@@ -230,7 +230,7 @@ static void irdma_fill_device_info(struct irdma_device *iwdev, struct ice_pf *pf
rf->gen_ops.unregister_qset = irdma_lan_unregister_qset;
rf->hw.hw_addr = pf->hw.hw_addr;
rf->pcidev = pf->pdev;
- rf->msix_count = pf->num_rdma_msix;
+ rf->msix_count = pf->msix.rdma;
rf->pf_id = pf->hw.pf_id;
rf->msix_entries = &pf->msix_entries[pf->rdma_base_vector];
rf->default_vsi.vsi_idx = vsi->vsi_num;
@@ -518,6 +518,14 @@ struct ice_agg_node {
u8 valid;
};
+struct ice_msix {
+ u16 all; /* All reserved MSI-X */
+ u16 misc; /* MSI-X reserved for misc */
+ u16 eth; /* MSI-X reserved for eth */
+ u16 rdma; /* MSI-X reserved for RDMA */
+ u16 vf; /* MSI-X reserved for VF */
+};
+
struct ice_pf {
struct pci_dev *pdev;
@@ -531,6 +539,10 @@ struct ice_pf {
/* OS reserved IRQ details */
struct msix_entry *msix_entries;
struct ice_res_tracker *irq_tracker;
+ /* Store MSI-X requested by user via devlink */
+ struct ice_msix req_msix;
+ /* Store currently used MSI-X */
+ struct ice_msix msix;
/* First MSIX vector used by SR-IOV VFs. Calculated by subtracting the
* number of MSIX vectors needed for all SR-IOV VFs from the number of
* MSIX vectors allowed on this PF.
@@ -561,7 +573,6 @@ struct ice_pf {
struct tty_driver *ice_gnss_tty_driver;
struct tty_port *gnss_tty_port[ICE_GNSS_TTY_MINOR_DEVICES];
struct gnss_serial *gnss_serial[ICE_GNSS_TTY_MINOR_DEVICES];
- u16 num_rdma_msix; /* Total MSIX vectors for RDMA driver */
u16 rdma_base_vector;
/* spinlock to protect the AdminQ wait list */
@@ -578,7 +589,6 @@ struct ice_pf {
u16 num_avail_sw_msix; /* remaining MSIX SW vectors left unclaimed */
u16 max_pf_txqs; /* Total Tx queues PF wide */
u16 max_pf_rxqs; /* Total Rx queues PF wide */
- u16 num_lan_msix; /* Total MSIX vectors for base driver */
u16 num_lan_tx; /* num LAN Tx queues setup */
u16 num_lan_rx; /* num LAN Rx queues setup */
u16 next_vsi; /* Next free slot in pf->vsi[] - 0-based! */
@@ -934,7 +944,7 @@ void ice_unload(struct ice_pf *pf);
*/
static inline void ice_set_rdma_cap(struct ice_pf *pf)
{
- if (pf->hw.func_caps.common_cap.rdma && pf->num_rdma_msix) {
+ if (pf->hw.func_caps.common_cap.rdma && pf->msix.rdma) {
set_bit(ICE_FLAG_RDMA_ENA, pf->flags);
set_bit(ICE_FLAG_PLUG_AUX_DEV, pf->flags);
}
@@ -3513,7 +3513,7 @@ ice_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
*/
static int ice_get_max_txq(struct ice_pf *pf)
{
- return min3(pf->num_lan_msix, (u16)num_online_cpus(),
+ return min3(pf->msix.eth, (u16)num_online_cpus(),
(u16)pf->hw.func_caps.common_cap.num_txq);
}
@@ -3523,7 +3523,7 @@ static int ice_get_max_txq(struct ice_pf *pf)
*/
static int ice_get_max_rxq(struct ice_pf *pf)
{
- return min3(pf->num_lan_msix, (u16)num_online_cpus(),
+ return min3(pf->msix.eth, (u16)num_online_cpus(),
(u16)pf->hw.func_caps.common_cap.num_rxq);
}
@@ -237,11 +237,11 @@ static int ice_reserve_rdma_qvector(struct ice_pf *pf)
if (ice_is_rdma_ena(pf)) {
int index;
- index = ice_get_res(pf, pf->irq_tracker, pf->num_rdma_msix,
+ index = ice_get_res(pf, pf->irq_tracker, pf->msix.rdma,
ICE_RES_RDMA_VEC_ID);
if (index < 0)
return index;
- pf->num_avail_sw_msix -= pf->num_rdma_msix;
+ pf->num_avail_sw_msix -= pf->msix.rdma;
pf->rdma_base_vector = (u16)index;
}
return 0;
@@ -253,7 +253,7 @@ static int ice_reserve_rdma_qvector(struct ice_pf *pf)
*/
static void ice_free_rdma_qvector(struct ice_pf *pf)
{
- pf->num_avail_sw_msix -= pf->num_rdma_msix;
+ pf->num_avail_sw_msix -= pf->msix.rdma;
ice_free_res(pf->irq_tracker, pf->rdma_base_vector,
ICE_RES_RDMA_VEC_ID);
}
@@ -184,7 +184,7 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, struct ice_vf *vf)
vsi->alloc_txq = vsi->req_txq;
vsi->num_txq = vsi->req_txq;
} else {
- vsi->alloc_txq = min3(pf->num_lan_msix,
+ vsi->alloc_txq = min3(pf->msix.eth,
ice_get_avail_txq_count(pf),
(u16)num_online_cpus());
}
@@ -199,7 +199,7 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, struct ice_vf *vf)
vsi->alloc_rxq = vsi->req_rxq;
vsi->num_rxq = vsi->req_rxq;
} else {
- vsi->alloc_rxq = min3(pf->num_lan_msix,
+ vsi->alloc_rxq = min3(pf->msix.eth,
ice_get_avail_rxq_count(pf),
(u16)num_online_cpus());
}
@@ -207,7 +207,7 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, struct ice_vf *vf)
pf->num_lan_rx = vsi->alloc_rxq;
- vsi->num_q_vectors = min_t(int, pf->num_lan_msix,
+ vsi->num_q_vectors = min_t(int, pf->msix.eth,
max_t(int, vsi->alloc_rxq,
vsi->alloc_txq));
break;
@@ -1100,7 +1100,7 @@ ice_chnl_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt)
u8 offset = 0;
int pow;
- qcount = min_t(int, vsi->num_rxq, pf->num_lan_msix);
+ qcount = min_t(int, vsi->num_rxq, pf->msix.eth);
pow = order_base_2(qcount);
qmap = ((offset << ICE_AQ_VSI_TC_Q_OFFSET_S) &
@@ -3818,119 +3818,193 @@ static int ice_init_pf(struct ice_pf *pf)
}
/**
- * ice_reduce_msix_usage - Reduce usage of MSI-X vectors
+ * ice_distribute_msix - distribute amount of MSI-X across all features
* @pf: board private structure
- * @v_remain: number of remaining MSI-X vectors to be distributed
+ * @msix: structure that stores MSI-X amount for all features
+ * @num_msix: amount of MSI-X to distribute
+ * @req_lan: amount of requested LAN MSI-X
+ * @req_rdma: amount of requested RDMA MSI-X
*
- * Reduce the usage of MSI-X vectors when entire request cannot be fulfilled.
- * pf->num_lan_msix and pf->num_rdma_msix values are set based on number of
- * remaining vectors.
+ * The function is trying to distribute MSI-X across all supported features.
+ * As a side effect the function can turn off not critical features if the
+ * number of msix is too low.
+ *
+ * Function returns -ERANGE if the amount of msix is too low to support
+ * basic driver operations. It is critical error.
+ *
+ * In rest cases function returns number of used MSI-X in total,
+ *
+ * First try to distribute minimum amount of MSI-X across all features. If
+ * there still are remaing MSI-X distribute it equally.
+ *
+ * Function can be called to distribute MSI-X amount received from hardware
+ * or from kernel. To support both cases MSI-X for VF isn't set here. This
+ * should be set outside this function based on returned value.
*/
-static void ice_reduce_msix_usage(struct ice_pf *pf, int v_remain)
+static int
+ice_distribute_msix(struct ice_pf *pf, struct ice_msix *msix, int num_msix,
+ int req_lan, int req_rdma)
{
- int v_rdma;
+ struct device *dev = ice_pf_to_dev(pf);
+ int rem_msix = num_msix;
+ int additional_msix;
- if (!ice_is_rdma_ena(pf)) {
- pf->num_lan_msix = v_remain;
- return;
+ msix->all = num_msix;
+ msix->eth = 0;
+ msix->rdma = 0;
+ msix->misc = 0;
+
+ if (num_msix < ICE_MIN_MSIX) {
+ dev_warn(dev, "not enough MSI-X vectors. wanted = %d, available = %d\n",
+ ICE_MIN_MSIX, num_msix);
+ return -ERANGE;
}
- /* RDMA needs at least 1 interrupt in addition to AEQ MSIX */
- v_rdma = ICE_RDMA_NUM_AEQ_MSIX + 1;
+ msix->misc = ICE_MIN_LAN_OICR_MSIX;
+ /* Minimum LAN */
+ msix->eth = ICE_MIN_LAN_TXRX_MSIX;
+ req_lan -= ICE_MIN_LAN_TXRX_MSIX;
+ rem_msix = rem_msix - msix->misc - msix->eth;
+ additional_msix = req_lan;
- if (v_remain < ICE_MIN_LAN_TXRX_MSIX + ICE_MIN_RDMA_MSIX) {
- dev_warn(ice_pf_to_dev(pf), "Not enough MSI-X vectors to support RDMA.\n");
- clear_bit(ICE_FLAG_RDMA_ENA, pf->flags);
+ /* MSI-X distribution for misc */
+ if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) {
+ if (rem_msix >= ICE_FDIR_MSIX) {
+ msix->misc += ICE_FDIR_MSIX;
+ rem_msix -= ICE_FDIR_MSIX;
+ } else {
+ dev_warn(dev, "not enough MSI-X vectors. wanted = %d, available = %d\n",
+ ICE_FDIR_MSIX, rem_msix);
+ dev_warn(dev, "turning flow director off\n");
+ clear_bit(ICE_FLAG_FD_ENA, pf->flags);
+ goto exit;
+ }
+ }
+
+ if (test_bit(ICE_FLAG_ESWITCH_CAPABLE, pf->flags)) {
+ if (rem_msix >= ICE_ESWITCH_MSIX) {
+ msix->misc += ICE_ESWITCH_MSIX;
+ rem_msix -= ICE_ESWITCH_MSIX;
+ } else {
+ dev_warn(dev, "not enough MSI-X vectors. wanted = %d, available = %d\n",
+ ICE_ESWITCH_MSIX, rem_msix);
+ dev_warn(dev, "turning eswitch off\n");
+ clear_bit(ICE_FLAG_ESWITCH_CAPABLE, pf->flags);
+ goto exit;
+ }
+ }
- pf->num_rdma_msix = 0;
- pf->num_lan_msix = ICE_MIN_LAN_TXRX_MSIX;
- } else if ((v_remain < ICE_MIN_LAN_TXRX_MSIX + v_rdma) ||
- (v_remain - v_rdma < v_rdma)) {
- /* Support minimum RDMA and give remaining vectors to LAN MSIX */
- pf->num_rdma_msix = ICE_MIN_RDMA_MSIX;
- pf->num_lan_msix = v_remain - ICE_MIN_RDMA_MSIX;
+ /* Minimum RDMA
+ * Minimum MSI-X for RDMA is ICE_MIN_RDMA_MSIX, but it is better to
+ * have ICE_RDMA_NUM_AEQ_MSIX + 1 for queue. Instead of starting
+ * with ICE_MIN_RDMA_MSIX and going to distribution for LAN, try
+ * to start with ICE_RDMA_NUM_AEQ_MSIX + 1 if there are MSI-X for it
+ */
+ if (ice_is_rdma_ena(pf)) {
+ if (rem_msix >= ICE_RDMA_NUM_AEQ_MSIX + 1 &&
+ req_rdma >= ICE_RDMA_NUM_AEQ_MSIX + 1) {
+ msix->rdma = ICE_RDMA_NUM_AEQ_MSIX + 1;
+ rem_msix -= msix->rdma;
+ req_rdma -= msix->rdma;
+ additional_msix += req_rdma;
+ } else if (rem_msix >= ICE_MIN_RDMA_MSIX) {
+ msix->rdma = ICE_MIN_RDMA_MSIX;
+ rem_msix -= ICE_MIN_RDMA_MSIX;
+ req_rdma -= ICE_MIN_RDMA_MSIX;
+ additional_msix += req_rdma;
+ } else {
+ dev_warn(dev, "not enough MSI-X vectors. wanted = %d, available = %d\n",
+ ICE_MIN_RDMA_MSIX, rem_msix);
+ dev_warn(dev, "turning RDMA off\n");
+ clear_bit(ICE_FLAG_RDMA_ENA, pf->flags);
+ goto exit;
+ }
} else {
- /* Split remaining MSIX with RDMA after accounting for AEQ MSIX
- */
- pf->num_rdma_msix = (v_remain - ICE_RDMA_NUM_AEQ_MSIX) / 2 +
- ICE_RDMA_NUM_AEQ_MSIX;
- pf->num_lan_msix = v_remain - pf->num_rdma_msix;
+ req_rdma = 0;
}
-}
-/**
- * ice_ena_msix_range - Request a range of MSIX vectors from the OS
- * @pf: board private structure
- *
- * Compute the number of MSIX vectors wanted and request from the OS. Adjust
- * device usage if there are not enough vectors. Return the number of vectors
- * reserved or negative on failure.
- */
-static int ice_ena_msix_range(struct ice_pf *pf)
-{
- int num_cpus, hw_num_msix, v_other, v_wanted, v_actual;
- struct device *dev = ice_pf_to_dev(pf);
- int err, i;
+ /* Check if there is enough additional MSI-X for all used features */
+ if (rem_msix >= additional_msix) {
+ /* Best case scenario */
+ msix->eth += req_lan;
+ rem_msix -= req_lan;
- hw_num_msix = pf->hw.func_caps.common_cap.num_msix_vectors;
- num_cpus = num_online_cpus();
+ msix->rdma += req_rdma;
+ rem_msix -= req_rdma;
+ } else if (rem_msix) {
+ int factor = DIV_ROUND_UP(additional_msix, rem_msix);
+ int rdma = req_rdma / factor;
- /* LAN miscellaneous handler */
- v_other = ICE_MIN_LAN_OICR_MSIX;
+ dev_warn(dev, "not enough MSI-X vectors. requested = %d, obtained = %d\n",
+ additional_msix, rem_msix);
- /* Flow Director */
- if (test_bit(ICE_FLAG_FD_ENA, pf->flags))
- v_other += ICE_FDIR_MSIX;
+ msix->rdma += rdma;
+ rem_msix -= rdma;
- /* switchdev */
- if (test_bit(ICE_FLAG_ESWITCH_CAPABLE, pf->flags))
- v_other += ICE_ESWITCH_MSIX;
+ msix->eth += rem_msix;
+ rem_msix = 0;
- v_wanted = v_other;
+ dev_notice(dev, "enable %d MSI-X vectors for LAN and %d for RDMA\n",
+ msix->eth, msix->rdma);
+ }
- /* LAN traffic */
- pf->num_lan_msix = num_cpus;
- v_wanted += pf->num_lan_msix;
+ /* It is fine if rem_msix isn't 0 here, as it can be used for VF.
+ * It isn't done here as this function also cover MSI-X distribution
+ * for MSI-X amount returned from kernel
+ */
- /* RDMA auxiliary driver */
- if (ice_is_rdma_ena(pf)) {
- pf->num_rdma_msix = num_cpus + ICE_RDMA_NUM_AEQ_MSIX;
- v_wanted += pf->num_rdma_msix;
- }
+exit:
+ /* Return number of used MSI-X instead of remaing */
+ return num_msix - rem_msix;
+}
- if (v_wanted > hw_num_msix) {
- int v_remain;
+/**
+ * ice_set_def_msix - set default value for MSI-X based on hardware
+ * @pf: board private structure
+ *
+ * Function will return -ERANGE when there is critical error and
+ * minimum amount of MSI-X can't be used. In success returns 0.
+ */
+static int
+ice_set_def_msix(struct ice_pf *pf)
+{
+ int hw_msix = pf->hw.func_caps.common_cap.num_msix_vectors;
+ int num_cpus = num_present_cpus();
+ int used_msix;
- dev_warn(dev, "not enough device MSI-X vectors. wanted = %d, available = %d\n",
- v_wanted, hw_num_msix);
+ /* set default only if it wasn't previously set in devlink API */
+ if (pf->req_msix.eth)
+ return 0;
- if (hw_num_msix < ICE_MIN_MSIX) {
- err = -ERANGE;
- goto exit_err;
- }
+ used_msix = ice_distribute_msix(pf, &pf->req_msix, hw_msix, num_cpus,
+ num_cpus + ICE_RDMA_NUM_AEQ_MSIX);
- v_remain = hw_num_msix - v_other;
- if (v_remain < ICE_MIN_LAN_TXRX_MSIX) {
- v_other = ICE_MIN_MSIX - ICE_MIN_LAN_TXRX_MSIX;
- v_remain = ICE_MIN_LAN_TXRX_MSIX;
- }
+ if (used_msix < 0)
+ return -ERANGE;
- ice_reduce_msix_usage(pf, v_remain);
- v_wanted = pf->num_lan_msix + pf->num_rdma_msix + v_other;
+ pf->req_msix.vf = hw_msix - used_msix;
- dev_notice(dev, "Reducing request to %d MSI-X vectors for LAN traffic.\n",
- pf->num_lan_msix);
- if (ice_is_rdma_ena(pf))
- dev_notice(dev, "Reducing request to %d MSI-X vectors for RDMA.\n",
- pf->num_rdma_msix);
- }
+ return 0;
+}
+/**
+ * ice_ena_msix_range - Request a range of MSIX vectors from the OS
+ * @pf: board private structure
+ *
+ * Compute the number of MSIX vectors wanted and request from the OS. Adjust
+ * device usage if there are not enough vectors. Return the number of vectors
+ * reserved or negative on failure.
+ */
+static int ice_ena_msix_range(struct ice_pf *pf)
+{
+ struct device *dev = ice_pf_to_dev(pf);
+ int v_wanted, v_actual, err, i;
+
+ v_wanted = pf->req_msix.eth + pf->req_msix.rdma + pf->req_msix.misc;
pf->msix_entries = devm_kcalloc(dev, v_wanted,
sizeof(*pf->msix_entries), GFP_KERNEL);
- if (!pf->msix_entries) {
- err = -ENOMEM;
- goto exit_err;
- }
+ if (!pf->msix_entries)
+ return -ENOMEM;
for (i = 0; i < v_wanted; i++)
pf->msix_entries[i].entry = i;
@@ -3938,46 +4012,17 @@ static int ice_ena_msix_range(struct ice_pf *pf)
/* actually reserve the vectors */
v_actual = pci_enable_msix_range(pf->pdev, pf->msix_entries,
ICE_MIN_MSIX, v_wanted);
- if (v_actual < 0) {
- dev_err(dev, "unable to reserve MSI-X vectors\n");
- err = v_actual;
- goto msix_err;
- }
- if (v_actual < v_wanted) {
- dev_warn(dev, "not enough OS MSI-X vectors. requested = %d, obtained = %d\n",
- v_wanted, v_actual);
-
- if (v_actual < ICE_MIN_MSIX) {
- /* error if we can't get minimum vectors */
- pci_disable_msix(pf->pdev);
- err = -ERANGE;
- goto msix_err;
- } else {
- int v_remain = v_actual - v_other;
-
- if (v_remain < ICE_MIN_LAN_TXRX_MSIX)
- v_remain = ICE_MIN_LAN_TXRX_MSIX;
-
- ice_reduce_msix_usage(pf, v_remain);
-
- dev_notice(dev, "Enabled %d MSI-X vectors for LAN traffic.\n",
- pf->num_lan_msix);
-
- if (ice_is_rdma_ena(pf))
- dev_notice(dev, "Enabled %d MSI-X vectors for RDMA.\n",
- pf->num_rdma_msix);
- }
- }
+ v_actual = ice_distribute_msix(pf, &pf->msix, v_actual,
+ pf->req_msix.eth, pf->req_msix.rdma);
+ if (v_actual < 0)
+ goto msix_err;
return v_actual;
msix_err:
devm_kfree(dev, pf->msix_entries);
-exit_err:
- pf->num_rdma_msix = 0;
- pf->num_lan_msix = 0;
return err;
}
@@ -4014,6 +4059,9 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf)
{
int vectors;
+ if (ice_set_def_msix(pf))
+ return -ERANGE;
+
vectors = ice_ena_msix_range(pf);
if (vectors < 0)