New kernels

[xeno74]

So tested the latest RC5 on Fienix, MATE and Mint. Working normally from what I can tell.

Thank you for testing!

On Mint I keep seeing some crash in dmesg that was also present in the last 6.16 RC5:

Code: Select all

[  116.258486] at-spi-bus-laun[5738]: segfault (11) at 2c6d48 nip f75dcdb4 lr 140b60 code 1 in libc.so.6[ccdb4,f7510000+225000]

Maybe an userland issue?

On MATE from which I'm typing this now I see an error about zram:

Code: Select all

Sep 11 01:40:23 ubuntu init-zram-swapping[1602]: modinfo: ERROR: Module zram not found.
Sep 11 01:40:24 ubuntu init-zram-swapping[1602]: modinfo: ERROR: Module zram not found.

Do you need the kernel module zram? (Virtual Swap Compressed in RAM)

Also see this which looks close to at-spi:

Code: Select all

Sep 11 01:42:08 ubuntu at-spi2-registr[3204]: Failed to register client: GDBus.Error:org.gnome.SessionManager.AlreadyRegistered: Unable to register client
Sep 11 01:42:08 ubuntu at-spi2-registr[3204]: Failed to register client: GDBus.Error:org.gnome.SessionManager.AlreadyRegistered: Unable to register client

Do you see these error messages in old kernels?

[xeno74]

Hypex,

We have the same at-spi-bus messages with MintPPC on the X5000.

Seems to be OK here:
neowofetch_kernel6.17rc2.png

I do get the following error on boot, but I think it was there before, probably related to the systemd problems and not unique to this kernel. It repeats from time to time.

Code: Select all

[  759.765434] at-spi-bus-laun[2710]: segfault (11) at 146d48 nip f76acdb4 lr f7c00ae0 code 1 in libc.so.6[ccdb4,f75e0000+225000]
[  759.765468] at-spi-bus-laun[2710]: code: 4182ffcc 7f85e378 7fa4eb78 7f63db78 7f9be214 4bffb309 39400000 995cffff 
[  759.765476] at-spi-bus-laun[2710]: code: 4bffffac 5464003a 3ce07f7f 54651ef8 <81040000> 3920ffff 38e77f7f 7d292c30 

Will continue to user this kernel unless something bad happens

Cheers,
Christian

[xeno74]

Hypex,

FYI:

We have the following error messages on MintPPC currently.

Code: Select all

[  759.765434] at-spi-bus-laun[2710]: segfault (11) at 146d48 nip f76acdb4 lr f7c00ae0 code 1 in libc.so.6[ccdb4,f75e0000+225000]
[  759.765468] at-spi-bus-laun[2710]: code: 4182ffcc 7f85e378 7fa4eb78 7f63db78 7f9be214 4bffb309 39400000 995cffff 
[  759.765476] at-spi-bus-laun[2710]: code: 4bffffac 5464003a 3ce07f7f 54651ef8 <81040000> 3920ffff 38e77f7f 7d292c30 

Do you have the same error messages on your machines?

I see the same error.

I don't think you need this service. I think it is needed for people who need screen readers, when being blind.
I filed a bug report.

Cheers,
Christian

[Hypex]

Maybe an userland issue?

Possible. I did have some issues with Debian before I had a compatible kernel. Mint continues to have problems with LLVM relocation errors.

Actually I still have issues with Debian. It still randomly reboots. I've locked it to kernel 6.15 for now as it seems stable. The reality is no kernel is stable as it could be 30 seconds or 30 minutes before it suddenly reboots so I can't trust it. The shell forgetting all history by default making it harder. I've suspected a hardware error but a Debian is the only one that reboots. Even Mint that is more buggy doesn't suddenly reboot.

Do you need the kernel module zram? (Virtual Swap Compressed in RAM)

Not for myself. I wonder if it's something the Zesty MATE setup needs or has installed? It, for example, doesn't appear in a Mint dmesg log.

Do you see these error messages in old kernels?

I don't. Just in recent kernels. Comparing with kernel 6.15.0 on Debian I don't see it. In kernel 6.16-rc5 on Mint it is there in a captured log.

We have the same at-spi-bus messages with MintPPC on the X5000.

Okay, we have a match. I also found a message about it in the Debian mailing list. So it looks like a universal crash:

https://lists.debian.org/debian-powerpc ... 00091.html