Re: [PATCH] driver core: Disable driver deferred probe timeout by default

[Javier Martinez Canillas]

On 11/14/22 12:38, Greg Kroah-Hartman wrote:
> On Mon, Nov 14, 2022 at 12:13:15PM +0100, Javier Martinez Canillas wrote:
>> Hello Greg,

[...]

>> I even gave an example about general purpose distributions that build as
>> much as possible as a module. What more info do you think that is missing?
> 
> Exact systems that this is failing on would be great to have.
>

The exact system is a Snapdragon SC7180 based HP X2 Chromebook with latest
Fedora Rawhide image (kernel version 6.1-rc4). The reason why is timing out
is that the arm_smmu driver is built-in (CONFIG_ARM_SMMU=y) but it depends
on gpucc-sc7180 clk driver that's built as module (CONFIG_SC_GPUCC_7180=m).

 >>> failing on the current value?  What drivers are causing the long delay
>>> here?  No one should be having to wait 10 seconds for a deferred delay
>>> on a real system as that feels really wrong.
>>>
>>
>> Not really, it depends if the drivers are built-in, built as modules, in
>> the initramfs or in the rootfs. A 10 seconds might not be enough if these
>> modules are in the root partition and need to wait for this to be mounted
>> and udev to load the modules, etc.
> 
> How does it take 10 seconds to load the initramfs for a system that
> requires deferred probe devices?  What typs of hardware is this?
>

That could depend on may things. The dependency of the systemd unit files,
whether NetworkManager-wait-online.service is enabled or not, etc. It can
really take more than 10 seconds on some systems to load all the modules.
 
[...]

>>
>> A nice feature of the probe deferral mechanism is that it was simple and
>> reliable. Adding a timeout makes it non-deterministic and more fragile IMO.
> 
> deferred probe was never simple or reliable or determinisitic.  It was a
> hack we had to implement to handle complex hardware situations and
> loadable drivers.  Let's not try to paper over driver bugs here by
> making the timeout "forever" but rather fix the root problem in the
> broken drivers.
>

I don't understand how adding a 10 secs timeout would make it more robust than
just letting the driver core to attempt probing the deferred drivers again for
every driver (or device) that gets registered.
 
> So, what drivers do we need to fix up?
> 

So what exactly needs to get fixed on the arm_smmu and gpucc-sc7180 drivers
mentioned? Yes, we could built both of them as =y and make sure that drivers
are registered and probed before the initcalls are done, but if we do that,
we will need to have most of the drivers built-in in the Fedora kernel. That
does not scale for all the platforms that we need to support.

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat