Re: [Intel-wired-lan] [PATCH iwl-next v1] ice: lower default irq/queue counts on high-core systems

[Michal Swiatkowski <michal.swiatkowski@linux.intel.com>]

On Thu, Oct 16, 2025 at 09:44:43AM +0200, Paul Menzel wrote:
> Dear Michal,
> 
> 
> Thank you for the patch. I’d mention the 64 in the summary:
> 

Sure, I will add it.

> > ice: lower default irq/queue counts to 64 on > 64 core systems
> 
> 
> Am 16.10.25 um 08:22 schrieb Michal Swiatkowski:
> > On some high-core systems loading ice driver with default values can
> > lead to queue/irq exhaustion. It will result in no additional resources
> > for SR-IOV.
> > 
> > In most cases there is no performance reason for more than 64 queues.
> > Limit the default value to 64. Still, using ethtool the number of
> > queues can be changed up to num_online_cpus().
> > 
> > This change affects only the default queue amount on systems with more
> > than 64 cores.
> 
> Please document a specific system and steps to reproduce the issue.
> 
> Please also document how to override the value.

Ok, will add.

> 
> > Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> > Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
> > ---
> >   drivers/net/ethernet/intel/ice/ice.h     | 20 ++++++++++++++++++++
> >   drivers/net/ethernet/intel/ice/ice_irq.c |  6 ++++--
> >   drivers/net/ethernet/intel/ice/ice_lib.c |  8 ++++----
> >   3 files changed, 28 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
> > index 3d4d8b88631b..354ec2950ff3 100644
> > --- a/drivers/net/ethernet/intel/ice/ice.h
> > +++ b/drivers/net/ethernet/intel/ice/ice.h
> > @@ -1133,4 +1133,24 @@ static inline struct ice_hw *ice_get_primary_hw(struct ice_pf *pf)
> >   	else
> >   		return &pf->adapter->ctrl_pf->hw;
> >   }
> > +
> > +/**
> > + * ice_capped_num_cpus - normalize the number of CPUs to a reasonable limit
> > + *
> > + * This function returns the number of online CPUs, but caps it at suitable
> > + * default to prevent excessive resource allocation on systems with very high
> > + * CPU counts.
> > + *
> > + * Note: suitable default is currently at 64, which is reflected in default_cpus
> > + * constant. In most cases there is no much benefit for more than 64 and it is a
> 
> no*t* much
>

Will fix

> > + * power of 2 number.
> > + *
> > + * Return: number of online CPUs, capped at suitable default.
> > + */
> > +static inline u16 ice_capped_num_cpus(void)
> 
> Why not return `unsigned int` or `size_t`?
>

Just because u16 is used for queue counts, but I can go with unsigned
int, makes more sense as num_online_cpus() is returning unsigned int.

> > +{
> > +	const int default_cpus = 64;
> > +
> > +	return min(num_online_cpus(), default_cpus);
> > +}
> >   #endif /* _ICE_H_ */
> > diff --git a/drivers/net/ethernet/intel/ice/ice_irq.c b/drivers/net/ethernet/intel/ice/ice_irq.c
> > index 30801fd375f0..df4d847ca858 100644
> > --- a/drivers/net/ethernet/intel/ice/ice_irq.c
> > +++ b/drivers/net/ethernet/intel/ice/ice_irq.c
> > @@ -106,9 +106,11 @@ static struct ice_irq_entry *ice_get_irq_res(struct ice_pf *pf,
> >   #define ICE_RDMA_AEQ_MSIX 1
> >   static int ice_get_default_msix_amount(struct ice_pf *pf)
> >   {
> > -	return ICE_MIN_LAN_OICR_MSIX + num_online_cpus() +
> > +	u16 cpus = ice_capped_num_cpus();
> > +
> > +	return ICE_MIN_LAN_OICR_MSIX + cpus +
> >   	       (test_bit(ICE_FLAG_FD_ENA, pf->flags) ? ICE_FDIR_MSIX : 0) +
> > -	       (ice_is_rdma_ena(pf) ? num_online_cpus() + ICE_RDMA_AEQ_MSIX : 0);
> > +	       (ice_is_rdma_ena(pf) ? cpus + ICE_RDMA_AEQ_MSIX : 0);
> >   }
> >   /**
> > diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
> > index bac481e8140d..3c5f8a4b6c6d 100644
> > --- a/drivers/net/ethernet/intel/ice/ice_lib.c
> > +++ b/drivers/net/ethernet/intel/ice/ice_lib.c
> > @@ -159,12 +159,12 @@ static void ice_vsi_set_num_desc(struct ice_vsi *vsi)
> >   static u16 ice_get_rxq_count(struct ice_pf *pf)
> >   {
> > -	return min(ice_get_avail_rxq_count(pf), num_online_cpus());
> > +	return min(ice_get_avail_rxq_count(pf), ice_capped_num_cpus());
> >   }
> >   static u16 ice_get_txq_count(struct ice_pf *pf)
> >   {
> > -	return min(ice_get_avail_txq_count(pf), num_online_cpus());
> > +	return min(ice_get_avail_txq_count(pf), ice_capped_num_cpus());
> >   }
> >   /**
> > @@ -907,13 +907,13 @@ static void ice_vsi_set_rss_params(struct ice_vsi *vsi)
> >   		if (vsi->type == ICE_VSI_CHNL)
> >   			vsi->rss_size = min_t(u16, vsi->num_rxq, max_rss_size);
> >   		else
> > -			vsi->rss_size = min_t(u16, num_online_cpus(),
> > +			vsi->rss_size = min_t(u16, ice_capped_num_cpus(),
> >   					      max_rss_size);
> >   		vsi->rss_lut_type = ICE_LUT_PF;
> >   		break;
> >   	case ICE_VSI_SF:
> >   		vsi->rss_table_size = ICE_LUT_VSI_SIZE;
> > -		vsi->rss_size = min_t(u16, num_online_cpus(), max_rss_size);
> > +		vsi->rss_size = min_t(u16, ice_capped_num_cpus(), max_rss_size);
> >   		vsi->rss_lut_type = ICE_LUT_VSI;
> >   		break;
> >   	case ICE_VSI_VF:
> 
> With the changes addressed, feel free to add:
> 
> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
> 

Thanks

> 
> Kind regards,
> 
> Paul