New kernels

[xeno74]

Is this intentional for the alpha 5 archive?

Yes, it is intended. Now, the kernel developers know the commit with the issue. Thanks a lot for testing!

[Hypex]

Yes, it is intended. Now, the kernel developers know the commit with the issue. Thanks a lot for testing!

Okay. Great! I did a check but didn't see any failed module load errors.

[xeno74]

FYI: radeon_fbdev_river_fbdev: failed to initialize framebuffer and setup emulation

[xeno74]

We have a patch!!! Link

[xeno74]

We have a patch!!! Link

I compiled the alpha6 of kernel 6.17 with the new Radeon patch today.

Please test.

[Hypex]

I compiled the alpha6 of kernel 6.17 with the new Radeon patch today.

Please test.

It works!

So I tested on Fienix. All looks good. Dmesg appears normal. I did notice this message about device nodes:

Code: Select all

[   13.196763] systemd[1]: Create List of Static Device Nodes was skipped because of an unmet condition check (ConditionFileNotEmpty=/lib/modules/6.17.0-a6-powerpc64-smp-g37816488247d-dirty/modules.devname).

Also the i2c-pasemi messages about NACK and resetting kept appearing. I'm typing this on Debian now with 6.15.0-powerpc64-smp. And the messages don't appear. So it would be with 6.16 and above by the looks of it. I would have posted when I noticed it.

P.S.:
Just checked. It was 6.15 alpha 5 I noticed. Need to check what that device is but the messages don't look good to me:
viewtopic.php?p=60443#p60443

Log extract:

Code: Select all

[    9.459706] i2c-pasemi 0000:00:1c.0: NACK, status 0x08210100
[    9.649736] i2c-pasemi 0000:00:1c.0: Issuing reset due to status 0x08210100 (xfstatus 0x00000001)
[    9.690696] i2c-pasemi 0000:00:1c.1: NACK, status 0x08210100
[    9.708651] i2c-pasemi 0000:00:1c.1: Issuing reset due to status 0x08210100 (xfstatus 0x00000001)
[    9.892693] i2c-pasemi 0000:00:1c.2: NACK, status 0x08210100
[    9.911535] i2c-pasemi 0000:00:1c.2: Issuing reset due to status 0x08210100 (xfstatus 0x00000001)

Okay looked it up and there appears to be an issue with reset of my SMBus:
https://devicehunt.com/view/type/pci/ve ... evice/A003

Device log from 6.15.0-powerpc64-smp:

Code: Select all

[    0.166709] pci 0000:00:1c.0: [1959:a003] type 00 class 0x0c0500 conventional PCI endpoint
[    0.166750] pci 0000:00:1c.0: BAR 0 [io  0x870200-0x87023f]
[    0.166866] pci 0000:00:1c.1: [1959:a003] type 00 class 0x0c0500 conventional PCI endpoint
[    0.166901] pci 0000:00:1c.1: BAR 0 [io  0x870240-0x87027f]
[    0.167017] pci 0000:00:1c.2: [1959:a003] type 00 class 0x0c0500 conventional PCI endpoint
[    0.167053] pci 0000:00:1c.2: BAR 0 [io  0x870280-0x8702bf]

[xeno74]

Hi Hypex,

It works!

Great!

I did notice this message about device nodes:

Code: Select all

[   13.196763] systemd[1]: Create List of Static Device Nodes was skipped because of an unmet condition check (ConditionFileNotEmpty=/lib/modules/6.17.0-a6-powerpc64-smp-g37816488247d-dirty/modules.devname).

I don’t have this message. Are the kernel modules loaded? Please check them with lsmod.

Thanks,
Christian

[xeno74]

Okay looked it up and there appears to be an issue with reset of my SMBus

Maybe Sven can help us. (Sven Peter sven at svenpeter.dev)

Links:

• i2c: pasemi: Split pci driver to its own file
• i2c-pasemi-pci.c

[Hypex]

Hi Hypex,

I don’t have this message. Are the kernel modules loaded? Please check them with lsmod.

Thanks,
Christian

You're right. They weren't loaded. Somehow I missed copying the modules over. I haven't missed in a while. Thanks!

Maybe Sven can help us. (Sven Peter sven at svenpeter.dev)

So there was a change. Excellent. That will help find it.

[xeno74]

FYI: i2c-pasemi messages since kernel 6.16