Re: [PATCH v17 30/32] fs/resctrl: Provide interface to create architecture specific debugfs area

["Luck, Tony" <tony.luck@intel.com>]

On Sat, Jan 10, 2026 at 11:57:58AM +0100, Borislav Petkov wrote:
> On Wed, Dec 17, 2025 at 09:21:17AM -0800, Tony Luck wrote:
> > All files below /sys/fs/resctrl are considered user ABI.
> > 
> > This leaves no place for architectures to provide additional interfaces.
> > 
> > Add resctrl_debugfs_mon_info_arch_mkdir() which creates a directory in
> > the debugfs file system for a monitoring resource. Naming follows the
> > layout of the main resctrl hierarchy:
> > 
> > 	/sys/kernel/debug/resctrl/info/{resource}_MON/{arch}
> > 
> > The {arch} last level directory name matches the output of the user level
> > "uname -m" command.
> > 
> > Architecture code may use this directory for debug information, or for minor
> > tuning of features. It must not be used for basic feature enabling as debugfs
> > may not be configured/mounted on production systems.
> 
> I, like you guys, thought that debugfs is "safe" in the sense, stuff there
> can't really be an ABI but just recently at LPC I got schooled about it and,
> basically, if anything in luserspace starts using debugfs, it will be
> considered an ABI.
> 
> And this thinking has been there for a looong time now:
> 
> https://lwn.net/Articles/309298/
> 
> So, before you put anything there, think again because you might end up
> supporting it just like an ABI.

Please drop patches 30, 31, and the debugfs hunk to Documentation from patch 32
while I look at options. The debugfs bits aren't needed for normal
operation, just for debugging if things don't appear to be working as
expected. No need to hold up inclusion waiting for a solution.

Telemetry based events tied to RMIDs are new in Clearwater Forest. I
have line of sight to a second platorm. But extrapolating from two
data points to decide on a stable ABI seems very risky.

Thanks

-Tony