Re: [PATCH 7/8] drm/i915: Assert that our hpd pin bitmasks don't overflow

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Thu, Jul 05, 2018 at 05:52:16PM +0100, Chris Wilson wrote:
> Quoting Ville Syrjala (2018-07-05 17:43:56)
> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > 
> > Make sure our hpd pin count doesn't exceed the bitmasks we use
> > for tracking pending hotplugs. Not ever close to the limit yet,
> > but no harm in making sure either.
> > 
> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/i915_irq.c | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> > index bb7c754979f8..c107e0837026 100644
> > --- a/drivers/gpu/drm/i915/i915_irq.c
> > +++ b/drivers/gpu/drm/i915/i915_irq.c
> > @@ -1713,6 +1713,8 @@ static void intel_get_hpd_pins(struct drm_i915_private *dev_priv,
> >  {
> >         enum hpd_pin pin;
> >  
> > +       BUILD_BUG_ON(HPD_NUM_PINS > 32);
> 
> sizeof(*pin_mask) * CHAR_BIT ?

For a slightly more pleasing appearance I suppose we might want
to steal BITS_PER_TYPE() from include/linux/net_dim.h. A quick 
grep didn't seem to reveal anything similar in a more general
location.

> 
> Just looking for some explanation as where the limit comes from. Now
> obviously u32, but why was u32 chosen?

Had the right kind of smell perhaps?

-- 
Ville Syrjälä
Intel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx