Re: PowerMac G5 fans run out of control with kernel4.17.0-3-powerpc64but not with 4.16.0-1-powerpc64
> On Aug 27, 2018, at 10:11 AM, Rick Thomas <rbthomas@pobox.com> wrote:
>
>
>> On Aug 26, 2018, at 1:25 AM, Mark G.B. <mark.balantzyan@gmail.com> wrote:
>>
>> Hi Rick and everyone,
>>
>> It’s a bug (IMHO). Feel free to correct me if I’m wrong, though.
>>
>> Attached here are two images of the Debian-powerpc64 specific kernel-config extracted from the .deb package of the Linux-image package off of packages.debian.org, .de server.
>>
>> Debian powerpc64 kernel 4.17 (port) has no mention of windfarm_core being compiled in EITHER AS internal-functionally or as a module (nocore.png). Not trying to be whistleblower on the porters! Actually, IMHO, it’s just a minor bug to me and can easily be fixed with a modprobe 😊. The “tornado” was probably from all the other windfarm modules being included/started W/O the core.
>>
>> For reference, i2c_powermac is compiled in with respect to this kernel revision for powerpc64, as shown in i2cyes.png.
>>
>> Cheers,
>> Mark
>
> Hi All,
>
> I observe that the windfarm configuration in /boot/config-4.17.0-3-powerpc64 and /boot/config-4.16.0-1-powerpc64 are identical:
Not surprising. The configuration is defined by the Debian packaging of the Linux kernel. And since we didn’t change anything, it’s not unusual for the configs to match.
> Adrian hinted that it has something to do with a name change, but I’m unable to figure out what that change could be???
It could be that a module in the kernel got split into two and hence the Debian kernel configuration needs to be updated.
Again, if upstream makes some changes, it’s our duty in Debian to adapt these changes by updating our configuration.
It can also be a bug, of course.
> If it is a bug, it’s a minor one, and it’s easily worked around by the addition to /etc/modules. But I’m curious to know what changed? If it is a name change, as Adrian suggests, I’m curious to know what changed and why it was changed…
Changes like these happen in the kernel all the time. Usually for refactoring of code or updates due to internal API changes in the kernel.
Adrian
Reply to: