Re: [PATCH 2/2] thermal: intel: int340x: Allow temperature override

[srinivas pandruvada <srinivas.pandruvada@linux.intel.com>]

On Fri, 2025-06-06 at 07:22 +0000, Zhang, Rui wrote:
> On Thu, 2025-06-05 at 10:20 -0700, srinivas pandruvada wrote:
> > > >  
> > > >  int proc_thermal_ptc_add(struct pci_dev *pdev, struct
> > > > proc_thermal_device *proc_priv)
> > > >  {
> > > > @@ -230,10 +289,13 @@ int proc_thermal_ptc_add(struct pci_dev
> > > > *pdev,
> > > > struct proc_thermal_device *proc_
> > > >  
> > > >  		for (i = 0; i < PTC_MAX_INSTANCES; i++) {
> > > >  			ptc_instance[i].offset =
> > > > ptc_offsets[i];
> > > > +			ptc_instance[i].pdev = pdev;
> > > >  			ptc_create_groups(pdev, i,
> > > > &ptc_instance[i]);
> > > >  		}
> > > >  	}
> > > >  
> > > > +	ptc_create_debugfs();
> > > > +
> > > 
> > > should we create the debugfs only when PROC_THERMAL_FEATURE_PTC
> > > is
> > > set?
> > 
> > This function is only called when
> >  if (feature_mask & PROC_THERMAL_FEATURE_PTC) {
> > }
> > 
> > 
> right, then the
>    if (proc_priv->mmio_feature_mask & PROC_THERMAL_FEATURE_PTC)
> check in proc_thermal_ptc_add() is redundant.
Yes.

> 
> 
> BTW, in proc_thermal_mmio_add() and proc_thermal_mmio_remove(), we
> 
> 1. call the rapl/ptc/rfim/wt_req/wt_hint .add() functions with the
> feature mask check in proc_thermal_mmio_add()
> 
> 2. call the .remove() functions without the feature mask check in
> failure
> cases in proc_thermal_mmio_add().
The current functions in the upstream code are for:
	proc_thermal_rfim_remove(pdev);
This is already protected inside as it has to check each feature.

	proc_thermal_rapl_remove();

This is by virtue of rapl_mmio_priv.control_type is NULL on error, so
the above function will return.

proc_thermal_ptc_remove() is also protected by the flag. But you are
right for the debugfs part added with this patch, which should be
inside the flag check.


> 
> 3. call the .remove() functions with feature mask check in
> proc_thermal_mmio_remove()
> 
> This is inconsistent. If you agree, I'd like to propose a cleanup
> patch
> to make them work in a unified way.
You can make a cleanup patch to be consistent.

Thanks,
Srinivas



> 
> thanks,
> rui