RE: [PATCH net-next 09/11] net: enetc: optimize the allocation of tx_bdr

[Wei Fang]

> -----Original Message-----
> From: Frank Li <frank.li@xxxxxxx>
> Sent: 2024年10月10日 1:25
> To: Wei Fang <wei.fang@xxxxxxx>
> Cc: davem@xxxxxxxxxxxxx; edumazet@xxxxxxxxxx; kuba@xxxxxxxxxx;
> pabeni@xxxxxxxxxx; robh@xxxxxxxxxx; krzk+dt@xxxxxxxxxx;
> conor+dt@xxxxxxxxxx; Vladimir Oltean <vladimir.oltean@xxxxxxx>; Claudiu
> Manoil <claudiu.manoil@xxxxxxx>; Clark Wang <xiaoning.wang@xxxxxxx>;
> christophe.leroy@xxxxxxxxxx; linux@xxxxxxxxxxxxxxx; bhelgaas@xxxxxxxxxx;
> imx@xxxxxxxxxxxxxxx; netdev@xxxxxxxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx;
> linux-kernel@xxxxxxxxxxxxxxx; linux-pci@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH net-next 09/11] net: enetc: optimize the allocation of
> tx_bdr
> 
> On Wed, Oct 09, 2024 at 05:51:14PM +0800, Wei Fang wrote:
> > From: Clark Wang <xiaoning.wang@xxxxxxx>
> >
> > There is a situation where num_tx_rings cannot be divided by
> > bdr_int_num. For example, num_tx_rings is 8 and bdr_int_num
> > is 3. According to the previous logic, this results in two
> > tx_bdr corresponding memories not being allocated, so when
> > sending packets to tx BD ring 6 or 7, wild pointers will be
> > accessed. Of course, this issue does not exist for LS1028A,
> > because its num_tx_rings is 8, and bdr_int_num is either 1
> > or 2. So there is no situation where it cannot be divided.
> > However, there is a risk for the upcoming i.MX95, so the
> > allocation of tx_bdr is optimized to ensure that each tx_bdr
> > can be allocated to the corresponding memory.
> >
> > Signed-off-by: Clark Wang <xiaoning.wang@xxxxxxx>
> > Signed-off-by: Wei Fang <wei.fang@xxxxxxx>
> > Reviewed-by: Claudiu Manoil <claudiu.manoil@xxxxxxx>
> > ---
> >  drivers/net/ethernet/freescale/enetc/enetc.c | 121 ++++++++++---------
> >  1 file changed, 62 insertions(+), 59 deletions(-)
> >
> > diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c
> b/drivers/net/ethernet/freescale/enetc/enetc.c
> > index 032d8eadd003..b84c88a76762 100644
> > --- a/drivers/net/ethernet/freescale/enetc/enetc.c
> > +++ b/drivers/net/ethernet/freescale/enetc/enetc.c
> > @@ -2965,13 +2965,70 @@ int enetc_ioctl(struct net_device *ndev, struct
> ifreq *rq, int cmd)
> >  }
> >  EXPORT_SYMBOL_GPL(enetc_ioctl);
> >
> > +static int enetc_bdr_init(struct enetc_ndev_priv *priv, int i, int v_tx_rings)
> > +{
> > +	struct enetc_int_vector *v __free(kfree);
> > +	struct enetc_bdr *bdr;
> > +	int j, err;
> > +
> > +	v = kzalloc(struct_size(v, tx_ring, v_tx_rings), GFP_KERNEL);
> > +	if (!v)
> > +		return -ENOMEM;
> > +
> > +	bdr = &v->rx_ring;
> > +	bdr->index = i;
> > +	bdr->ndev = priv->ndev;
> > +	bdr->dev = priv->dev;
> > +	bdr->bd_count = priv->rx_bd_count;
> > +	bdr->buffer_offset = ENETC_RXB_PAD;
> > +	priv->rx_ring[i] = bdr;
> > +
> > +	err = xdp_rxq_info_reg(&bdr->xdp.rxq, priv->ndev, i, 0);
> > +	if (err)
> > +		return err;
> > +
> > +	err = xdp_rxq_info_reg_mem_model(&bdr->xdp.rxq,
> > +					 MEM_TYPE_PAGE_SHARED, NULL);
> > +	if (err) {
> > +		xdp_rxq_info_unreg(&bdr->xdp.rxq);
> > +		return err;
> > +	}
> > +
> > +	/* init defaults for adaptive IC */
> > +	if (priv->ic_mode & ENETC_IC_RX_ADAPTIVE) {
> > +		v->rx_ictt = 0x1;
> > +		v->rx_dim_en = true;
> > +	}
> > +	INIT_WORK(&v->rx_dim.work, enetc_rx_dim_work);
> > +	netif_napi_add(priv->ndev, &v->napi, enetc_poll);
> > +	v->count_tx_rings = v_tx_rings;
> > +
> > +	for (j = 0; j < v_tx_rings; j++) {
> > +		int idx;
> > +
> > +		/* default tx ring mapping policy */
> > +		idx = priv->bdr_int_num * j + i;
> > +		__set_bit(idx, &v->tx_rings_map);
> > +		bdr = &v->tx_ring[j];
> > +		bdr->index = idx;
> > +		bdr->ndev = priv->ndev;
> > +		bdr->dev = priv->dev;
> > +		bdr->bd_count = priv->tx_bd_count;
> > +		priv->tx_ring[idx] = bdr;
> > +	}
> > +
> > +	priv->int_vector[i] = no_free_ptr(v);
> > +
> > +	return 0;
> > +}
> > +
> >  int enetc_alloc_msix(struct enetc_ndev_priv *priv)
> >  {
> >  	struct pci_dev *pdev = priv->si->pdev;
> > +	int v_tx_rings, v_remainder;
> >  	int num_stack_tx_queues;
> >  	int first_xdp_tx_ring;
> >  	int i, n, err, nvec;
> > -	int v_tx_rings;
> >
> >  	nvec = ENETC_BDR_INT_BASE_IDX + priv->bdr_int_num;
> >  	/* allocate MSIX for both messaging and Rx/Tx interrupts */
> > @@ -2985,65 +3042,11 @@ int enetc_alloc_msix(struct enetc_ndev_priv
> *priv)
> >
> >  	/* # of tx rings per int vector */
> >  	v_tx_rings = priv->num_tx_rings / priv->bdr_int_num;
> > +	v_remainder = priv->num_tx_rings % priv->bdr_int_num;
> >
> > -	for (i = 0; i < priv->bdr_int_num; i++) {
> > -		struct enetc_int_vector *v;
> > -		struct enetc_bdr *bdr;
> > -		int j;
> > -
> > -		v = kzalloc(struct_size(v, tx_ring, v_tx_rings), GFP_KERNEL);
> > -		if (!v) {
> > -			err = -ENOMEM;
> > -			goto fail;
> > -		}
> > -
> > -		priv->int_vector[i] = v;
> > -
> > -		bdr = &v->rx_ring;
> > -		bdr->index = i;
> > -		bdr->ndev = priv->ndev;
> > -		bdr->dev = priv->dev;
> > -		bdr->bd_count = priv->rx_bd_count;
> > -		bdr->buffer_offset = ENETC_RXB_PAD;
> > -		priv->rx_ring[i] = bdr;
> > -
> > -		err = xdp_rxq_info_reg(&bdr->xdp.rxq, priv->ndev, i, 0);
> > -		if (err) {
> > -			kfree(v);
> > -			goto fail;
> > -		}
> > -
> > -		err = xdp_rxq_info_reg_mem_model(&bdr->xdp.rxq,
> > -						 MEM_TYPE_PAGE_SHARED, NULL);
> > -		if (err) {
> > -			xdp_rxq_info_unreg(&bdr->xdp.rxq);
> > -			kfree(v);
> > -			goto fail;
> > -		}
> > -
> > -		/* init defaults for adaptive IC */
> > -		if (priv->ic_mode & ENETC_IC_RX_ADAPTIVE) {
> > -			v->rx_ictt = 0x1;
> > -			v->rx_dim_en = true;
> > -		}
> > -		INIT_WORK(&v->rx_dim.work, enetc_rx_dim_work);
> > -		netif_napi_add(priv->ndev, &v->napi, enetc_poll);
> > -		v->count_tx_rings = v_tx_rings;
> > -
> > -		for (j = 0; j < v_tx_rings; j++) {
> > -			int idx;
> > -
> > -			/* default tx ring mapping policy */
> > -			idx = priv->bdr_int_num * j + i;
> > -			__set_bit(idx, &v->tx_rings_map);
> > -			bdr = &v->tx_ring[j];
> > -			bdr->index = idx;
> > -			bdr->ndev = priv->ndev;
> > -			bdr->dev = priv->dev;
> > -			bdr->bd_count = priv->tx_bd_count;
> > -			priv->tx_ring[idx] = bdr;
> > -		}
> > -	}
> > +	for (i = 0; i < priv->bdr_int_num; i++)
> > +		enetc_bdr_init(priv, i,
> > +			       i < v_remainder ? v_tx_rings + 1 : v_tx_rings);
> 
> suggest you create two patches, one just move to help function to
> enetc_bdr_init(), the another is for real fixes.

Sure, thanks

> 
> >
> >  	num_stack_tx_queues = enetc_num_stack_tx_queues(priv);
> >
> > --
> > 2.34.1
> >