and I'm not sure about "pci=nomsi".
I should have removed those kernel parameters...
Thanks so much everyone for your help!
Hyprland crashes though, I'll have to fix that.
Last edited by liaminventions (2023-08-29 16:58:59)
Offline