Page MenuHomeSolus

Linux LTS 5.15 Issues
Open, HighPublic

Description

This is a tracking issue for all issues users on the LTS kernel have after upgrading to the 5.15 kernel. We will try to resolve as many issues as possible but cannot guarantee that all will be solved.

Event Timeline

Plume added a subscriber: Plume.Tue, Nov 8, 2:05 PM

As I mentionned elsewhere in the forum, my machine have graphical issues with kernel 5.x. Every now end then, the screen turns black when the mouse cursor is in the lower part of the screen and remains black until the mouse is moved again. This trouble doesn't occure with kernel 4.14 nor with Debian 5.x.
CPU i3-6100U, graphic chipset i915

graphic chipset i915

Have you tried the most recent linux-current (5.15.77)? There was an old patch for the i915 that I discovered had already been upstreamed in a different form so I removed it. It's possible that there was a conflict between that old patch and the upstreamed fix which was causing your issue. If that doesn't resolve it I can dig into it further.

Plume added a comment.Tue, Nov 8, 3:31 PM

Thank's for your answer.
The problem still remains with 5.15.77-219.

By the way, the upgrade to 5.15.77 went wrong on my machine (tried twice). /lib/modules/5.15.77-219.current is only 63674 ko, compared to /lib/modules.5.15.68-218.current which was 230960 ko. As a result, no network connection. However 5.15.77 boots and I experimented the same graphic issue.

Fabio added a subscriber: Fabio.EditedWed, Nov 16, 5:20 PM

In the Forum you ask about kernel problems, I have always encountered problems with the flash card reader and I think they are due to the kernel module.

This is lshw :

...      description: MMC Host
  product: AU6625 PCI-E Flash card reader controller
  vendor: Alcor Micro
  physical id: 0
  bus info: pci@0000:04:00.0
  logical name: mmc0
  version: 00
  width: 64 bits
  clock: 33MHz
  capabilities: cap_list
  configuration: driver=alcor_pci latency=0
  resources: irq:18 memory:b4300000-b4300fff
*-device
     description: SD Card
     product: SL16G
     vendor: SanDisk
     physical id: aaaa
     logical name: /dev/mmcblk0
     version: 8.0
     date: 03/2019
     serial: 277286665
     capabilities: sd ...

Quoted Text

This is a test of mount:

sudo mount /dev/mmcblk0 provamount
mount: /home/fabio/provamount: can't read superblock on /dev/mmcblk0.
dmesg(1) may have more information after failed mount system call.

dmesg:

[21893.385331] mmcblk0: recovery failed!
[21893.385337] print_req_error: 8 callbacks suppressed
[21893.385338] blk_update_request: I/O error, dev mmcblk0, sector 31116160 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
[21893.385428] mmcblk0: recovery failed!
[21893.385430] blk_update_request: I/O error, dev mmcblk0, sector 31116160 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[21893.385433] Buffer I/O error on dev mmcblk0, logical block 3889520, async page read
[21893.385649] mmcblk0: recovery failed!
[21893.385652] blk_update_request: I/O error, dev mmcblk0, sector 2 op 0x0:(READ) flags 0x1000 phys_seg 1 prio class 0
[21893.385656] EXT4-fs (mmcblk0): unable to read superblock

Other information:

Host: solus Kernel: 5.15.77-219.current arch: x86_64 bits
Machine:

Type: Laptop System: HP product: HP Pavilion Gaming Laptop 15-dk0xxx
  v: Type1

Thanks!

joebonrichie added a subscriber: joebonrichie.EditedWed, Nov 16, 6:17 PM

I know this is not the exact same laptop but according to this: https://wiki.archlinux.org/title/HP_Pavilion_15-cx0xxx

"Detected in kernel versions 5.6 and newer. If mounting the card does not work try booting with the kernel parameter intel_iommu=on and/or with the kernel parameter pci=nommconf. "

I think both models use the same card reader (AU6625)

So for example, create a file /etc/kernel/cmdline.d/sd-card-reader.conf with:
intel_iommu=on

Then run clr-boot-manager update, reboot, then retest the sd card reader.

From a quick search on the net I read quite a few reports of peeps having issues with the AU6625 card reader. It smells like an upstream issue to me.

Fabio added a comment.Wed, Nov 16, 7:33 PM

I have tried

  • intel_iommu=on
  • pci=nommconf
  • intel_iommu=o pci=nommconf

All 3 don't work, but the third is less verbose and "red" in dmesg:
[ 564.805781] mmc0: error -110 whilst initialising SD card
[ 568.380182] mmc0: Skipping voltage switch
[ 568.581615] mmc0: new high speed SDHC card at address aaaa
[ 568.582346] mmcblk0: mmc0:aaaa SL16G 14.8 GiB

Anyway thanks

ReillyBrogan triaged this task as High priority.

By the way, the upgrade to 5.15.77 went wrong on my machine (tried twice). /lib/modules/5.15.77-219.current is only 63674 ko, compared to /lib/modules.5.15.68-218.current which was 230960 ko. As a result, no network connection. However 5.15.77 boots and I experimented the same graphic issue.

I don't understand where you are getting this number. By my count there are 2705 kernel modules in the 5.15.77-99 kernel.

As I mentionned elsewhere in the forum, my machine have graphical issues with kernel 5.x. Every now end then, the screen turns black when the mouse cursor is in the lower part of the screen and remains black until the mouse is moved again. This trouble doesn't occure with kernel 4.14 nor with Debian 5.x.
CPU i3-6100U, graphic chipset i915

Are there any errors in the kernel log?

Staudey added a subscriber: Staudey.EditedWed, Nov 16, 8:26 PM

I don't understand where you are getting this number. By my count there are 2705 kernel modules in the 5.15.77-99 kernel.

I think they're just talking about the size of the modules (ko for kb in French/France), and I guess the reduction is due to the new zstd compression

In the Forum you ask about kernel problems, I have always encountered problems with the flash card reader and I think they are due to the kernel module.

So to be clear is this issue present in 4.14 as well? Does it work in a different distro (such as live booting from Ubuntu 22.10)?

I don't understand where you are getting this number. By my count there are 2705 kernel modules in the 5.15.77-99 kernel.

I think they're just talking about the size of the modules (ko for kb in French/France), and I guess the reduction is due to the new zstd compression

That makes sense. Yes, the reduction in size is to be expected.

Plume added a comment.Wed, Nov 16, 8:47 PM

By the way, the upgrade to 5.15.77 went wrong on my machine (tried twice). /lib/modules/5.15.77-219.current is only 63674 ko, compared to /lib/modules.5.15.68-218.current which was 230960 ko. As a result, no network connection. However 5.15.77 boots and I experimented the same graphic issue.

I don't understand where you are getting this number. By my count there are 2705 kernel modules in the 5.15.77-99 kernel.

philippe@hystou /data/home/philippe_solus $ du -s /lib/modules/*
184856 /lib/modules/4.14.298-188.lts
230960 /lib/modules/5.15.68-218.current
71116 /lib/modules/5.15.77-219.current (with broadcom-sta-current installed)

Fabio added a comment.EditedWed, Nov 16, 10:48 PM

In the Forum you ask about kernel problems, I have always encountered problems with the flash card reader and I think they are due to the kernel module.

So to be clear is this issue present in 4.14 as well?

I don't know

Does it work in a different distro (such as live booting from Ubuntu 22.10)?

I only used Solus on this laptop so,now, I just downloaded the live Ubuntu 22.10 and apparently it won't read the flash card exactly like Solus.