Re: [PATCH v5 4/4] PCI: keystone: Add support to build as a loadable module

[Bjorn Helgaas <helgaas@kernel.org>]

On Thu, Nov 13, 2025 at 06:35:13PM +0000, Russell King (Oracle) wrote:
> On Thu, Nov 13, 2025 at 12:13:55PM -0600, Bjorn Helgaas wrote:
> > >  config PCI_KEYSTONE_HOST
> > > -	bool "TI Keystone PCIe controller (host mode)"
> > > +	tristate "TI Keystone PCIe controller (host mode)"
> > >  	depends on ARCH_KEYSTONE || ARCH_K3 || COMPILE_TEST
> > >  	depends on PCI_MSI
> > >  	select PCIE_DW_HOST
> > > -	select PCI_KEYSTONE
> > > +	select PCI_KEYSTONE if ARM
> > > +	select PCI_KEYSTONE_TRISTATE if !ARM
> > 
> > This is kind of a lot of dancing to make keystone built-in on ARM32
> > because hook_fault_code() is __init, while making it modular
> > everywhere else.
> > 
> > Is hook_fault_code() __init for some intrinsic reason?  All the
> > existing callers are __init, so that's one reason.  But could it be
> > made non-__init?
> 
> Yes. To discourage use in modules, because there is *no* way to safely
> remove a hook.
> 
> While one can call hook_fault_code() with a NULL handler, that doesn't
> mean that another CPU isn't executing in that function. If that code
> gets unmapped while another CPU is executing it (because of a module
> being unmapped) then we'll get another fault.
> 
> Trying to throw locks at this doesn't help - not without holding locks
> over the execution of the called function, which *will* be extremely
> detrimental on all fault handling, and probably introduce deadlocks.

Ah, thanks, I hadn't thought about the removal problem.