Re: [PATCH v6 1/2] iommu/arm-smmu-v3: Add a cmdq_max_n_shift module parameter

[Will Deacon]

On Fri, Sep 25, 2026 at 11:01:31AM +0100, Kiryl Shutsemau wrote:
> On Wed, Sep 23, 2026 at 02:03:36PM -0300, Jason Gunthorpe wrote:
> > > [ ... 34 lines skipped ... ]
> > > @@ -40,6 +40,11 @@ module_param(disable_msipolling, bool, 0444);
> > >  MODULE_PARM_DESC(disable_msipolling,
> > >  	"Disable MSI-based polling for CMD_SYNC completion.");
> > >  
> > > +static unsigned int cmdq_max_n_shift;
> > > +module_param(cmdq_max_n_shift, uint, 0444);
> > > +MODULE_PARM_DESC(cmdq_max_n_shift,
> > > +	"Cap on the command queue depth, as log2 of the number of entries. Zero means the hardware maximum; the queue never shrinks below one page.");
> > 
> > The patch looks ok
> > 
> > Reviewed-by: Jason Gunthorpe <jgg@xxxxxxxxxx>
> > 
> > But I'd rather use lg2 as name, especially user visible, instead of
> > n_shift. "shift" is a word for bitwise stuff, this is a 2**N log2
> > encoded value.
> 
> This name has changed twice already. v5 took the value in entries.
> Nicolin asked for a log2 value and named it after max_n_shift, which is
> what the driver calls this quantity everywhere else.
> 
> I would rather not rename it a third time without input from the
> maintainer.
> 
> Will, this is the user-visible name. Do you have a preference here?
> Unless you want it changed I will leave v6 as it is.

I don't have really strong opinions on naming, but since Jason has given
a reasonably complete outline of the change and I have a small preference
for the lg2-style naming (rather than exposing the name of an internal
driver variable), it would be great if you can spin a v7 with that
included so that I can pick it up.

Thanks,

Will