i40e: allow for more VSIs

The number of VSIs that the firmware reports to us is a guaranteed
minimum, not an absolute maximum. The hardware actually supports far
more  than the reported value, which we often need.

To allow for this, we allocate space for a larger number of VSIs than is
guaranteed by the firmware, with the knowledge that we may fail to get
them all in the future.

Note that we are just allocating pointers here, the actual (much larger)
VSI structures are allocated on demand.

Change-ID: I6f4e535ce39d3bf417aef78306e04fbc7505140e
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

drivers/net/ethernet/intel/i40e/i40e.h

@@ -72,6 +72,7 @@
 #define I40E_MIN_NUM_DESCRIPTORS      64
 #define I40E_MIN_MSIX                 2
 #define I40E_DEFAULT_NUM_VMDQ_VSI     8 /* max 256 VSIs */
+#define I40E_MIN_VSI_ALLOC            51 /* LAN, ATR, FCOE, 32 VF, 16 VMDQ */
 #define I40E_DEFAULT_QUEUES_PER_VMDQ  2 /* max 16 qps */
 #define I40E_DEFAULT_QUEUES_PER_VF    4
 #define I40E_DEFAULT_QUEUES_PER_TC    1 /* should be a power of 2 */
@@ -215,6 +216,7 @@ struct i40e_pf {
 	u16 rss_size;              /* num queues in the RSS array */
 	u16 rss_size_max;          /* HW defined max RSS queues */
 	u16 fdir_pf_filter_count;  /* num of guaranteed filters for this PF */
+	u16 num_alloc_vsi;         /* num VSIs this driver supports */
 	u8 atr_sample_rate;
 	bool wol_en;
 

drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c

@@ -232,7 +232,7 @@ static void i40e_dcbnl_del_app(struct i40e_pf *pf,
 			      struct i40e_ieee_app_priority_table *app)
 {
 	int v, err;
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] && pf->vsi[v]->netdev) {
 			err = i40e_dcbnl_vsi_del_app(pf->vsi[v], app);
 			if (err)

drivers/net/ethernet/intel/i40e/i40e_debugfs.c

@@ -45,7 +45,7 @@ static struct i40e_vsi *i40e_dbg_find_vsi(struct i40e_pf *pf, int seid)
 	if (seid < 0)
 		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
 	else
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+		for (i = 0; i < pf->num_alloc_vsi; i++)
 			if (pf->vsi[i] && (pf->vsi[i]->seid == seid))
 				return pf->vsi[i];
 
@@ -843,7 +843,7 @@ static void i40e_dbg_dump_vsi_no_seid(struct i40e_pf *pf)
 {
 	int i;
 
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			dev_info(&pf->pdev->dev, "dump vsi[%d]: %d\n",
 				 i, pf->vsi[i]->seid);
@@ -1526,7 +1526,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			cnt = sscanf(&cmd_buf[15], "%i", &vsi_seid);
 			if (cnt == 0) {
 				int i;
-				for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+				for (i = 0; i < pf->num_alloc_vsi; i++)
 					i40e_vsi_reset_stats(pf->vsi[i]);
 				dev_info(&pf->pdev->dev, "vsi clear stats called for all vsi's\n");
 			} else if (cnt == 1) {

drivers/net/ethernet/intel/i40e/i40e_main.c

@@ -652,7 +652,7 @@ static void i40e_update_link_xoff_rx(struct i40e_pf *pf)
 		return;
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for all Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
 		if (!vsi || !vsi->tx_rings[0])
@@ -706,7 +706,7 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
 	}
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
 		if (!vsi || !vsi->tx_rings[0])
@@ -1734,7 +1734,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 		return;
 	pf->flags &= ~I40E_FLAG_FILTER_SYNC;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] &&
 		    (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED))
 			i40e_sync_vsi_filters(pf->vsi[v]);
@@ -3524,7 +3524,7 @@ static void i40e_clear_interrupt_scheme(struct i40e_pf *pf)
 	int i;
 
 	i40e_put_lump(pf->irq_pile, 0, I40E_PILE_VALID_BIT-1);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			i40e_vsi_free_q_vectors(pf->vsi[i]);
 	i40e_reset_interrupt_capability(pf);
@@ -3614,7 +3614,7 @@ static void i40e_pf_quiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_quiesce_vsi(pf->vsi[v]);
 	}
@@ -3628,7 +3628,7 @@ static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_unquiesce_vsi(pf->vsi[v]);
 	}
@@ -4069,7 +4069,7 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf)
 	}
 
 	/* Update each VSI */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v])
 			continue;
 
@@ -4592,7 +4592,7 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags)
 		/* Find the VSI(s) that requested a re-init */
 		dev_info(&pf->pdev->dev,
 			 "VSI reinit requested\n");
-		for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+		for (v = 0; v < pf->num_alloc_vsi; v++) {
 			struct i40e_vsi *vsi = pf->vsi[v];
 			if (vsi != NULL &&
 			    test_bit(__I40E_REINIT_REQUESTED, &vsi->state)) {
@@ -4919,7 +4919,7 @@ static void i40e_veb_link_event(struct i40e_veb *veb, bool link_up)
 			i40e_veb_link_event(pf->veb[i], link_up);
 
 	/* ... now the local VSIs */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && (pf->vsi[i]->uplink_seid == veb->seid))
 			i40e_vsi_link_event(pf->vsi[i], link_up);
 }
@@ -4976,7 +4976,7 @@ static void i40e_check_hang_subtask(struct i40e_pf *pf)
 	 *     for each q_vector
 	 *         force an interrupt
 	 */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 		int armed = 0;
 
@@ -5026,7 +5026,7 @@ static void i40e_watchdog_subtask(struct i40e_pf *pf)
 	/* Update the stats for active netdevs so the network stack
 	 * can look at updated numbers whenever it cares to
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->netdev)
 			i40e_update_stats(pf->vsi[i]);
 
@@ -5278,7 +5278,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 	int ret;
 
 	/* build VSI that owns this VEB, temporarily attached to base VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis && !ctl_vsi; v++) {
+	for (v = 0; v < pf->num_alloc_vsi && !ctl_vsi; v++) {
 		if (pf->vsi[v] &&
 		    pf->vsi[v]->veb_idx == veb->idx &&
 		    pf->vsi[v]->flags & I40E_VSI_FLAG_VEB_OWNER) {
@@ -5308,7 +5308,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 		goto end_reconstitute;
 
 	/* create the remaining VSIs attached to this VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v] || pf->vsi[v] == ctl_vsi)
 			continue;
 
@@ -5421,7 +5421,7 @@ static void i40e_fdir_sb_setup(struct i40e_pf *pf)
 
 	/* find existing VSI and see if it needs configuring */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			vsi = pf->vsi[i];
 			break;
@@ -5451,7 +5451,7 @@ static void i40e_fdir_teardown(struct i40e_pf *pf)
 	int i;
 
 	i40e_fdir_filter_exit(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			i40e_vsi_release(pf->vsi[i]);
 			break;
@@ -5480,7 +5480,7 @@ static int i40e_prep_for_reset(struct i40e_pf *pf)
 	/* quiesce the VSIs and their queues that are not already DOWN */
 	i40e_pf_quiesce_all_vsi(pf);
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			pf->vsi[v]->seid = 0;
 	}
@@ -5960,15 +5960,15 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
 	 * find next empty vsi slot, looping back around if necessary
 	 */
 	i = pf->next_vsi;
-	while (i < pf->hw.func_caps.num_vsis && pf->vsi[i])
+	while (i < pf->num_alloc_vsi && pf->vsi[i])
 		i++;
-	if (i >= pf->hw.func_caps.num_vsis) {
+	if (i >= pf->num_alloc_vsi) {
 		i = 0;
 		while (i < pf->next_vsi && pf->vsi[i])
 			i++;
 	}
 
-	if (i < pf->hw.func_caps.num_vsis && !pf->vsi[i]) {
+	if (i < pf->num_alloc_vsi && !pf->vsi[i]) {
 		vsi_idx = i;             /* Found one! */
 	} else {
 		ret = -ENODEV;
@@ -7229,7 +7229,7 @@ int i40e_vsi_release(struct i40e_vsi *vsi)
 	 * the orphan VEBs yet.  We'll wait for an explicit remove request
 	 * from up the network stack.
 	 */
-	for (n = 0, i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (n = 0, i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] &&
 		    pf->vsi[i]->uplink_seid == uplink_seid &&
 		    (pf->vsi[i]->flags & I40E_VSI_FLAG_VEB_OWNER) == 0) {
@@ -7408,7 +7408,7 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 
 	if (!veb && uplink_seid != pf->mac_seid) {
 
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+		for (i = 0; i < pf->num_alloc_vsi; i++) {
 			if (pf->vsi[i] && pf->vsi[i]->seid == uplink_seid) {
 				vsi = pf->vsi[i];
 				break;
@@ -7651,7 +7651,7 @@ static void i40e_switch_branch_release(struct i40e_veb *branch)
 	 * NOTE: Removing the last VSI on a VEB has the SIDE EFFECT of removing
 	 *       the VEB itself, so don't use (*branch) after this loop.
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (!pf->vsi[i])
 			continue;
 		if (pf->vsi[i]->uplink_seid == branch_seid &&
@@ -7703,7 +7703,7 @@ void i40e_veb_release(struct i40e_veb *veb)
 	pf = veb->pf;
 
 	/* find the remaining VSI and check for extras */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->uplink_seid == veb->seid) {
 			n++;
 			vsi = pf->vsi[i];
@@ -7815,10 +7815,10 @@ struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags,
 	}
 
 	/* make sure there is such a vsi and uplink */
-	for (vsi_idx = 0; vsi_idx < pf->hw.func_caps.num_vsis; vsi_idx++)
+	for (vsi_idx = 0; vsi_idx < pf->num_alloc_vsi; vsi_idx++)
 		if (pf->vsi[vsi_idx] && pf->vsi[vsi_idx]->seid == vsi_seid)
 			break;
-	if (vsi_idx >= pf->hw.func_caps.num_vsis && vsi_seid != 0) {
+	if (vsi_idx >= pf->num_alloc_vsi && vsi_seid != 0) {
 		dev_info(&pf->pdev->dev, "vsi seid %d not found\n",
 			 vsi_seid);
 		return NULL;
@@ -8484,10 +8484,18 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i40e_determine_queue_usage(pf);
 	i40e_init_interrupt_scheme(pf);
 
-	/* Set up the *vsi struct based on the number of VSIs in the HW,
-	 * and set up our local tracking of the MAIN PF vsi.
+	/* The number of VSIs reported by the FW is the minimum guaranteed
+	 * to us; HW supports far more and we share the remaining pool with
+	 * the other PFs. We allocate space for more than the guarantee with
+	 * the understanding that we might not get them all later.
 	 */
-	len = sizeof(struct i40e_vsi *) * pf->hw.func_caps.num_vsis;
+	if (pf->hw.func_caps.num_vsis < I40E_MIN_VSI_ALLOC)
+		pf->num_alloc_vsi = I40E_MIN_VSI_ALLOC;
+	else
+		pf->num_alloc_vsi = pf->hw.func_caps.num_vsis;
+
+	/* Set up the *vsi struct and our local tracking of the MAIN PF vsi. */
+	len = sizeof(struct i40e_vsi *) * pf->num_alloc_vsi;
 	pf->vsi = kzalloc(len, GFP_KERNEL);
 	if (!pf->vsi) {
 		err = -ENOMEM;
@@ -8500,7 +8508,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_vsis;
 	}
 	/* if FDIR VSI was set up, start it now */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			i40e_vsi_open(pf->vsi[i]);
 			break;
@@ -8695,7 +8703,7 @@ static void i40e_remove(struct pci_dev *pdev)
 
 	/* Clear all dynamic memory lists of rings, q_vectors, and VSIs */
 	i40e_clear_interrupt_scheme(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i]) {
 			i40e_vsi_clear_rings(pf->vsi[i]);
 			i40e_vsi_clear(pf->vsi[i]);

drivers/net/ethernet/intel/i40e/i40e_txrx.c

@@ -62,7 +62,7 @@ int i40e_program_fdir_filter(struct i40e_fdir_filter *fdir_data, u8 *raw_packet,
 
 	/* find existing FDIR VSI */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR)
 			vsi = pf->vsi[i];
 	if (!vsi)