* [PATCH for bnxt_re V4 08/21] RDMA/bnxt_re: Adding support for port related verbs
From: Selvin Xavier @ 2016-12-21 11:41 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
michael.chan-dY08KVG/lbpWk0Htik3J/w, Selvin Xavier, Eddie Wai,
Devesh Sharma, Somnath Kotur, Sriharsha Basavapatna
In-Reply-To: <1482320530-5344-1-git-send-email-selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Implements query_port, modify_port and port_immutable verbs
Signed-off-by: Eddie Wai <eddie.wai-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Devesh Sharma <devesh.sharma-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Somnath Kotur <somnath.kotur-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Selvin Xavier <selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
---
drivers/infiniband/hw/bnxt_re/ib_verbs.c | 122 +++++++++++++++++++++++++++++++
drivers/infiniband/hw/bnxt_re/ib_verbs.h | 7 ++
drivers/infiniband/hw/bnxt_re/main.c | 4 +
3 files changed, 133 insertions(+)
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 3c964d7..97172c9 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -150,6 +150,128 @@ int bnxt_re_modify_device(struct ib_device *ibdev,
return 0;
}
+static void __to_ib_speed_width(struct net_device *netdev, u8 *speed, u8 *width)
+{
+ struct ethtool_link_ksettings lksettings;
+ u32 espeed;
+
+ if (netdev->ethtool_ops && netdev->ethtool_ops->get_link_ksettings) {
+ memset(&lksettings, 0, sizeof(lksettings));
+ rtnl_lock();
+ netdev->ethtool_ops->get_link_ksettings(netdev, &lksettings);
+ rtnl_unlock();
+ espeed = lksettings.base.speed;
+ } else {
+ espeed = SPEED_UNKNOWN;
+ }
+ switch (espeed) {
+ case SPEED_1000:
+ *speed = IB_SPEED_SDR;
+ *width = IB_WIDTH_1X;
+ break;
+ case SPEED_10000:
+ *speed = IB_SPEED_QDR;
+ *width = IB_WIDTH_1X;
+ break;
+ case SPEED_20000:
+ *speed = IB_SPEED_DDR;
+ *width = IB_WIDTH_4X;
+ break;
+ case SPEED_25000:
+ *speed = IB_SPEED_EDR;
+ *width = IB_WIDTH_1X;
+ break;
+ case SPEED_40000:
+ *speed = IB_SPEED_QDR;
+ *width = IB_WIDTH_4X;
+ break;
+ case SPEED_50000:
+ break;
+ default:
+ *speed = IB_SPEED_SDR;
+ *width = IB_WIDTH_1X;
+ break;
+ }
+}
+
+/* Port */
+int bnxt_re_query_port(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_attr *port_attr)
+{
+ struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev);
+ struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr;
+
+ memset(port_attr, 0, sizeof(*port_attr));
+
+ if (netif_running(rdev->netdev) && netif_carrier_ok(rdev->netdev)) {
+ port_attr->state = IB_PORT_ACTIVE;
+ port_attr->phys_state = 5;
+ } else {
+ port_attr->state = IB_PORT_DOWN;
+ port_attr->phys_state = 3;
+ }
+ port_attr->max_mtu = IB_MTU_4096;
+ port_attr->active_mtu = iboe_get_mtu(rdev->netdev->mtu);
+ port_attr->gid_tbl_len = dev_attr->max_sgid;
+ port_attr->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_REINIT_SUP |
+ IB_PORT_DEVICE_MGMT_SUP |
+ IB_PORT_VENDOR_CLASS_SUP |
+ IB_PORT_IP_BASED_GIDS;
+
+ /* Max MSG size set to 2G for now */
+ port_attr->max_msg_sz = 0x80000000;
+ port_attr->bad_pkey_cntr = 0;
+ port_attr->qkey_viol_cntr = 0;
+ port_attr->pkey_tbl_len = dev_attr->max_pkey;
+ port_attr->lid = 0;
+ port_attr->sm_lid = 0;
+ port_attr->lmc = 0;
+ port_attr->max_vl_num = 4;
+ port_attr->sm_sl = 0;
+ port_attr->subnet_timeout = 0;
+ port_attr->init_type_reply = 0;
+ /* call the underlying netdev's ethtool hooks to query speed settings
+ * for which we acquire rtnl_lock _only_ if it's registered with
+ * IB stack to avoid race in the NETDEV_UNREG path
+ */
+ if (test_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags))
+ __to_ib_speed_width(rdev->netdev, &port_attr->active_speed,
+ &port_attr->active_width);
+ return 0;
+}
+
+int bnxt_re_modify_port(struct ib_device *ibdev, u8 port_num,
+ int port_modify_mask,
+ struct ib_port_modify *port_modify)
+{
+ switch (port_modify_mask) {
+ case IB_PORT_SHUTDOWN:
+ break;
+ case IB_PORT_INIT_TYPE:
+ break;
+ case IB_PORT_RESET_QKEY_CNTR:
+ break;
+ default:
+ break;
+ }
+ return 0;
+}
+
+int bnxt_re_get_port_immutable(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_immutable *immutable)
+{
+ struct ib_port_attr port_attr;
+
+ if (bnxt_re_query_port(ibdev, port_num, &port_attr))
+ return -EINVAL;
+
+ immutable->pkey_tbl_len = port_attr.pkey_tbl_len;
+ immutable->gid_tbl_len = port_attr.gid_tbl_len;
+ immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+ immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
+ immutable->max_mad_size = IB_MGMT_MAD_SIZE;
+ return 0;
+}
/* Protection Domains */
int bnxt_re_dealloc_pd(struct ib_pd *ib_pd)
{
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
index 45f9253..255a0ad 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
@@ -60,6 +60,13 @@ int bnxt_re_query_device(struct ib_device *ibdev,
int bnxt_re_modify_device(struct ib_device *ibdev,
int device_modify_mask,
struct ib_device_modify *device_modify);
+int bnxt_re_query_port(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_attr *port_attr);
+int bnxt_re_modify_port(struct ib_device *ibdev, u8 port_num,
+ int port_modify_mask,
+ struct ib_port_modify *port_modify);
+int bnxt_re_get_port_immutable(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_immutable *immutable);
struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev,
struct ib_ucontext *context,
struct ib_udata *udata);
diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
index 038279b..e7942a6 100644
--- a/drivers/infiniband/hw/bnxt_re/main.c
+++ b/drivers/infiniband/hw/bnxt_re/main.c
@@ -438,6 +438,10 @@ static int bnxt_re_register_ib(struct bnxt_re_dev *rdev)
ibdev->dma_device = &rdev->en_dev->pdev->dev;
ibdev->query_device = bnxt_re_query_device;
ibdev->modify_device = bnxt_re_modify_device;
+
+ ibdev->query_port = bnxt_re_query_port;
+ ibdev->modify_port = bnxt_re_modify_port;
+ ibdev->get_port_immutable = bnxt_re_get_port_immutable;
ibdev->alloc_pd = bnxt_re_alloc_pd;
ibdev->dealloc_pd = bnxt_re_dealloc_pd;
ibdev->alloc_ucontext = bnxt_re_alloc_ucontext;
--
2.5.5
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH for bnxt_re V4 05/21] RDMA/bnxt_re: Adding Notification Queue support
From: Selvin Xavier @ 2016-12-21 11:41 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
michael.chan-dY08KVG/lbpWk0Htik3J/w, Selvin Xavier, Eddie Wai,
Devesh Sharma, Somnath Kotur, Sriharsha Basavapatna
In-Reply-To: <1482320530-5344-1-git-send-email-selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Completion Notifcations are handled by Notification Queue (NQ). This
patch configures the NQs. Also, configures the Door bell page mapping
v3: Fixes some sparse warnings related to endianness checks
v4: Change include file names
Signed-off-by: Eddie Wai <eddie.wai-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Devesh Sharma <devesh.sharma-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Somnath Kotur <somnath.kotur-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Selvin Xavier <selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
---
drivers/infiniband/hw/bnxt_re/bnxt_re.h | 8 ++
drivers/infiniband/hw/bnxt_re/main.c | 52 +++++++++-
drivers/infiniband/hw/bnxt_re/qplib_fp.c | 161 ++++++++++++++++++++++++++++++
drivers/infiniband/hw/bnxt_re/qplib_fp.h | 60 +++++++++++
drivers/infiniband/hw/bnxt_re/qplib_res.h | 6 ++
5 files changed, 286 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
index 78e95a5..8d2f42f 100644
--- a/drivers/infiniband/hw/bnxt_re/bnxt_re.h
+++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
@@ -60,6 +60,8 @@ struct bnxt_re_work {
#define BNXT_RE_MIN_MSIX 2
#define BNXT_RE_MAX_MSIX 16
#define BNXT_RE_AEQ_IDX 0
+#define BNXT_RE_NQ_IDX 1
+
struct bnxt_re_dev {
struct ib_device ibdev;
struct list_head list;
@@ -78,9 +80,15 @@ struct bnxt_re_dev {
int id;
+ /* FP Notification Queue (CQ & SRQ) */
+ struct tasklet_struct nq_task;
+
/* RCFW Channel */
struct bnxt_qplib_rcfw rcfw;
+ /* NQ */
+ struct bnxt_qplib_nq nq;
+
/* Device Resources */
struct bnxt_qplib_dev_attr dev_attr;
struct bnxt_qplib_ctx qplib_ctx;
diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
index a8d0f9b..bef654e 100644
--- a/drivers/infiniband/hw/bnxt_re/main.c
+++ b/drivers/infiniband/hw/bnxt_re/main.c
@@ -568,6 +568,9 @@ static int bnxt_re_aeq_handler(struct bnxt_qplib_rcfw *rcfw,
static void bnxt_re_cleanup_res(struct bnxt_re_dev *rdev)
{
+ if (rdev->nq.hwq.max_elements)
+ bnxt_qplib_disable_nq(&rdev->nq);
+
if (rdev->qplib_res.rcfw)
bnxt_qplib_cleanup_res(&rdev->qplib_res);
}
@@ -578,11 +581,32 @@ static int bnxt_re_init_res(struct bnxt_re_dev *rdev)
bnxt_qplib_init_res(&rdev->qplib_res);
+ if (rdev->msix_entries[BNXT_RE_NQ_IDX].vector <= 0)
+ return -EINVAL;
+
+ rc = bnxt_qplib_enable_nq(rdev->en_dev->pdev, &rdev->nq,
+ rdev->msix_entries[BNXT_RE_NQ_IDX].vector,
+ rdev->msix_entries[BNXT_RE_NQ_IDX].db_offset,
+ NULL,
+ NULL);
+
+ if (rc)
+ dev_err(rdev_to_dev(rdev), "Failed to enable NQ: %#x", rc);
+
return rc;
}
static void bnxt_re_free_res(struct bnxt_re_dev *rdev, bool lock_wait)
{
+ if (rdev->nq.hwq.max_elements) {
+ bnxt_re_net_ring_free(rdev, rdev->nq.ring_id, lock_wait);
+ bnxt_qplib_free_nq(&rdev->nq);
+ }
+ if (rdev->qplib_res.dpi_tbl.max) {
+ bnxt_qplib_dealloc_dpi(&rdev->qplib_res,
+ &rdev->qplib_res.dpi_tbl,
+ &rdev->dpi_privileged);
+ }
if (rdev->qplib_res.rcfw) {
bnxt_qplib_free_res(&rdev->qplib_res);
rdev->qplib_res.rcfw = NULL;
@@ -604,8 +628,34 @@ static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
if (rc)
goto fail;
- return 0;
+ rc = bnxt_qplib_alloc_dpi(&rdev->qplib_res.dpi_tbl,
+ &rdev->dpi_privileged,
+ rdev);
+ if (rc)
+ goto fail;
+ rdev->nq.hwq.max_elements = BNXT_RE_MAX_CQ_COUNT +
+ BNXT_RE_MAX_SRQC_COUNT + 2;
+ rc = bnxt_qplib_alloc_nq(rdev->en_dev->pdev, &rdev->nq);
+ if (rc) {
+ dev_err(rdev_to_dev(rdev),
+ "Failed to allocate NQ memory: %#x", rc);
+ goto fail;
+ }
+ rc = bnxt_re_net_ring_alloc
+ (rdev, rdev->nq.hwq.pbl[PBL_LVL_0].pg_map_arr,
+ rdev->nq.hwq.pbl[rdev->nq.hwq.level].pg_count,
+ HWRM_RING_ALLOC_CMPL, BNXT_QPLIB_NQE_MAX_CNT - 1,
+ rdev->msix_entries[BNXT_RE_NQ_IDX].ring_idx,
+ &rdev->nq.ring_id);
+ if (rc) {
+ dev_err(rdev_to_dev(rdev),
+ "Failed to allocate NQ ring: %#x", rc);
+ goto free_nq;
+ }
+ return 0;
+free_nq:
+ bnxt_qplib_free_nq(&rdev->nq);
fail:
rdev->qplib_res.rcfw = NULL;
return rc;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
index 36c4b81..a33960b 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
@@ -35,3 +35,164 @@
*
* Description: Fast Path Operators
*/
+
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <linux/prefetch.h>
+
+#include "roce_hsi.h"
+
+#include "qplib_res.h"
+#include "qplib_rcfw.h"
+#include "qplib_sp.h"
+#include "qplib_fp.h"
+
+static void bnxt_qplib_service_nq(unsigned long data)
+{
+ struct bnxt_qplib_nq *nq = (struct bnxt_qplib_nq *)data;
+ struct bnxt_qplib_hwq *hwq = &nq->hwq;
+ struct nq_base *nqe, **nq_ptr;
+ u32 sw_cons, raw_cons;
+ u16 type;
+ int budget = nq->budget;
+
+ /* Service the NQ until empty */
+ raw_cons = hwq->cons;
+ while (budget--) {
+ sw_cons = HWQ_CMP(raw_cons, hwq);
+ nq_ptr = (struct nq_base **)hwq->pbl_ptr;
+ nqe = &nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)];
+ if (!NQE_CMP_VALID(nqe, raw_cons, hwq->max_elements))
+ break;
+
+ type = le16_to_cpu(nqe->info10_type) & NQ_BASE_TYPE_MASK;
+ switch (type) {
+ case NQ_BASE_TYPE_CQ_NOTIFICATION:
+ break;
+ case NQ_BASE_TYPE_DBQ_EVENT:
+ break;
+ default:
+ dev_warn(&nq->pdev->dev,
+ "QPLIB: nqe with type = 0x%x not handled",
+ type);
+ break;
+ }
+ raw_cons++;
+ }
+ if (hwq->cons != raw_cons) {
+ hwq->cons = raw_cons;
+ NQ_DB_REARM(nq->bar_reg_iomem, hwq->cons, hwq->max_elements);
+ }
+}
+
+static irqreturn_t bnxt_qplib_nq_irq(int irq, void *dev_instance)
+{
+ struct bnxt_qplib_nq *nq = dev_instance;
+ struct bnxt_qplib_hwq *hwq = &nq->hwq;
+ struct nq_base **nq_ptr;
+ u32 sw_cons;
+
+ /* Prefetch the NQ element */
+ sw_cons = HWQ_CMP(hwq->cons, hwq);
+ nq_ptr = (struct nq_base **)nq->hwq.pbl_ptr;
+ prefetch(&nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)]);
+
+ /* Fan out to CPU affinitized kthreads? */
+ tasklet_schedule(&nq->worker);
+
+ return IRQ_HANDLED;
+}
+
+void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq)
+{
+ /* Make sure the HW is stopped! */
+ synchronize_irq(nq->vector);
+ tasklet_disable(&nq->worker);
+ tasklet_kill(&nq->worker);
+
+ if (nq->requested) {
+ free_irq(nq->vector, nq);
+ nq->requested = false;
+ }
+ if (nq->bar_reg_iomem)
+ iounmap(nq->bar_reg_iomem);
+ nq->bar_reg_iomem = NULL;
+
+ nq->cqn_handler = NULL;
+ nq->srqn_handler = NULL;
+ nq->vector = 0;
+}
+
+int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq,
+ int msix_vector, int bar_reg_offset,
+ int (*cqn_handler)(struct bnxt_qplib_nq *nq,
+ void *),
+ int (*srqn_handler)(struct bnxt_qplib_nq *nq,
+ void *, u8 event))
+{
+ resource_size_t nq_base;
+ int rc;
+
+ nq->pdev = pdev;
+ nq->vector = msix_vector;
+
+ nq->cqn_handler = cqn_handler;
+
+ nq->srqn_handler = srqn_handler;
+
+ tasklet_init(&nq->worker, bnxt_qplib_service_nq, (unsigned long)nq);
+
+ nq->requested = false;
+ rc = request_irq(nq->vector, bnxt_qplib_nq_irq, 0, "bnxt_qplib_nq", nq);
+ if (rc) {
+ dev_err(&nq->pdev->dev,
+ "Failed to request IRQ for NQ: %#x", rc);
+ bnxt_qplib_disable_nq(nq);
+ goto fail;
+ }
+ nq->requested = true;
+ nq->bar_reg = NQ_CONS_PCI_BAR_REGION;
+ nq->bar_reg_off = bar_reg_offset;
+ nq_base = pci_resource_start(pdev, nq->bar_reg);
+ if (!nq_base) {
+ rc = -ENOMEM;
+ goto fail;
+ }
+ nq->bar_reg_iomem = ioremap_nocache(nq_base + nq->bar_reg_off, 4);
+ if (!nq->bar_reg_iomem) {
+ rc = -ENOMEM;
+ goto fail;
+ }
+ NQ_DB_REARM(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements);
+
+ return 0;
+fail:
+ bnxt_qplib_disable_nq(nq);
+ return rc;
+}
+
+void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq)
+{
+ if (nq->hwq.max_elements)
+ bnxt_qplib_free_hwq(nq->pdev, &nq->hwq);
+}
+
+int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq)
+{
+ nq->pdev = pdev;
+ if (!nq->hwq.max_elements ||
+ nq->hwq.max_elements > BNXT_QPLIB_NQE_MAX_CNT)
+ nq->hwq.max_elements = BNXT_QPLIB_NQE_MAX_CNT;
+
+ if (bnxt_qplib_alloc_init_hwq(nq->pdev, &nq->hwq, NULL, 0,
+ &nq->hwq.max_elements,
+ BNXT_QPLIB_MAX_NQE_ENTRY_SIZE, 0,
+ PAGE_SIZE, HWQ_TYPE_L2_CMPL))
+ return -ENOMEM;
+
+ nq->budget = 8;
+ return 0;
+}
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
index 0983465..25a4494 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
@@ -39,4 +39,64 @@
#ifndef __BNXT_QPLIB_FP_H__
#define __BNXT_QPLIB_FP_H__
+#define BNXT_QPLIB_MAX_NQE_ENTRY_SIZE sizeof(struct nq_base)
+
+#define NQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_NQE_ENTRY_SIZE)
+#define NQE_MAX_IDX_PER_PG (NQE_CNT_PER_PG - 1)
+#define NQE_PG(x) (((x) & ~NQE_MAX_IDX_PER_PG) / NQE_CNT_PER_PG)
+#define NQE_IDX(x) ((x) & NQE_MAX_IDX_PER_PG)
+
+#define NQE_CMP_VALID(hdr, raw_cons, cp_bit) \
+ (!!(le32_to_cpu((hdr)->info63_v[0]) & NQ_BASE_V) == \
+ !((raw_cons) & (cp_bit)))
+
+#define BNXT_QPLIB_NQE_MAX_CNT (128 * 1024)
+
+#define NQ_CONS_PCI_BAR_REGION 2
+#define NQ_DB_KEY_CP (0x2 << CMPL_DOORBELL_KEY_SFT)
+#define NQ_DB_IDX_VALID CMPL_DOORBELL_IDX_VALID
+#define NQ_DB_IRQ_DIS CMPL_DOORBELL_MASK
+#define NQ_DB_CP_FLAGS_REARM (NQ_DB_KEY_CP | \
+ NQ_DB_IDX_VALID)
+#define NQ_DB_CP_FLAGS (NQ_DB_KEY_CP | \
+ NQ_DB_IDX_VALID | \
+ NQ_DB_IRQ_DIS)
+#define NQ_DB_REARM(db, raw_cons, cp_bit) \
+ writel(NQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db)
+#define NQ_DB(db, raw_cons, cp_bit) \
+ writel(NQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db)
+
+struct bnxt_qplib_nq {
+ struct pci_dev *pdev;
+
+ int vector;
+ int budget;
+ bool requested;
+ struct tasklet_struct worker;
+ struct bnxt_qplib_hwq hwq;
+
+ u16 bar_reg;
+ u16 bar_reg_off;
+ u16 ring_id;
+ void __iomem *bar_reg_iomem;
+
+ int (*cqn_handler)
+ (struct bnxt_qplib_nq *nq,
+ void *cq);
+ int (*srqn_handler)
+ (struct bnxt_qplib_nq *nq,
+ void *srq,
+ u8 event);
+};
+
+void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq);
+int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq,
+ int msix_vector, int bar_reg_offset,
+ int (*cqn_handler)(struct bnxt_qplib_nq *nq,
+ void *cq),
+ int (*srqn_handler)(struct bnxt_qplib_nq *nq,
+ void *srq,
+ u8 event));
+void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq);
+int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq);
#endif /* __BNXT_QPLIB_FP_H__ */
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h
index ce122cf..571feda 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h
@@ -193,6 +193,12 @@ int bnxt_qplib_alloc_init_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq,
struct scatterlist *sl, int nmap, u32 *elements,
u32 elements_per_page, u32 aux, u32 pg_size,
enum bnxt_qplib_hwq_type hwq_type);
+int bnxt_qplib_alloc_dpi(struct bnxt_qplib_dpi_tbl *dpit,
+ struct bnxt_qplib_dpi *dpi,
+ void *app);
+int bnxt_qplib_dealloc_dpi(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_dpi_tbl *dpi_tbl,
+ struct bnxt_qplib_dpi *dpi);
void bnxt_qplib_cleanup_res(struct bnxt_qplib_res *res);
int bnxt_qplib_init_res(struct bnxt_qplib_res *res);
void bnxt_qplib_free_res(struct bnxt_qplib_res *res);
--
2.5.5
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH for bnxt_re V4 04/21] RDMA/bnxt_re: Enabling RoCE control path
From: Selvin Xavier @ 2016-12-21 11:41 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
michael.chan-dY08KVG/lbpWk0Htik3J/w, Selvin Xavier, Eddie Wai,
Devesh Sharma, Somnath Kotur, Sriharsha Basavapatna
In-Reply-To: <1482320530-5344-1-git-send-email-selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
This patch covers the basic initialization of the HW interface.
Implements some of the slow path FW commands required for the
HW intialization. It also handles registration with the IB stack.
v2: Fix some of the sparse warnings
v3: Removed some more smatch and sparse warnings related to endianness.
Fixes cross compilation failure warnings. Also, fixes the retry logic
to avoid system hangs in case of delay or no response for the FW commands.
v4: Changes header file names
Signed-off-by: Eddie Wai <eddie.wai-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Devesh Sharma <devesh.sharma-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Somnath Kotur <somnath.kotur-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
Signed-off-by: Selvin Xavier <selvin.xavier-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
---
drivers/infiniband/hw/bnxt_re/bnxt_re.h | 15 +
drivers/infiniband/hw/bnxt_re/main.c | 428 ++++++++++++++++-
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c | 608 ++++++++++++++++++++++++
drivers/infiniband/hw/bnxt_re/qplib_rcfw.h | 189 ++++++++
drivers/infiniband/hw/bnxt_re/qplib_res.c | 738 +++++++++++++++++++++++++++++
drivers/infiniband/hw/bnxt_re/qplib_res.h | 165 +++++++
drivers/infiniband/hw/bnxt_re/qplib_sp.c | 163 +++++++
drivers/infiniband/hw/bnxt_re/qplib_sp.h | 42 ++
8 files changed, 2343 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
index 8b73e3d..78e95a5 100644
--- a/drivers/infiniband/hw/bnxt_re/bnxt_re.h
+++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
@@ -45,6 +45,11 @@
#define BNXT_RE_REF_WAIT_COUNT 10
#define BNXT_RE_DESC "Broadcom NetXtreme-C/E RoCE Driver"
+#define BNXT_RE_MAX_QPC_COUNT (64 * 1024)
+#define BNXT_RE_MAX_MRW_COUNT (64 * 1024)
+#define BNXT_RE_MAX_SRQC_COUNT (64 * 1024)
+#define BNXT_RE_MAX_CQ_COUNT (64 * 1024)
+
struct bnxt_re_work {
struct work_struct work;
unsigned long event;
@@ -54,6 +59,7 @@ struct bnxt_re_work {
#define BNXT_RE_MIN_MSIX 2
#define BNXT_RE_MAX_MSIX 16
+#define BNXT_RE_AEQ_IDX 0
struct bnxt_re_dev {
struct ib_device ibdev;
struct list_head list;
@@ -72,6 +78,15 @@ struct bnxt_re_dev {
int id;
+ /* RCFW Channel */
+ struct bnxt_qplib_rcfw rcfw;
+
+ /* Device Resources */
+ struct bnxt_qplib_dev_attr dev_attr;
+ struct bnxt_qplib_ctx qplib_ctx;
+ struct bnxt_qplib_res qplib_res;
+ struct bnxt_qplib_dpi dpi_privileged;
+
atomic_t qp_count;
struct mutex qp_lock; /* protect qp list */
struct list_head qp_list;
diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
index 71d7501..a8d0f9b 100644
--- a/drivers/infiniband/hw/bnxt_re/main.c
+++ b/drivers/infiniband/hw/bnxt_re/main.c
@@ -54,6 +54,10 @@
#include "bnxt_ulp.h"
#include "roce_hsi.h"
+#include "qplib_res.h"
+#include "qplib_sp.h"
+#include "qplib_fp.h"
+#include "qplib_rcfw.h"
#include "bnxt_re.h"
#include "bnxt.h"
static char version[] =
@@ -192,6 +196,160 @@ static int bnxt_re_request_msix(struct bnxt_re_dev *rdev)
return rc;
}
+static void bnxt_re_init_hwrm_hdr(struct bnxt_re_dev *rdev, struct input *hdr,
+ u16 opcd, u16 crid, u16 trid)
+{
+ hdr->req_type = cpu_to_le16(opcd);
+ hdr->cmpl_ring = cpu_to_le16(crid);
+ hdr->target_id = cpu_to_le16(trid);
+}
+
+static void bnxt_re_fill_fw_msg(struct bnxt_fw_msg *fw_msg, void *msg,
+ int msg_len, void *resp, int resp_max_len,
+ int timeout)
+{
+ fw_msg->msg = msg;
+ fw_msg->msg_len = msg_len;
+ fw_msg->resp = resp;
+ fw_msg->resp_max_len = resp_max_len;
+ fw_msg->timeout = timeout;
+}
+
+static int bnxt_re_net_ring_free(struct bnxt_re_dev *rdev, u16 fw_ring_id,
+ bool lock_wait)
+{
+ struct bnxt_en_dev *en_dev = rdev->en_dev;
+ struct hwrm_ring_free_input req = {0};
+ struct hwrm_ring_free_output resp;
+ struct bnxt_fw_msg fw_msg;
+ bool do_unlock = false;
+ int rc = -EINVAL;
+
+ if (!en_dev)
+ return rc;
+
+ memset(&fw_msg, 0, sizeof(fw_msg));
+ if (lock_wait) {
+ rtnl_lock();
+ do_unlock = true;
+ }
+
+ bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_RING_FREE, -1, -1);
+ req.ring_type = RING_ALLOC_REQ_RING_TYPE_CMPL;
+ req.ring_id = cpu_to_le16(fw_ring_id);
+ bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp,
+ sizeof(resp), DFLT_HWRM_CMD_TIMEOUT);
+ rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg);
+ if (rc)
+ dev_err(rdev_to_dev(rdev),
+ "Failed to free HW ring:%d :%#x", req.ring_id, rc);
+ if (do_unlock)
+ rtnl_unlock();
+ return rc;
+}
+
+static int bnxt_re_net_ring_alloc(struct bnxt_re_dev *rdev, dma_addr_t *dma_arr,
+ int pages, int type, u32 ring_mask,
+ u32 map_index, u16 *fw_ring_id)
+{
+ struct bnxt_en_dev *en_dev = rdev->en_dev;
+ struct hwrm_ring_alloc_input req = {0};
+ struct hwrm_ring_alloc_output resp;
+ struct bnxt_fw_msg fw_msg;
+ int rc = -EINVAL;
+
+ if (!en_dev)
+ return rc;
+
+ memset(&fw_msg, 0, sizeof(fw_msg));
+ rtnl_lock();
+ bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_RING_ALLOC, -1, -1);
+ req.enables = 0;
+ req.page_tbl_addr = cpu_to_le64(dma_arr[0]);
+ if (pages > 1) {
+ /* Page size is in log2 units */
+ req.page_size = BNXT_PAGE_SHIFT;
+ req.page_tbl_depth = 1;
+ }
+ req.fbo = 0;
+ /* Association of ring index with doorbell index and MSIX number */
+ req.logical_id = cpu_to_le16(map_index);
+ req.length = cpu_to_le32(ring_mask + 1);
+ req.ring_type = RING_ALLOC_REQ_RING_TYPE_CMPL;
+ req.int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
+ bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp,
+ sizeof(resp), DFLT_HWRM_CMD_TIMEOUT);
+ rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg);
+ if (!rc)
+ *fw_ring_id = le16_to_cpu(resp.ring_id);
+
+ rtnl_unlock();
+ return rc;
+}
+
+static int bnxt_re_net_stats_ctx_free(struct bnxt_re_dev *rdev,
+ u32 fw_stats_ctx_id, bool lock_wait)
+{
+ struct bnxt_en_dev *en_dev = rdev->en_dev;
+ struct hwrm_stat_ctx_free_input req = {0};
+ struct bnxt_fw_msg fw_msg;
+ bool do_unlock = false;
+ int rc = -EINVAL;
+
+ if (!en_dev)
+ return rc;
+
+ memset(&fw_msg, 0, sizeof(fw_msg));
+ if (lock_wait) {
+ rtnl_lock();
+ do_unlock = true;
+ }
+
+ bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_STAT_CTX_FREE, -1, -1);
+ req.stat_ctx_id = cpu_to_le32(fw_stats_ctx_id);
+ bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&req,
+ sizeof(req), DFLT_HWRM_CMD_TIMEOUT);
+ rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg);
+ if (rc)
+ dev_err(rdev_to_dev(rdev),
+ "Failed to free HW stats context %#x", rc);
+
+ if (do_unlock)
+ rtnl_unlock();
+ return rc;
+}
+
+static int bnxt_re_net_stats_ctx_alloc(struct bnxt_re_dev *rdev,
+ dma_addr_t dma_map,
+ u32 *fw_stats_ctx_id)
+{
+ struct hwrm_stat_ctx_alloc_output resp = {0};
+ struct hwrm_stat_ctx_alloc_input req = {0};
+ struct bnxt_en_dev *en_dev = rdev->en_dev;
+ struct bnxt_fw_msg fw_msg;
+ int rc = -EINVAL;
+
+ *fw_stats_ctx_id = INVALID_STATS_CTX_ID;
+
+ if (!en_dev)
+ return rc;
+
+ memset(&fw_msg, 0, sizeof(fw_msg));
+ rtnl_lock();
+
+ bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_STAT_CTX_ALLOC, -1, -1);
+ req.update_period_ms = cpu_to_le32(1000);
+ req.stats_dma_addr = cpu_to_le64(dma_map);
+ bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp,
+ sizeof(resp), DFLT_HWRM_CMD_TIMEOUT);
+ rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg);
+ if (!rc)
+ *fw_stats_ctx_id = le32_to_cpu(resp.stat_ctx_id);
+
+ rtnl_unlock();
+ return rc;
+}
+
/* Device */
static bool is_bnxt_re_dev(struct net_device *netdev)
@@ -258,6 +416,62 @@ static struct bnxt_en_dev *bnxt_re_dev_probe(struct net_device *netdev)
return en_dev;
}
+static void bnxt_re_unregister_ib(struct bnxt_re_dev *rdev)
+{
+ ib_unregister_device(&rdev->ibdev);
+}
+
+static int bnxt_re_register_ib(struct bnxt_re_dev *rdev)
+{
+ struct ib_device *ibdev = &rdev->ibdev;
+
+ /* ib device init */
+ ibdev->owner = THIS_MODULE;
+ ibdev->node_type = RDMA_NODE_IB_CA;
+ strlcpy(ibdev->name, "bnxt_re%d", IB_DEVICE_NAME_MAX);
+ strlcpy(ibdev->node_desc, BNXT_RE_DESC " HCA",
+ strlen(BNXT_RE_DESC) + 5);
+ ibdev->phys_port_cnt = 1;
+
+ ibdev->num_comp_vectors = 1;
+ ibdev->dma_device = &rdev->en_dev->pdev->dev;
+ return ib_register_device(ibdev, NULL);
+}
+
+static ssize_t show_rev(struct device *device, struct device_attribute *attr,
+ char *buf)
+{
+ struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev);
+
+ return scnprintf(buf, PAGE_SIZE, "0x%x\n", rdev->en_dev->pdev->vendor);
+}
+
+static ssize_t show_fw_ver(struct device *device, struct device_attribute *attr,
+ char *buf)
+{
+ struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev);
+
+ return scnprintf(buf, PAGE_SIZE, "%s\n", rdev->dev_attr.fw_ver);
+}
+
+static ssize_t show_hca(struct device *device, struct device_attribute *attr,
+ char *buf)
+{
+ struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev);
+
+ return scnprintf(buf, PAGE_SIZE, "%s\n", rdev->ibdev.node_desc);
+}
+
+static DEVICE_ATTR(hw_rev, 0444, show_rev, NULL);
+static DEVICE_ATTR(fw_rev, 0444, show_fw_ver, NULL);
+static DEVICE_ATTR(hca_type, 0444, show_hca, NULL);
+
+static struct device_attribute *bnxt_re_attributes[] = {
+ &dev_attr_hw_rev,
+ &dev_attr_fw_rev,
+ &dev_attr_hca_type
+};
+
static void bnxt_re_dev_remove(struct bnxt_re_dev *rdev)
{
int i = BNXT_RE_REF_WAIT_COUNT;
@@ -320,10 +534,109 @@ static struct bnxt_re_dev *bnxt_re_dev_add(struct net_device *netdev,
return rdev;
}
+static int bnxt_re_aeq_handler(struct bnxt_qplib_rcfw *rcfw,
+ struct creq_func_event *aeqe)
+{
+ switch (aeqe->event) {
+ case CREQ_FUNC_EVENT_EVENT_TX_WQE_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TX_DATA_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_RX_WQE_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_RX_DATA_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CQ_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TQM_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCQ_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCS_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCC_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCM_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TIM_ERROR:
+ break;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static void bnxt_re_cleanup_res(struct bnxt_re_dev *rdev)
+{
+ if (rdev->qplib_res.rcfw)
+ bnxt_qplib_cleanup_res(&rdev->qplib_res);
+}
+
+static int bnxt_re_init_res(struct bnxt_re_dev *rdev)
+{
+ int rc = 0;
+
+ bnxt_qplib_init_res(&rdev->qplib_res);
+
+ return rc;
+}
+
+static void bnxt_re_free_res(struct bnxt_re_dev *rdev, bool lock_wait)
+{
+ if (rdev->qplib_res.rcfw) {
+ bnxt_qplib_free_res(&rdev->qplib_res);
+ rdev->qplib_res.rcfw = NULL;
+ }
+}
+
+static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
+{
+ int rc = 0;
+
+ /* Configure and allocate resources for qplib */
+ rdev->qplib_res.rcfw = &rdev->rcfw;
+ rc = bnxt_qplib_get_dev_attr(&rdev->rcfw, &rdev->dev_attr);
+ if (rc)
+ goto fail;
+
+ rc = bnxt_qplib_alloc_res(&rdev->qplib_res, rdev->en_dev->pdev,
+ rdev->netdev, &rdev->dev_attr);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ rdev->qplib_res.rcfw = NULL;
+ return rc;
+}
+
static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev, bool lock_wait)
{
- int rc;
+ int i, rc;
+
+ if (test_and_clear_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags)) {
+ for (i = 0; i < ARRAY_SIZE(bnxt_re_attributes); i++)
+ device_remove_file(&rdev->ibdev.dev,
+ bnxt_re_attributes[i]);
+ /* Cleanup ib dev */
+ bnxt_re_unregister_ib(rdev);
+ }
+ bnxt_re_cleanup_res(rdev);
+ bnxt_re_free_res(rdev, lock_wait);
+ if (test_and_clear_bit(BNXT_RE_FLAG_RCFW_CHANNEL_EN, &rdev->flags)) {
+ rc = bnxt_qplib_deinit_rcfw(&rdev->rcfw);
+ if (rc)
+ dev_warn(rdev_to_dev(rdev),
+ "Failed to deinitialize RCFW: %#x", rc);
+ bnxt_re_net_stats_ctx_free(rdev, rdev->qplib_ctx.stats.fw_id,
+ lock_wait);
+ bnxt_qplib_free_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx);
+ bnxt_qplib_disable_rcfw_channel(&rdev->rcfw);
+ bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, lock_wait);
+ bnxt_qplib_free_rcfw_channel(&rdev->rcfw);
+ }
if (test_and_clear_bit(BNXT_RE_FLAG_GOT_MSIX, &rdev->flags)) {
rc = bnxt_re_free_msix(rdev, lock_wait);
if (rc)
@@ -338,6 +651,19 @@ static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev, bool lock_wait)
}
}
+static void bnxt_re_set_resource_limits(struct bnxt_re_dev *rdev)
+{
+ u32 i;
+
+ rdev->qplib_ctx.qpc_count = BNXT_RE_MAX_QPC_COUNT;
+ rdev->qplib_ctx.mrw_count = BNXT_RE_MAX_MRW_COUNT;
+ rdev->qplib_ctx.srqc_count = BNXT_RE_MAX_SRQC_COUNT;
+ rdev->qplib_ctx.cq_count = BNXT_RE_MAX_CQ_COUNT;
+ for (i = 0; i < MAX_TQM_ALLOC_REQ; i++)
+ rdev->qplib_ctx.tqm_count[i] =
+ rdev->dev_attr.tqm_alloc_reqs[i];
+}
+
static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
{
int i, j, rc;
@@ -358,7 +684,103 @@ static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
}
set_bit(BNXT_RE_FLAG_GOT_MSIX, &rdev->flags);
+ /* Establish RCFW Communication Channel to initialize the context
+ * memory for the function and all child VFs
+ */
+ rc = bnxt_qplib_alloc_rcfw_channel(rdev->en_dev->pdev, &rdev->rcfw);
+ if (rc)
+ goto fail;
+
+ rc = bnxt_re_net_ring_alloc
+ (rdev, rdev->rcfw.creq.pbl[PBL_LVL_0].pg_map_arr,
+ rdev->rcfw.creq.pbl[rdev->rcfw.creq.level].pg_count,
+ HWRM_RING_ALLOC_CMPL, BNXT_QPLIB_CREQE_MAX_CNT - 1,
+ rdev->msix_entries[BNXT_RE_AEQ_IDX].ring_idx,
+ &rdev->rcfw.creq_ring_id);
+ if (rc) {
+ pr_err("Failed to allocate CREQ: %#x\n", rc);
+ goto free_rcfw;
+ }
+ rc = bnxt_qplib_enable_rcfw_channel
+ (rdev->en_dev->pdev, &rdev->rcfw,
+ rdev->msix_entries[BNXT_RE_AEQ_IDX].vector,
+ rdev->msix_entries[BNXT_RE_AEQ_IDX].db_offset,
+ 0, &bnxt_re_aeq_handler);
+ if (rc) {
+ pr_err("Failed to enable RCFW channel: %#x\n", rc);
+ goto free_ring;
+ }
+
+ rc = bnxt_qplib_get_dev_attr(&rdev->rcfw, &rdev->dev_attr);
+ if (rc)
+ goto disable_rcfw;
+ bnxt_re_set_resource_limits(rdev);
+
+ rc = bnxt_qplib_alloc_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx, 0);
+ if (rc) {
+ pr_err("Failed to allocate QPLIB context: %#x\n", rc);
+ goto disable_rcfw;
+ }
+ rc = bnxt_re_net_stats_ctx_alloc(rdev,
+ rdev->qplib_ctx.stats.dma_map,
+ &rdev->qplib_ctx.stats.fw_id);
+ if (rc) {
+ pr_err("Failed to allocate stats context: %#x\n", rc);
+ goto free_ctx;
+ }
+
+ rc = bnxt_qplib_init_rcfw(&rdev->rcfw, &rdev->qplib_ctx, 0);
+ if (rc) {
+ pr_err("Failed to initialize RCFW: %#x\n", rc);
+ goto free_sctx;
+ }
+ set_bit(BNXT_RE_FLAG_RCFW_CHANNEL_EN, &rdev->flags);
+
+ /* Resources based on the 'new' device caps */
+ rc = bnxt_re_alloc_res(rdev);
+ if (rc) {
+ pr_err("Failed to allocate resources: %#x\n", rc);
+ goto fail;
+ }
+ rc = bnxt_re_init_res(rdev);
+ if (rc) {
+ pr_err("Failed to initialize resources: %#x\n", rc);
+ goto fail;
+ }
+
+ /* Register ib dev */
+ rc = bnxt_re_register_ib(rdev);
+ if (rc) {
+ pr_err("Failed to register with IB: %#x\n", rc);
+ goto fail;
+ }
+ dev_info(rdev_to_dev(rdev), "Device registered successfully");
+ for (i = 0; i < ARRAY_SIZE(bnxt_re_attributes); i++) {
+ rc = device_create_file(&rdev->ibdev.dev,
+ bnxt_re_attributes[i]);
+ if (rc) {
+ dev_err(rdev_to_dev(rdev),
+ "Failed to create IB sysfs: %#x", rc);
+ /* Must clean up all created device files */
+ for (j = 0; j < i; j++)
+ device_remove_file(&rdev->ibdev.dev,
+ bnxt_re_attributes[j]);
+ bnxt_re_unregister_ib(rdev);
+ goto fail;
+ }
+ }
+ set_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags);
return 0;
+free_sctx:
+ bnxt_re_net_stats_ctx_free(rdev, rdev->qplib_ctx.stats.fw_id, true);
+free_ctx:
+ bnxt_qplib_free_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx);
+disable_rcfw:
+ bnxt_qplib_disable_rcfw_channel(&rdev->rcfw);
+free_ring:
+ bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, true);
+free_rcfw:
+ bnxt_qplib_free_rcfw_channel(&rdev->rcfw);
fail:
bnxt_re_ib_unreg(rdev, true);
return rc;
@@ -427,14 +849,10 @@ static void bnxt_re_task(struct work_struct *work)
"Failed to register with IB: %#x", rc);
break;
case NETDEV_UP:
-
break;
case NETDEV_DOWN:
-
break;
-
case NETDEV_CHANGE:
-
break;
default:
break;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
index 4029935..9144b5a 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
@@ -35,3 +35,611 @@
*
* Description: RDMA Controller HW interface
*/
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/pci.h>
+#include <linux/prefetch.h>
+#include "roce_hsi.h"
+#include "qplib_res.h"
+#include "qplib_rcfw.h"
+static void bnxt_qplib_service_creq(unsigned long data);
+
+/* Hardware communication channel */
+int bnxt_qplib_rcfw_wait_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie)
+{
+ u16 cbit;
+ int rc;
+
+ cookie &= RCFW_MAX_COOKIE_VALUE;
+ cbit = cookie % RCFW_MAX_OUTSTANDING_CMD;
+ if (!test_bit(cbit, rcfw->cmdq_bitmap))
+ dev_warn(&rcfw->pdev->dev,
+ "QPLIB: CMD bit %d for cookie 0x%x is not set?",
+ cbit, cookie);
+
+ rc = wait_event_timeout(rcfw->waitq,
+ !test_bit(cbit, rcfw->cmdq_bitmap),
+ msecs_to_jiffies(RCFW_CMD_WAIT_TIME_MS));
+ if (!rc) {
+ dev_warn(&rcfw->pdev->dev,
+ "QPLIB: Bono Error: timeout %d msec, msg {0x%x}\n",
+ RCFW_CMD_WAIT_TIME_MS, cookie);
+ }
+
+ return rc;
+};
+
+int bnxt_qplib_rcfw_block_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie)
+{
+ u32 count = -1;
+ u16 cbit;
+
+ cookie &= RCFW_MAX_COOKIE_VALUE;
+ cbit = cookie % RCFW_MAX_OUTSTANDING_CMD;
+ if (!test_bit(cbit, rcfw->cmdq_bitmap))
+ goto done;
+ do {
+ bnxt_qplib_service_creq((unsigned long)rcfw);
+ } while (test_bit(cbit, rcfw->cmdq_bitmap) && --count);
+done:
+ return count;
+};
+
+void *bnxt_qplib_rcfw_send_message(struct bnxt_qplib_rcfw *rcfw,
+ struct cmdq_base *req, void **crsbe,
+ u8 is_block)
+{
+ struct bnxt_qplib_crsq *crsq = &rcfw->crsq;
+ struct bnxt_qplib_cmdqe *cmdqe, **cmdq_ptr;
+ struct bnxt_qplib_hwq *cmdq = &rcfw->cmdq;
+ struct bnxt_qplib_hwq *crsb = &rcfw->crsb;
+ struct bnxt_qplib_crsqe *crsqe = NULL;
+ struct bnxt_qplib_crsbe **crsb_ptr;
+ u32 sw_prod, cmdq_prod;
+ u8 retry_cnt = 0xFF;
+ dma_addr_t dma_addr;
+ unsigned long flags;
+ u32 size, opcode;
+ u16 cookie, cbit;
+ int pg, idx;
+ u8 *preq;
+
+retry:
+ opcode = req->opcode;
+ if (!test_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags) &&
+ (opcode != CMDQ_BASE_OPCODE_QUERY_FUNC &&
+ opcode != CMDQ_BASE_OPCODE_INITIALIZE_FW)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW not initialized, reject opcode 0x%x",
+ opcode);
+ return NULL;
+ }
+
+ if (test_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags) &&
+ opcode == CMDQ_BASE_OPCODE_INITIALIZE_FW) {
+ dev_err(&rcfw->pdev->dev, "QPLIB: RCFW already initialized!");
+ return NULL;
+ }
+
+ /* Cmdq are in 16-byte units, each request can consume 1 or more
+ * cmdqe
+ */
+ spin_lock_irqsave(&cmdq->lock, flags);
+ if (req->cmd_size > cmdq->max_elements -
+ ((HWQ_CMP(cmdq->prod, cmdq) - HWQ_CMP(cmdq->cons, cmdq)) &
+ (cmdq->max_elements - 1))) {
+ dev_err(&rcfw->pdev->dev, "QPLIB: RCFW: CMDQ is full!");
+ spin_unlock_irqrestore(&cmdq->lock, flags);
+
+ if (!retry_cnt--)
+ return NULL;
+ goto retry;
+ }
+
+ retry_cnt = 0xFF;
+
+ cookie = atomic_inc_return(&rcfw->seq_num) & RCFW_MAX_COOKIE_VALUE;
+ cbit = cookie % RCFW_MAX_OUTSTANDING_CMD;
+ if (is_block)
+ cookie |= RCFW_CMD_IS_BLOCKING;
+ req->cookie = cpu_to_le16(cookie);
+ if (test_and_set_bit(cbit, rcfw->cmdq_bitmap)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW MAX outstanding cmd reached!");
+ atomic_dec(&rcfw->seq_num);
+ spin_unlock_irqrestore(&cmdq->lock, flags);
+
+ if (!retry_cnt--)
+ return NULL;
+ goto retry;
+ }
+ /* Reserve a resp buffer slot if requested */
+ if (req->resp_size && crsbe) {
+ spin_lock(&crsb->lock);
+ sw_prod = HWQ_CMP(crsb->prod, crsb);
+ crsb_ptr = (struct bnxt_qplib_crsbe **)crsb->pbl_ptr;
+ *crsbe = (void *)&crsb_ptr[get_crsb_pg(sw_prod)]
+ [get_crsb_idx(sw_prod)];
+ bnxt_qplib_crsb_dma_next(crsb->pbl_dma_ptr, sw_prod, &dma_addr);
+ req->resp_addr = cpu_to_le64(dma_addr);
+ crsb->prod++;
+ spin_unlock(&crsb->lock);
+
+ req->resp_size = (sizeof(struct bnxt_qplib_crsbe) +
+ BNXT_QPLIB_CMDQE_UNITS - 1) /
+ BNXT_QPLIB_CMDQE_UNITS;
+ }
+ cmdq_ptr = (struct bnxt_qplib_cmdqe **)cmdq->pbl_ptr;
+ preq = (u8 *)req;
+ size = req->cmd_size * BNXT_QPLIB_CMDQE_UNITS;
+ do {
+ pg = 0;
+ idx = 0;
+
+ /* Locate the next cmdq slot */
+ sw_prod = HWQ_CMP(cmdq->prod, cmdq);
+ cmdqe = &cmdq_ptr[get_cmdq_pg(sw_prod)][get_cmdq_idx(sw_prod)];
+ if (!cmdqe) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW request failed with no cmdqe!");
+ goto done;
+ }
+ /* Copy a segment of the req cmd to the cmdq */
+ memset(cmdqe, 0, sizeof(*cmdqe));
+ memcpy(cmdqe, preq, min_t(u32, size, sizeof(*cmdqe)));
+ preq += min_t(u32, size, sizeof(*cmdqe));
+ size -= min_t(u32, size, sizeof(*cmdqe));
+ cmdq->prod++;
+ } while (size > 0);
+
+ cmdq_prod = cmdq->prod;
+ if (rcfw->flags & FIRMWARE_FIRST_FLAG) {
+ /* The very first doorbell write is required to set this flag
+ * which prompts the FW to reset its internal pointers
+ */
+ cmdq_prod |= FIRMWARE_FIRST_FLAG;
+ rcfw->flags &= ~FIRMWARE_FIRST_FLAG;
+ }
+ sw_prod = HWQ_CMP(crsq->prod, crsq);
+ crsqe = &crsq->crsq[sw_prod];
+ memset(crsqe, 0, sizeof(*crsqe));
+ crsq->prod++;
+ crsqe->req_size = req->cmd_size;
+
+ /* ring CMDQ DB */
+ writel(cmdq_prod, rcfw->cmdq_bar_reg_iomem +
+ rcfw->cmdq_bar_reg_prod_off);
+ writel(RCFW_CMDQ_TRIG_VAL, rcfw->cmdq_bar_reg_iomem +
+ rcfw->cmdq_bar_reg_trig_off);
+done:
+ spin_unlock_irqrestore(&cmdq->lock, flags);
+ /* Return the CREQ response pointer */
+ return crsqe ? &crsqe->qp_event : NULL;
+}
+
+/* Completions */
+static int bnxt_qplib_process_func_event(struct bnxt_qplib_rcfw *rcfw,
+ struct creq_func_event *func_event)
+{
+ switch (func_event->event) {
+ case CREQ_FUNC_EVENT_EVENT_TX_WQE_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TX_DATA_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_RX_WQE_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_RX_DATA_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CQ_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TQM_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCQ_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCS_ERROR:
+ /* SRQ ctx error, call srq_handler??
+ * But there's no SRQ handle!
+ */
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCC_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_CFCM_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_TIM_ERROR:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_VF_COMM_REQUEST:
+ break;
+ case CREQ_FUNC_EVENT_EVENT_RESOURCE_EXHAUSTED:
+ break;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+}
+
+/* SP - CREQ Completion handlers */
+static void bnxt_qplib_service_creq(unsigned long data)
+{
+ struct bnxt_qplib_rcfw *rcfw = (struct bnxt_qplib_rcfw *)data;
+ struct bnxt_qplib_hwq *creq = &rcfw->creq;
+ struct creq_base *creqe, **creq_ptr;
+ u32 sw_cons, raw_cons;
+ unsigned long flags;
+ u32 type;
+
+ /* Service the CREQ until empty */
+ spin_lock_irqsave(&creq->lock, flags);
+ raw_cons = creq->cons;
+ while (1) {
+ sw_cons = HWQ_CMP(raw_cons, creq);
+ creq_ptr = (struct creq_base **)creq->pbl_ptr;
+ creqe = &creq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)];
+ if (!CREQ_CMP_VALID(creqe, raw_cons, creq->max_elements))
+ break;
+
+ type = creqe->type & CREQ_BASE_TYPE_MASK;
+ switch (type) {
+ case CREQ_BASE_TYPE_QP_EVENT:
+ break;
+ case CREQ_BASE_TYPE_FUNC_EVENT:
+ if (!bnxt_qplib_process_func_event
+ (rcfw, (struct creq_func_event *)creqe))
+ rcfw->creq_func_event_processed++;
+ else
+ dev_warn
+ (&rcfw->pdev->dev, "QPLIB:aeqe:%#x Not handled",
+ type);
+ break;
+ default:
+ dev_warn(&rcfw->pdev->dev, "QPLIB: creqe with ");
+ dev_warn(&rcfw->pdev->dev,
+ "QPLIB: op_event = 0x%x not handled", type);
+ break;
+ }
+ raw_cons++;
+ }
+ if (creq->cons != raw_cons) {
+ creq->cons = raw_cons;
+ CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, raw_cons,
+ creq->max_elements);
+ }
+ spin_unlock_irqrestore(&creq->lock, flags);
+}
+
+static irqreturn_t bnxt_qplib_creq_irq(int irq, void *dev_instance)
+{
+ struct bnxt_qplib_rcfw *rcfw = dev_instance;
+ struct bnxt_qplib_hwq *creq = &rcfw->creq;
+ struct creq_base **creq_ptr;
+ u32 sw_cons;
+
+ /* Prefetch the CREQ element */
+ sw_cons = HWQ_CMP(creq->cons, creq);
+ creq_ptr = (struct creq_base **)rcfw->creq.pbl_ptr;
+ prefetch(&creq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)]);
+
+ tasklet_schedule(&rcfw->worker);
+
+ return IRQ_HANDLED;
+}
+
+/* RCFW */
+int bnxt_qplib_deinit_rcfw(struct bnxt_qplib_rcfw *rcfw)
+{
+ struct creq_deinitialize_fw_resp *resp;
+ struct cmdq_deinitialize_fw req;
+ u16 cmd_flags = 0;
+
+ RCFW_CMD_PREP(req, DEINITIALIZE_FW, cmd_flags);
+ resp = (struct creq_deinitialize_fw_resp *)
+ bnxt_qplib_rcfw_send_message(rcfw, (void *)&req,
+ NULL, 0);
+ if (!resp)
+ return -EINVAL;
+
+ if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie)))
+ return -ETIMEDOUT;
+
+ if (resp->status ||
+ le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie))
+ return -EFAULT;
+
+ clear_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags);
+ return 0;
+}
+
+static int __get_pbl_pg_idx(struct bnxt_qplib_pbl *pbl)
+{
+ return (pbl->pg_size == ROCE_PG_SIZE_4K ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K :
+ pbl->pg_size == ROCE_PG_SIZE_8K ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8K :
+ pbl->pg_size == ROCE_PG_SIZE_64K ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_64K :
+ pbl->pg_size == ROCE_PG_SIZE_2M ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_2M :
+ pbl->pg_size == ROCE_PG_SIZE_8M ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8M :
+ pbl->pg_size == ROCE_PG_SIZE_1G ?
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_1G :
+ CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K);
+}
+
+int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw,
+ struct bnxt_qplib_ctx *ctx, int is_virtfn)
+{
+ struct creq_initialize_fw_resp *resp;
+ struct cmdq_initialize_fw req;
+ u16 cmd_flags = 0, level;
+
+ RCFW_CMD_PREP(req, INITIALIZE_FW, cmd_flags);
+
+ /*
+ * VFs need not setup the HW context area, PF
+ * shall setup this area for VF. Skipping the
+ * HW programming
+ */
+ if (is_virtfn)
+ goto skip_ctx_setup;
+
+ level = ctx->qpc_tbl.level;
+ req.qpc_pg_size_qpc_lvl = (level << CMDQ_INITIALIZE_FW_QPC_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->qpc_tbl.pbl[level]);
+ level = ctx->mrw_tbl.level;
+ req.mrw_pg_size_mrw_lvl = (level << CMDQ_INITIALIZE_FW_MRW_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->mrw_tbl.pbl[level]);
+ level = ctx->srqc_tbl.level;
+ req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]);
+ level = ctx->cq_tbl.level;
+ req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]);
+ level = ctx->srqc_tbl.level;
+ req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]);
+ level = ctx->cq_tbl.level;
+ req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]);
+ level = ctx->tim_tbl.level;
+ req.tim_pg_size_tim_lvl = (level << CMDQ_INITIALIZE_FW_TIM_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->tim_tbl.pbl[level]);
+ level = ctx->tqm_pde_level;
+ req.tqm_pg_size_tqm_lvl = (level << CMDQ_INITIALIZE_FW_TQM_LVL_SFT) |
+ __get_pbl_pg_idx(&ctx->tqm_pde.pbl[level]);
+
+ req.qpc_page_dir =
+ cpu_to_le64(ctx->qpc_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
+ req.mrw_page_dir =
+ cpu_to_le64(ctx->mrw_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
+ req.srq_page_dir =
+ cpu_to_le64(ctx->srqc_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
+ req.cq_page_dir =
+ cpu_to_le64(ctx->cq_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
+ req.tim_page_dir =
+ cpu_to_le64(ctx->tim_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
+ req.tqm_page_dir =
+ cpu_to_le64(ctx->tqm_pde.pbl[PBL_LVL_0].pg_map_arr[0]);
+
+ req.number_of_qp = cpu_to_le32(ctx->qpc_tbl.max_elements);
+ req.number_of_mrw = cpu_to_le32(ctx->mrw_tbl.max_elements);
+ req.number_of_srq = cpu_to_le32(ctx->srqc_tbl.max_elements);
+ req.number_of_cq = cpu_to_le32(ctx->cq_tbl.max_elements);
+
+ req.max_qp_per_vf = cpu_to_le32(ctx->vf_res.max_qp_per_vf);
+ req.max_mrw_per_vf = cpu_to_le32(ctx->vf_res.max_mrw_per_vf);
+ req.max_srq_per_vf = cpu_to_le32(ctx->vf_res.max_srq_per_vf);
+ req.max_cq_per_vf = cpu_to_le32(ctx->vf_res.max_cq_per_vf);
+ req.max_gid_per_vf = cpu_to_le32(ctx->vf_res.max_gid_per_vf);
+
+skip_ctx_setup:
+ req.stat_ctx_id = cpu_to_le32(ctx->stats.fw_id);
+ resp = (struct creq_initialize_fw_resp *)
+ bnxt_qplib_rcfw_send_message(rcfw, (void *)&req,
+ NULL, 0);
+ if (!resp) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW: INITIALIZE_FW send failed");
+ return -EINVAL;
+ }
+ if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) {
+ /* Cmd timed out */
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW: INITIALIZE_FW timed out");
+ return -ETIMEDOUT;
+ }
+ if (resp->status ||
+ le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: RCFW: INITIALIZE_FW failed");
+ return -EINVAL;
+ }
+ set_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags);
+ return 0;
+}
+
+void bnxt_qplib_free_rcfw_channel(struct bnxt_qplib_rcfw *rcfw)
+{
+ bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->crsb);
+ kfree(rcfw->crsq.crsq);
+ bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->cmdq);
+ bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->creq);
+
+ rcfw->pdev = NULL;
+}
+
+int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev,
+ struct bnxt_qplib_rcfw *rcfw)
+{
+ rcfw->pdev = pdev;
+ rcfw->creq.max_elements = BNXT_QPLIB_CREQE_MAX_CNT;
+ if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->creq, NULL, 0,
+ &rcfw->creq.max_elements,
+ BNXT_QPLIB_CREQE_UNITS, 0, PAGE_SIZE,
+ HWQ_TYPE_L2_CMPL)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: HW channel CREQ allocation failed");
+ goto fail;
+ }
+ rcfw->cmdq.max_elements = BNXT_QPLIB_CMDQE_MAX_CNT;
+ if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->cmdq, NULL, 0,
+ &rcfw->cmdq.max_elements,
+ BNXT_QPLIB_CMDQE_UNITS, 0, PAGE_SIZE,
+ HWQ_TYPE_CTX)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: HW channel CMDQ allocation failed");
+ goto fail;
+ }
+
+ rcfw->crsq.max_elements = rcfw->cmdq.max_elements;
+ rcfw->crsq.crsq = kcalloc(rcfw->crsq.max_elements,
+ sizeof(*rcfw->crsq.crsq), GFP_KERNEL);
+ if (!rcfw->crsq.crsq)
+ goto fail;
+
+ rcfw->crsb.max_elements = BNXT_QPLIB_CRSBE_MAX_CNT;
+ if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->crsb, NULL, 0,
+ &rcfw->crsb.max_elements,
+ BNXT_QPLIB_CRSBE_UNITS, 0, PAGE_SIZE,
+ HWQ_TYPE_CTX)) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: HW channel CRSB allocation failed");
+ goto fail;
+ }
+ return 0;
+
+fail:
+ bnxt_qplib_free_rcfw_channel(rcfw);
+ return -ENOMEM;
+}
+
+void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw)
+{
+ unsigned long indx;
+
+ /* Make sure the HW channel is stopped! */
+ synchronize_irq(rcfw->vector);
+ tasklet_disable(&rcfw->worker);
+ tasklet_kill(&rcfw->worker);
+
+ if (rcfw->requested) {
+ free_irq(rcfw->vector, rcfw);
+ rcfw->requested = false;
+ }
+ if (rcfw->cmdq_bar_reg_iomem)
+ iounmap(rcfw->cmdq_bar_reg_iomem);
+ rcfw->cmdq_bar_reg_iomem = NULL;
+
+ if (rcfw->creq_bar_reg_iomem)
+ iounmap(rcfw->creq_bar_reg_iomem);
+ rcfw->creq_bar_reg_iomem = NULL;
+
+ indx = find_first_bit(rcfw->cmdq_bitmap, rcfw->bmap_size);
+ if (indx != rcfw->bmap_size)
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: disabling RCFW with pending cmd-bit %lx", indx);
+ kfree(rcfw->cmdq_bitmap);
+ rcfw->bmap_size = 0;
+
+ rcfw->aeq_handler = NULL;
+ rcfw->vector = 0;
+}
+
+int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev,
+ struct bnxt_qplib_rcfw *rcfw,
+ int msix_vector,
+ int cp_bar_reg_off, int virt_fn,
+ int (*aeq_handler)(struct bnxt_qplib_rcfw *,
+ struct creq_func_event *))
+{
+ resource_size_t res_base;
+ struct cmdq_init init;
+ u16 bmap_size;
+ int rc;
+
+ /* General */
+ atomic_set(&rcfw->seq_num, 0);
+ rcfw->flags = FIRMWARE_FIRST_FLAG;
+ bmap_size = BITS_TO_LONGS(RCFW_MAX_OUTSTANDING_CMD *
+ sizeof(unsigned long));
+ rcfw->cmdq_bitmap = kzalloc(bmap_size, GFP_KERNEL);
+ if (!rcfw->cmdq_bitmap)
+ return -ENOMEM;
+ rcfw->bmap_size = bmap_size;
+
+ /* CMDQ */
+ rcfw->cmdq_bar_reg = RCFW_COMM_PCI_BAR_REGION;
+ res_base = pci_resource_start(pdev, rcfw->cmdq_bar_reg);
+ if (!res_base)
+ return -ENOMEM;
+
+ rcfw->cmdq_bar_reg_iomem = ioremap_nocache(res_base +
+ RCFW_COMM_BASE_OFFSET,
+ RCFW_COMM_SIZE);
+ if (!rcfw->cmdq_bar_reg_iomem) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: CMDQ BAR region %d mapping failed",
+ rcfw->cmdq_bar_reg);
+ return -ENOMEM;
+ }
+
+ rcfw->cmdq_bar_reg_prod_off = virt_fn ? RCFW_VF_COMM_PROD_OFFSET :
+ RCFW_PF_COMM_PROD_OFFSET;
+
+ rcfw->cmdq_bar_reg_trig_off = RCFW_COMM_TRIG_OFFSET;
+
+ /* CRSQ */
+ rcfw->crsq.prod = 0;
+ rcfw->crsq.cons = 0;
+
+ /* CREQ */
+ rcfw->creq_bar_reg = RCFW_COMM_CONS_PCI_BAR_REGION;
+ res_base = pci_resource_start(pdev, rcfw->creq_bar_reg);
+ if (!res_base)
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: CREQ BAR region %d resc start is 0!",
+ rcfw->creq_bar_reg);
+ rcfw->creq_bar_reg_iomem = ioremap_nocache(res_base + cp_bar_reg_off,
+ 4);
+ if (!rcfw->creq_bar_reg_iomem) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: CREQ BAR region %d mapping failed",
+ rcfw->creq_bar_reg);
+ return -ENOMEM;
+ }
+ rcfw->creq_qp_event_processed = 0;
+ rcfw->creq_func_event_processed = 0;
+
+ rcfw->vector = msix_vector;
+ if (aeq_handler)
+ rcfw->aeq_handler = aeq_handler;
+
+ tasklet_init(&rcfw->worker, bnxt_qplib_service_creq,
+ (unsigned long)rcfw);
+
+ rcfw->requested = false;
+ rc = request_irq(rcfw->vector, bnxt_qplib_creq_irq, 0,
+ "bnxt_qplib_creq", rcfw);
+ if (rc) {
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: Failed to request IRQ for CREQ rc = 0x%x", rc);
+ bnxt_qplib_disable_rcfw_channel(rcfw);
+ return rc;
+ }
+ rcfw->requested = true;
+
+ init_waitqueue_head(&rcfw->waitq);
+
+ CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, 0, rcfw->creq.max_elements);
+
+ init.cmdq_pbl = cpu_to_le64(rcfw->cmdq.pbl[PBL_LVL_0].pg_map_arr[0]);
+ init.cmdq_size_cmdq_lvl = cpu_to_le16(
+ ((BNXT_QPLIB_CMDQE_MAX_CNT << CMDQ_INIT_CMDQ_SIZE_SFT) &
+ CMDQ_INIT_CMDQ_SIZE_MASK) |
+ ((rcfw->cmdq.level << CMDQ_INIT_CMDQ_LVL_SFT) &
+ CMDQ_INIT_CMDQ_LVL_MASK));
+ init.creq_ring_id = cpu_to_le16(rcfw->creq_ring_id);
+
+ /* Write to the Bono mailbox register */
+ __iowrite32_copy(rcfw->cmdq_bar_reg_iomem, &init, sizeof(init) / 4);
+ return 0;
+}
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
index 1f63956..b1e114e 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
@@ -39,4 +39,193 @@
#ifndef __BNXT_QPLIB_RCFW_H__
#define __BNXT_QPLIB_RCFW_H__
+#define RCFW_CMDQ_TRIG_VAL 1
+#define RCFW_COMM_PCI_BAR_REGION 0
+#define RCFW_COMM_CONS_PCI_BAR_REGION 2
+#define RCFW_COMM_BASE_OFFSET 0x600
+#define RCFW_PF_COMM_PROD_OFFSET 0x7c
+#define RCFW_VF_COMM_PROD_OFFSET 0xc
+#define RCFW_COMM_TRIG_OFFSET 0x100
+#define RCFW_COMM_SIZE 0x104
+
+#define RCFW_DBR_PCI_BAR_REGION 2
+
+#define RCFW_CMD_PREP(req, CMD, cmd_flags) \
+ do { \
+ memset(&(req), 0, sizeof((req))); \
+ (req).opcode = CMDQ_BASE_OPCODE_##CMD; \
+ (req).cmd_size = (sizeof((req)) + \
+ BNXT_QPLIB_CMDQE_UNITS - 1) / \
+ BNXT_QPLIB_CMDQE_UNITS; \
+ (req).flags = cpu_to_le16(cmd_flags); \
+ } while (0)
+
+#define RCFW_CMD_WAIT_TIME_MS 20000 /* 20 Seconds timeout */
+
+/* CMDQ elements */
+#define BNXT_QPLIB_CMDQE_MAX_CNT 256
+#define BNXT_QPLIB_CMDQE_UNITS sizeof(struct bnxt_qplib_cmdqe)
+#define BNXT_QPLIB_CMDQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CMDQE_UNITS)
+
+#define MAX_CMDQ_IDX (BNXT_QPLIB_CMDQE_MAX_CNT - 1)
+#define MAX_CMDQ_IDX_PER_PG (BNXT_QPLIB_CMDQE_CNT_PER_PG - 1)
+
+#define RCFW_MAX_OUTSTANDING_CMD BNXT_QPLIB_CMDQE_MAX_CNT
+#define RCFW_MAX_COOKIE_VALUE 0x7FFF
+#define RCFW_CMD_IS_BLOCKING 0x8000
+
+/* Cmdq contains a fix number of a 16-Byte slots */
+struct bnxt_qplib_cmdqe {
+ u8 data[16];
+};
+
+static inline u32 get_cmdq_pg(u32 val)
+{
+ return (val & ~MAX_CMDQ_IDX_PER_PG) / BNXT_QPLIB_CMDQE_CNT_PER_PG;
+}
+
+static inline u32 get_cmdq_idx(u32 val)
+{
+ return val & MAX_CMDQ_IDX_PER_PG;
+}
+
+/* Crsq buf is 1024-Byte */
+struct bnxt_qplib_crsbe {
+ u8 data[1024];
+};
+
+/* CRSQ SB */
+#define BNXT_QPLIB_CRSBE_MAX_CNT 4
+#define BNXT_QPLIB_CRSBE_UNITS sizeof(struct bnxt_qplib_crsbe)
+#define BNXT_QPLIB_CRSBE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CRSBE_UNITS)
+
+#define MAX_CRSB_IDX (BNXT_QPLIB_CRSBE_MAX_CNT - 1)
+#define MAX_CRSB_IDX_PER_PG (BNXT_QPLIB_CRSBE_CNT_PER_PG - 1)
+
+static inline u32 get_crsb_pg(u32 val)
+{
+ return (val & ~MAX_CRSB_IDX_PER_PG) / BNXT_QPLIB_CRSBE_CNT_PER_PG;
+}
+
+static inline u32 get_crsb_idx(u32 val)
+{
+ return val & MAX_CRSB_IDX_PER_PG;
+}
+
+static inline void bnxt_qplib_crsb_dma_next(dma_addr_t *pg_map_arr,
+ u32 prod, dma_addr_t *dma_addr)
+{
+ *dma_addr = pg_map_arr[(prod) / BNXT_QPLIB_CRSBE_CNT_PER_PG];
+ *dma_addr += ((prod) % BNXT_QPLIB_CRSBE_CNT_PER_PG) *
+ BNXT_QPLIB_CRSBE_UNITS;
+}
+
+/* CREQ */
+/* Allocate 1 per QP for async error notification for now */
+#define BNXT_QPLIB_CREQE_MAX_CNT (64 * 1024)
+#define BNXT_QPLIB_CREQE_UNITS 16 /* 16-Bytes per prod unit */
+#define BNXT_QPLIB_CREQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CREQE_UNITS)
+
+#define MAX_CREQ_IDX (BNXT_QPLIB_CREQE_MAX_CNT - 1)
+#define MAX_CREQ_IDX_PER_PG (BNXT_QPLIB_CREQE_CNT_PER_PG - 1)
+
+static inline u32 get_creq_pg(u32 val)
+{
+ return (val & ~MAX_CREQ_IDX_PER_PG) / BNXT_QPLIB_CREQE_CNT_PER_PG;
+}
+
+static inline u32 get_creq_idx(u32 val)
+{
+ return val & MAX_CREQ_IDX_PER_PG;
+}
+
+#define BNXT_QPLIB_CREQE_PER_PG (PAGE_SIZE / sizeof(struct creq_base))
+
+#define CREQ_CMP_VALID(hdr, raw_cons, cp_bit) \
+ (!!((hdr)->v & CREQ_BASE_V) == \
+ !((raw_cons) & (cp_bit)))
+
+#define CREQ_DB_KEY_CP (0x2 << CMPL_DOORBELL_KEY_SFT)
+#define CREQ_DB_IDX_VALID CMPL_DOORBELL_IDX_VALID
+#define CREQ_DB_IRQ_DIS CMPL_DOORBELL_MASK
+#define CREQ_DB_CP_FLAGS_REARM (CREQ_DB_KEY_CP | \
+ CREQ_DB_IDX_VALID)
+#define CREQ_DB_CP_FLAGS (CREQ_DB_KEY_CP | \
+ CREQ_DB_IDX_VALID | \
+ CREQ_DB_IRQ_DIS)
+#define CREQ_DB_REARM(db, raw_cons, cp_bit) \
+ writel(CREQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db)
+#define CREQ_DB(db, raw_cons, cp_bit) \
+ writel(CREQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db)
+
+/* HWQ */
+struct bnxt_qplib_crsqe {
+ struct creq_qp_event qp_event;
+ u32 req_size;
+};
+
+struct bnxt_qplib_crsq {
+ struct bnxt_qplib_crsqe *crsq;
+ u32 prod;
+ u32 cons;
+ u32 max_elements;
+};
+
+/* RCFW Communication Channels */
+struct bnxt_qplib_rcfw {
+ struct pci_dev *pdev;
+ int vector;
+ struct tasklet_struct worker;
+ bool requested;
+ unsigned long *cmdq_bitmap;
+ u32 bmap_size;
+ unsigned long flags;
+#define FIRMWARE_INITIALIZED_FLAG 1
+#define FIRMWARE_FIRST_FLAG BIT(31)
+ wait_queue_head_t waitq;
+ int (*aeq_handler)(struct bnxt_qplib_rcfw *,
+ struct creq_func_event *);
+ atomic_t seq_num;
+
+ /* Bar region info */
+ void __iomem *cmdq_bar_reg_iomem;
+ u16 cmdq_bar_reg;
+ u16 cmdq_bar_reg_prod_off;
+ u16 cmdq_bar_reg_trig_off;
+ u16 creq_ring_id;
+ u16 creq_bar_reg;
+ void __iomem *creq_bar_reg_iomem;
+
+ /* Cmd-Resp and Async Event notification queue */
+ struct bnxt_qplib_hwq creq;
+ u64 creq_qp_event_processed;
+ u64 creq_func_event_processed;
+
+ /* Actual Cmd and Resp Queues */
+ struct bnxt_qplib_hwq cmdq;
+ struct bnxt_qplib_crsq crsq;
+ struct bnxt_qplib_hwq crsb;
+};
+
+void bnxt_qplib_free_rcfw_channel(struct bnxt_qplib_rcfw *rcfw);
+int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev,
+ struct bnxt_qplib_rcfw *rcfw);
+void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw);
+int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev,
+ struct bnxt_qplib_rcfw *rcfw,
+ int msix_vector,
+ int cp_bar_reg_off, int virt_fn,
+ int (*aeq_handler)
+ (struct bnxt_qplib_rcfw *,
+ struct creq_func_event *));
+
+int bnxt_qplib_rcfw_block_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie);
+int bnxt_qplib_rcfw_wait_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie);
+void *bnxt_qplib_rcfw_send_message(struct bnxt_qplib_rcfw *rcfw,
+ struct cmdq_base *req, void **crsbe,
+ u8 is_block);
+
+int bnxt_qplib_deinit_rcfw(struct bnxt_qplib_rcfw *rcfw);
+int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw,
+ struct bnxt_qplib_ctx *ctx, int is_virtfn);
#endif /* __BNXT_QPLIB_RCFW_H__ */
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c
index 178eebd..c9ef4c2 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c
@@ -35,3 +35,741 @@
*
* Description: QPLib resource manager
*/
+
+#include <linux/spinlock.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <linux/inetdevice.h>
+#include <linux/dma-mapping.h>
+#include <linux/if_vlan.h>
+#include "roce_hsi.h"
+#include "qplib_res.h"
+#include "qplib_sp.h"
+#include "qplib_rcfw.h"
+
+static void bnxt_qplib_free_stats_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_stats *stats);
+static int bnxt_qplib_alloc_stats_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_stats *stats);
+
+/* PBL */
+static void __free_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl,
+ bool is_umem)
+{
+ int i;
+
+ if (!is_umem) {
+ for (i = 0; i < pbl->pg_count; i++) {
+ if (pbl->pg_arr[i])
+ dma_free_coherent(&pdev->dev, pbl->pg_size,
+ (void *)((unsigned long)
+ pbl->pg_arr[i] &
+ PAGE_MASK),
+ pbl->pg_map_arr[i]);
+ else
+ dev_warn(&pdev->dev,
+ "QPLIB: PBL free pg_arr[%d] empty?!",
+ i);
+ pbl->pg_arr[i] = NULL;
+ }
+ }
+ kfree(pbl->pg_arr);
+ pbl->pg_arr = NULL;
+ kfree(pbl->pg_map_arr);
+ pbl->pg_map_arr = NULL;
+ pbl->pg_count = 0;
+ pbl->pg_size = 0;
+}
+
+static int __alloc_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl,
+ struct scatterlist *sghead, u32 pages, u32 pg_size)
+{
+ struct scatterlist *sg;
+ bool is_umem = false;
+ int i;
+
+ /* page ptr arrays */
+ pbl->pg_arr = kcalloc(pages, sizeof(void *), GFP_KERNEL);
+ if (!pbl->pg_arr)
+ return -ENOMEM;
+
+ pbl->pg_map_arr = kcalloc(pages, sizeof(dma_addr_t), GFP_KERNEL);
+ if (!pbl->pg_map_arr) {
+ kfree(pbl->pg_arr);
+ pbl->pg_arr = NULL;
+ return -ENOMEM;
+ }
+ pbl->pg_count = 0;
+ pbl->pg_size = pg_size;
+
+ if (!sghead) {
+ for (i = 0; i < pages; i++) {
+ pbl->pg_arr[i] = dma_alloc_coherent(&pdev->dev,
+ pbl->pg_size,
+ &pbl->pg_map_arr[i],
+ GFP_KERNEL);
+ if (!pbl->pg_arr[i])
+ goto fail;
+ memset(pbl->pg_arr[i], 0, pbl->pg_size);
+ pbl->pg_count++;
+ }
+ } else {
+ i = 0;
+ is_umem = true;
+ for_each_sg(sghead, sg, pages, i) {
+ pbl->pg_map_arr[i] = sg_dma_address(sg);
+ pbl->pg_arr[i] = sg_virt(sg);
+ if (!pbl->pg_arr[i])
+ goto fail;
+
+ pbl->pg_count++;
+ }
+ }
+
+ return 0;
+
+fail:
+ __free_pbl(pdev, pbl, is_umem);
+ return -ENOMEM;
+}
+
+/* HWQ */
+void bnxt_qplib_free_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq)
+{
+ int i;
+
+ if (!hwq->max_elements)
+ return;
+ if (hwq->level >= PBL_LVL_MAX)
+ return;
+
+ for (i = 0; i < hwq->level + 1; i++) {
+ if (i == hwq->level)
+ __free_pbl(pdev, &hwq->pbl[i], hwq->is_user);
+ else
+ __free_pbl(pdev, &hwq->pbl[i], false);
+ }
+
+ hwq->level = PBL_LVL_MAX;
+ hwq->max_elements = 0;
+ hwq->element_size = 0;
+ hwq->prod = 0;
+ hwq->cons = 0;
+ hwq->cp_bit = 0;
+}
+
+/* All HWQs are power of 2 in size */
+int bnxt_qplib_alloc_init_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq,
+ struct scatterlist *sghead, int nmap,
+ u32 *elements, u32 element_size, u32 aux,
+ u32 pg_size, enum bnxt_qplib_hwq_type hwq_type)
+{
+ u32 pages, slots, size, aux_pages = 0, aux_size = 0;
+ dma_addr_t *src_phys_ptr, **dst_virt_ptr;
+ int i, rc;
+
+ hwq->level = PBL_LVL_MAX;
+
+ slots = roundup_pow_of_two(*elements);
+ if (aux) {
+ aux_size = roundup_pow_of_two(aux);
+ aux_pages = (slots * aux_size) / pg_size;
+ if ((slots * aux_size) % pg_size)
+ aux_pages++;
+ }
+ size = roundup_pow_of_two(element_size);
+
+ if (!sghead) {
+ hwq->is_user = false;
+ pages = (slots * size) / pg_size + aux_pages;
+ if ((slots * size) % pg_size)
+ pages++;
+ if (!pages)
+ return -EINVAL;
+ } else {
+ hwq->is_user = true;
+ pages = nmap;
+ }
+
+ /* Alloc the 1st memory block; can be a PDL/PTL/PBL */
+ if (sghead && (pages == MAX_PBL_LVL_0_PGS))
+ rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_0], sghead,
+ pages, pg_size);
+ else
+ rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_0], NULL, 1, pg_size);
+ if (rc)
+ goto fail;
+
+ hwq->level = PBL_LVL_0;
+
+ if (pages > MAX_PBL_LVL_0_PGS) {
+ if (pages > MAX_PBL_LVL_1_PGS) {
+ /* 2 levels of indirection */
+ rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_1], NULL,
+ MAX_PBL_LVL_1_PGS_FOR_LVL_2, pg_size);
+ if (rc)
+ goto fail;
+ /* Fill in lvl0 PBL */
+ dst_virt_ptr =
+ (dma_addr_t **)hwq->pbl[PBL_LVL_0].pg_arr;
+ src_phys_ptr = hwq->pbl[PBL_LVL_1].pg_map_arr;
+ for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count; i++)
+ dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] =
+ src_phys_ptr[i] | PTU_PDE_VALID;
+ hwq->level = PBL_LVL_1;
+
+ rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_2], sghead,
+ pages, pg_size);
+ if (rc)
+ goto fail;
+
+ /* Fill in lvl1 PBL */
+ dst_virt_ptr =
+ (dma_addr_t **)hwq->pbl[PBL_LVL_1].pg_arr;
+ src_phys_ptr = hwq->pbl[PBL_LVL_2].pg_map_arr;
+ for (i = 0; i < hwq->pbl[PBL_LVL_2].pg_count; i++) {
+ dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] =
+ src_phys_ptr[i] | PTU_PTE_VALID;
+ }
+ if (hwq_type == HWQ_TYPE_QUEUE) {
+ /* Find the last pg of the size */
+ i = hwq->pbl[PBL_LVL_2].pg_count;
+ dst_virt_ptr[PTR_PG(i - 1)][PTR_IDX(i - 1)] |=
+ PTU_PTE_LAST;
+ if (i > 1)
+ dst_virt_ptr[PTR_PG(i - 2)]
+ [PTR_IDX(i - 2)] |=
+ PTU_PTE_NEXT_TO_LAST;
+ }
+ hwq->level = PBL_LVL_2;
+ } else {
+ u32 flag = hwq_type == HWQ_TYPE_L2_CMPL ? 0 :
+ PTU_PTE_VALID;
+
+ /* 1 level of indirection */
+ rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_1], sghead,
+ pages, pg_size);
+ if (rc)
+ goto fail;
+ /* Fill in lvl0 PBL */
+ dst_virt_ptr =
+ (dma_addr_t **)hwq->pbl[PBL_LVL_0].pg_arr;
+ src_phys_ptr = hwq->pbl[PBL_LVL_1].pg_map_arr;
+ for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count; i++) {
+ dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] =
+ src_phys_ptr[i] | flag;
+ }
+ if (hwq_type == HWQ_TYPE_QUEUE) {
+ /* Find the last pg of the size */
+ i = hwq->pbl[PBL_LVL_1].pg_count;
+ dst_virt_ptr[PTR_PG(i - 1)][PTR_IDX(i - 1)] |=
+ PTU_PTE_LAST;
+ if (i > 1)
+ dst_virt_ptr[PTR_PG(i - 2)]
+ [PTR_IDX(i - 2)] |=
+ PTU_PTE_NEXT_TO_LAST;
+ }
+ hwq->level = PBL_LVL_1;
+ }
+ }
+ hwq->pdev = pdev;
+ spin_lock_init(&hwq->lock);
+ hwq->prod = 0;
+ hwq->cons = 0;
+ *elements = hwq->max_elements = slots;
+ hwq->element_size = size;
+
+ /* For direct access to the elements */
+ hwq->pbl_ptr = hwq->pbl[hwq->level].pg_arr;
+ hwq->pbl_dma_ptr = hwq->pbl[hwq->level].pg_map_arr;
+
+ return 0;
+
+fail:
+ bnxt_qplib_free_hwq(pdev, hwq);
+ return -ENOMEM;
+}
+
+/* Context Tables */
+void bnxt_qplib_free_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_ctx *ctx)
+{
+ int i;
+
+ bnxt_qplib_free_hwq(pdev, &ctx->qpc_tbl);
+ bnxt_qplib_free_hwq(pdev, &ctx->mrw_tbl);
+ bnxt_qplib_free_hwq(pdev, &ctx->srqc_tbl);
+ bnxt_qplib_free_hwq(pdev, &ctx->cq_tbl);
+ bnxt_qplib_free_hwq(pdev, &ctx->tim_tbl);
+ for (i = 0; i < MAX_TQM_ALLOC_REQ; i++)
+ bnxt_qplib_free_hwq(pdev, &ctx->tqm_tbl[i]);
+ bnxt_qplib_free_hwq(pdev, &ctx->tqm_pde);
+ bnxt_qplib_free_stats_ctx(pdev, &ctx->stats);
+}
+
+/*
+ * Routine: bnxt_qplib_alloc_ctx
+ * Description:
+ * Context tables are memories which are used by the chip fw.
+ * The 6 tables defined are:
+ * QPC ctx - holds QP states
+ * MRW ctx - holds memory region and window
+ * SRQ ctx - holds shared RQ states
+ * CQ ctx - holds completion queue states
+ * TQM ctx - holds Tx Queue Manager context
+ * TIM ctx - holds timer context
+ * Depending on the size of the tbl requested, either a 1 Page Buffer List
+ * or a 1-to-2-stage indirection Page Directory List + 1 PBL is used
+ * instead.
+ * Table might be employed as follows:
+ * For 0 < ctx size <= 1 PAGE, 0 level of ind is used
+ * For 1 PAGE < ctx size <= 512 entries size, 1 level of ind is used
+ * For 512 < ctx size <= MAX, 2 levels of ind is used
+ * Returns:
+ * 0 if success, else -ERRORS
+ */
+int bnxt_qplib_alloc_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_ctx *ctx,
+ bool virt_fn)
+{
+ int i, j, k, rc = 0;
+ int fnz_idx = -1;
+ __le64 **pbl_ptr;
+
+ if (virt_fn)
+ goto stats_alloc;
+
+ /* QPC Tables */
+ ctx->qpc_tbl.max_elements = ctx->qpc_count;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->qpc_tbl, NULL, 0,
+ &ctx->qpc_tbl.max_elements,
+ BNXT_QPLIB_MAX_QP_CTX_ENTRY_SIZE, 0,
+ PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+ /* MRW Tables */
+ ctx->mrw_tbl.max_elements = ctx->mrw_count;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->mrw_tbl, NULL, 0,
+ &ctx->mrw_tbl.max_elements,
+ BNXT_QPLIB_MAX_MRW_CTX_ENTRY_SIZE, 0,
+ PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+ /* SRQ Tables */
+ ctx->srqc_tbl.max_elements = ctx->srqc_count;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->srqc_tbl, NULL, 0,
+ &ctx->srqc_tbl.max_elements,
+ BNXT_QPLIB_MAX_SRQ_CTX_ENTRY_SIZE, 0,
+ PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+ /* CQ Tables */
+ ctx->cq_tbl.max_elements = ctx->cq_count;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->cq_tbl, NULL, 0,
+ &ctx->cq_tbl.max_elements,
+ BNXT_QPLIB_MAX_CQ_CTX_ENTRY_SIZE, 0,
+ PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+ /* TQM Buffer */
+ ctx->tqm_pde.max_elements = 512;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tqm_pde, NULL, 0,
+ &ctx->tqm_pde.max_elements, sizeof(u64),
+ 0, PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+ for (i = 0; i < MAX_TQM_ALLOC_REQ; i++) {
+ if (!ctx->tqm_count[i])
+ continue;
+ ctx->tqm_tbl[i].max_elements = ctx->qpc_count *
+ ctx->tqm_count[i];
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tqm_tbl[i], NULL, 0,
+ &ctx->tqm_tbl[i].max_elements, 1,
+ 0, PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+ }
+ pbl_ptr = (__le64 **)ctx->tqm_pde.pbl_ptr;
+ for (i = 0, j = 0; i < MAX_TQM_ALLOC_REQ;
+ i++, j += MAX_TQM_ALLOC_BLK_SIZE) {
+ if (!ctx->tqm_tbl[i].max_elements)
+ continue;
+ if (fnz_idx == -1)
+ fnz_idx = i;
+ switch (ctx->tqm_tbl[i].level) {
+ case PBL_LVL_2:
+ for (k = 0; k < ctx->tqm_tbl[i].pbl[PBL_LVL_1].pg_count;
+ k++)
+ pbl_ptr[PTR_PG(j + k)][PTR_IDX(j + k)] =
+ cpu_to_le64(
+ ctx->tqm_tbl[i].pbl[PBL_LVL_1].pg_map_arr[k]
+ | PTU_PTE_VALID);
+ break;
+ case PBL_LVL_1:
+ case PBL_LVL_0:
+ default:
+ pbl_ptr[PTR_PG(j)][PTR_IDX(j)] = cpu_to_le64(
+ ctx->tqm_tbl[i].pbl[PBL_LVL_0].pg_map_arr[0] |
+ PTU_PTE_VALID);
+ break;
+ }
+ }
+ if (fnz_idx == -1)
+ fnz_idx = 0;
+ ctx->tqm_pde_level = ctx->tqm_tbl[fnz_idx].level == PBL_LVL_2 ?
+ PBL_LVL_2 : ctx->tqm_tbl[fnz_idx].level + 1;
+
+ /* TIM Buffer */
+ ctx->tim_tbl.max_elements = ctx->qpc_count * 16;
+ rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tim_tbl, NULL, 0,
+ &ctx->tim_tbl.max_elements, 1,
+ 0, PAGE_SIZE, HWQ_TYPE_CTX);
+ if (rc)
+ goto fail;
+
+stats_alloc:
+ /* Stats */
+ rc = bnxt_qplib_alloc_stats_ctx(pdev, &ctx->stats);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ bnxt_qplib_free_ctx(pdev, ctx);
+ return rc;
+}
+
+static void bnxt_qplib_free_sgid_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_sgid_tbl *sgid_tbl)
+{
+ kfree(sgid_tbl->tbl);
+ kfree(sgid_tbl->hw_id);
+ kfree(sgid_tbl->ctx);
+ sgid_tbl->tbl = NULL;
+ sgid_tbl->hw_id = NULL;
+ sgid_tbl->ctx = NULL;
+ sgid_tbl->max = 0;
+ sgid_tbl->active = 0;
+}
+
+static int bnxt_qplib_alloc_sgid_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_sgid_tbl *sgid_tbl,
+ u16 max)
+{
+ sgid_tbl->tbl = kcalloc(max, sizeof(struct bnxt_qplib_gid), GFP_KERNEL);
+ if (!sgid_tbl->tbl)
+ return -ENOMEM;
+
+ sgid_tbl->hw_id = kcalloc(max, sizeof(u16), GFP_KERNEL);
+ if (!sgid_tbl->hw_id)
+ goto out_free1;
+
+ sgid_tbl->ctx = kcalloc(max, sizeof(void *), GFP_KERNEL);
+ if (!sgid_tbl->ctx)
+ goto out_free2;
+
+ sgid_tbl->max = max;
+ return 0;
+out_free2:
+ kfree(sgid_tbl->hw_id);
+ sgid_tbl->hw_id = NULL;
+out_free1:
+ kfree(sgid_tbl->tbl);
+ sgid_tbl->tbl = NULL;
+ return -ENOMEM;
+};
+
+static void bnxt_qplib_cleanup_sgid_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_sgid_tbl *sgid_tbl)
+{
+ int i;
+
+ memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max);
+ memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max);
+ sgid_tbl->active = 0;
+}
+
+static void bnxt_qplib_init_sgid_tbl(struct bnxt_qplib_sgid_tbl *sgid_tbl,
+ struct net_device *netdev)
+{
+ memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max);
+ memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max);
+}
+
+static void bnxt_qplib_free_pkey_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl)
+{
+ if (!pkey_tbl->tbl)
+ dev_dbg(&res->pdev->dev, "QPLIB: PKEY tbl not present");
+ else
+ kfree(pkey_tbl->tbl);
+
+ pkey_tbl->tbl = NULL;
+ pkey_tbl->max = 0;
+ pkey_tbl->active = 0;
+}
+
+static int bnxt_qplib_alloc_pkey_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl,
+ u16 max)
+{
+ pkey_tbl->tbl = kcalloc(max, sizeof(u16), GFP_KERNEL);
+ if (!pkey_tbl->tbl)
+ return -ENOMEM;
+
+ pkey_tbl->max = max;
+ return 0;
+};
+
+static void bnxt_qplib_free_pd_tbl(struct bnxt_qplib_pd_tbl *pdt)
+{
+ kfree(pdt->tbl);
+ pdt->tbl = NULL;
+ pdt->max = 0;
+}
+
+static int bnxt_qplib_alloc_pd_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pd_tbl *pdt,
+ u32 max)
+{
+ u32 bytes;
+
+ bytes = max >> 3;
+ if (!bytes)
+ bytes = 1;
+ pdt->tbl = kmalloc(bytes, GFP_KERNEL);
+ if (!pdt->tbl)
+ return -ENOMEM;
+
+ pdt->max = max;
+ memset((u8 *)pdt->tbl, 0xFF, bytes);
+
+ return 0;
+}
+
+/* DPIs */
+int bnxt_qplib_alloc_dpi(struct bnxt_qplib_dpi_tbl *dpit,
+ struct bnxt_qplib_dpi *dpi,
+ void *app)
+{
+ u32 bit_num;
+
+ bit_num = find_first_bit(dpit->tbl, dpit->max);
+ if (bit_num == dpit->max)
+ return -ENOMEM;
+
+ /* Found unused DPI */
+ clear_bit(bit_num, dpit->tbl);
+ dpit->app_tbl[bit_num] = app;
+
+ dpi->dpi = bit_num;
+ dpi->dbr = dpit->dbr_bar_reg_iomem + (bit_num * PAGE_SIZE);
+ dpi->umdbr = dpit->unmapped_dbr + (bit_num * PAGE_SIZE);
+
+ return 0;
+}
+
+int bnxt_qplib_dealloc_dpi(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_dpi_tbl *dpit,
+ struct bnxt_qplib_dpi *dpi)
+{
+ if (dpi->dpi >= dpit->max) {
+ dev_warn(&res->pdev->dev, "Invalid DPI? dpi = %d", dpi->dpi);
+ return -EINVAL;
+ }
+ if (test_and_set_bit(dpi->dpi, dpit->tbl)) {
+ dev_warn(&res->pdev->dev, "Freeing an unused DPI? dpi = %d",
+ dpi->dpi);
+ return -EINVAL;
+ }
+ if (dpit->app_tbl)
+ dpit->app_tbl[dpi->dpi] = NULL;
+ memset(dpi, 0, sizeof(*dpi));
+
+ return 0;
+}
+
+static void bnxt_qplib_free_dpi_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_dpi_tbl *dpit)
+{
+ kfree(dpit->tbl);
+ kfree(dpit->app_tbl);
+ if (dpit->dbr_bar_reg_iomem)
+ pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem);
+ memset(dpit, 0, sizeof(*dpit));
+}
+
+static int bnxt_qplib_alloc_dpi_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_dpi_tbl *dpit,
+ u32 dbr_offset)
+{
+ u32 dbr_bar_reg = RCFW_DBR_PCI_BAR_REGION;
+ resource_size_t bar_reg_base;
+ u32 dbr_len, bytes;
+
+ if (dpit->dbr_bar_reg_iomem) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: DBR BAR region %d already mapped", dbr_bar_reg);
+ return -EALREADY;
+ }
+
+ bar_reg_base = pci_resource_start(res->pdev, dbr_bar_reg);
+ if (!bar_reg_base) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: BAR region %d resc start failed", dbr_bar_reg);
+ return -ENOMEM;
+ }
+
+ dbr_len = pci_resource_len(res->pdev, dbr_bar_reg) - dbr_offset;
+ if (!dbr_len || ((dbr_len & (PAGE_SIZE - 1)) != 0)) {
+ dev_err(&res->pdev->dev, "QPLIB: Invalid DBR length %d",
+ dbr_len);
+ return -ENOMEM;
+ }
+
+ dpit->dbr_bar_reg_iomem = ioremap_nocache(bar_reg_base + dbr_offset,
+ dbr_len);
+ if (!dpit->dbr_bar_reg_iomem) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: FP: DBR BAR region %d mapping failed",
+ dbr_bar_reg);
+ return -ENOMEM;
+ }
+
+ dpit->unmapped_dbr = bar_reg_base + dbr_offset;
+ dpit->max = dbr_len / PAGE_SIZE;
+
+ dpit->app_tbl = kcalloc(dpit->max, sizeof(void *), GFP_KERNEL);
+ if (!dpit->app_tbl) {
+ pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem);
+ dev_err(&res->pdev->dev,
+ "QPLIB: DPI app tbl allocation failed");
+ return -ENOMEM;
+ }
+
+ bytes = dpit->max >> 3;
+ if (!bytes)
+ bytes = 1;
+
+ dpit->tbl = kmalloc(bytes, GFP_KERNEL);
+ if (!dpit->tbl) {
+ pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem);
+ kfree(dpit->app_tbl);
+ dpit->app_tbl = NULL;
+ dev_err(&res->pdev->dev,
+ "QPLIB: DPI tbl allocation failed for size = %d",
+ bytes);
+ return -ENOMEM;
+ }
+
+ memset((u8 *)dpit->tbl, 0xFF, bytes);
+
+ return 0;
+}
+
+/* PKEYs */
+static void bnxt_qplib_cleanup_pkey_tbl(struct bnxt_qplib_pkey_tbl *pkey_tbl)
+{
+ memset(pkey_tbl->tbl, 0, sizeof(u16) * pkey_tbl->max);
+ pkey_tbl->active = 0;
+}
+
+static void bnxt_qplib_init_pkey_tbl(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl)
+{
+ u16 pkey = 0xFFFF;
+
+ memset(pkey_tbl->tbl, 0, sizeof(u16) * pkey_tbl->max);
+
+ /* pkey default = 0xFFFF */
+ bnxt_qplib_add_pkey(res, pkey_tbl, &pkey, false);
+}
+
+/* Stats */
+static void bnxt_qplib_free_stats_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_stats *stats)
+{
+ if (stats->dma) {
+ dma_free_coherent(&pdev->dev, stats->size,
+ stats->dma, stats->dma_map);
+ }
+ memset(stats, 0, sizeof(*stats));
+ stats->fw_id = -1;
+}
+
+static int bnxt_qplib_alloc_stats_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_stats *stats)
+{
+ memset(stats, 0, sizeof(*stats));
+ stats->fw_id = -1;
+ stats->size = sizeof(struct ctx_hw_stats);
+ stats->dma = dma_alloc_coherent(&pdev->dev, stats->size,
+ &stats->dma_map, GFP_KERNEL);
+ if (!stats->dma) {
+ dev_err(&pdev->dev, "QPLIB: Stats DMA allocation failed");
+ return -ENOMEM;
+ }
+ return 0;
+}
+
+void bnxt_qplib_cleanup_res(struct bnxt_qplib_res *res)
+{
+ bnxt_qplib_cleanup_pkey_tbl(&res->pkey_tbl);
+ bnxt_qplib_cleanup_sgid_tbl(res, &res->sgid_tbl);
+}
+
+int bnxt_qplib_init_res(struct bnxt_qplib_res *res)
+{
+ bnxt_qplib_init_sgid_tbl(&res->sgid_tbl, res->netdev);
+ bnxt_qplib_init_pkey_tbl(res, &res->pkey_tbl);
+
+ return 0;
+}
+
+void bnxt_qplib_free_res(struct bnxt_qplib_res *res)
+{
+ bnxt_qplib_free_pkey_tbl(res, &res->pkey_tbl);
+ bnxt_qplib_free_sgid_tbl(res, &res->sgid_tbl);
+ bnxt_qplib_free_pd_tbl(&res->pd_tbl);
+ bnxt_qplib_free_dpi_tbl(res, &res->dpi_tbl);
+
+ res->netdev = NULL;
+ res->pdev = NULL;
+}
+
+int bnxt_qplib_alloc_res(struct bnxt_qplib_res *res, struct pci_dev *pdev,
+ struct net_device *netdev,
+ struct bnxt_qplib_dev_attr *dev_attr)
+{
+ int rc = 0;
+
+ res->pdev = pdev;
+ res->netdev = netdev;
+
+ rc = bnxt_qplib_alloc_sgid_tbl(res, &res->sgid_tbl, dev_attr->max_sgid);
+ if (rc)
+ goto fail;
+
+ rc = bnxt_qplib_alloc_pkey_tbl(res, &res->pkey_tbl, dev_attr->max_pkey);
+ if (rc)
+ goto fail;
+
+ rc = bnxt_qplib_alloc_pd_tbl(res, &res->pd_tbl, dev_attr->max_pd);
+ if (rc)
+ goto fail;
+
+ rc = bnxt_qplib_alloc_dpi_tbl(res, &res->dpi_tbl, dev_attr->l2_db_size);
+ if (rc)
+ goto fail;
+
+ return 0;
+fail:
+ bnxt_qplib_free_res(res);
+ return rc;
+}
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h
index 5fc4107..ce122cf 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h
@@ -39,4 +39,169 @@
#ifndef __BNXT_QPLIB_RES_H__
#define __BNXT_QPLIB_RES_H__
+extern const struct bnxt_qplib_gid bnxt_qplib_gid_zero;
+
+#define PTR_CNT_PER_PG (PAGE_SIZE / sizeof(void *))
+#define PTR_MAX_IDX_PER_PG (PTR_CNT_PER_PG - 1)
+#define PTR_PG(x) (((x) & ~PTR_MAX_IDX_PER_PG) / PTR_CNT_PER_PG)
+#define PTR_IDX(x) ((x) & PTR_MAX_IDX_PER_PG)
+
+#define HWQ_CMP(idx, hwq) ((idx) & ((hwq)->max_elements - 1))
+
+enum bnxt_qplib_hwq_type {
+ HWQ_TYPE_CTX,
+ HWQ_TYPE_QUEUE,
+ HWQ_TYPE_L2_CMPL
+};
+
+#define MAX_PBL_LVL_0_PGS 1
+#define MAX_PBL_LVL_1_PGS 512
+#define MAX_PBL_LVL_1_PGS_SHIFT 9
+#define MAX_PBL_LVL_1_PGS_FOR_LVL_2 256
+#define MAX_PBL_LVL_2_PGS (256 * 512)
+
+enum bnxt_qplib_pbl_lvl {
+ PBL_LVL_0,
+ PBL_LVL_1,
+ PBL_LVL_2,
+ PBL_LVL_MAX
+};
+
+#define ROCE_PG_SIZE_4K (4 * 1024)
+#define ROCE_PG_SIZE_8K (8 * 1024)
+#define ROCE_PG_SIZE_64K (64 * 1024)
+#define ROCE_PG_SIZE_2M (2 * 1024 * 1024)
+#define ROCE_PG_SIZE_8M (8 * 1024 * 1024)
+#define ROCE_PG_SIZE_1G (1024 * 1024 * 1024)
+
+struct bnxt_qplib_pbl {
+ u32 pg_count;
+ u32 pg_size;
+ void **pg_arr;
+ dma_addr_t *pg_map_arr;
+};
+
+struct bnxt_qplib_hwq {
+ struct pci_dev *pdev;
+ /* lock to protect qplib_hwq */
+ spinlock_t lock;
+ struct bnxt_qplib_pbl pbl[PBL_LVL_MAX];
+ enum bnxt_qplib_pbl_lvl level; /* 0, 1, or 2 */
+ /* ptr for easy access to the PBL entries */
+ void **pbl_ptr;
+ /* ptr for easy access to the dma_addr */
+ dma_addr_t *pbl_dma_ptr;
+ u32 max_elements;
+ u16 element_size; /* Size of each entry */
+
+ u32 prod; /* raw */
+ u32 cons; /* raw */
+ u8 cp_bit;
+ u8 is_user;
+};
+
+/* Tables */
+struct bnxt_qplib_pd_tbl {
+ unsigned long *tbl;
+ u32 max;
+};
+
+struct bnxt_qplib_sgid_tbl {
+ struct bnxt_qplib_gid *tbl;
+ u16 *hw_id;
+ u16 max;
+ u16 active;
+ void *ctx;
+};
+
+struct bnxt_qplib_pkey_tbl {
+ u16 *tbl;
+ u16 max;
+ u16 active;
+};
+
+struct bnxt_qplib_dpi {
+ u32 dpi;
+ void __iomem *dbr;
+ u64 umdbr;
+};
+
+struct bnxt_qplib_dpi_tbl {
+ void **app_tbl;
+ unsigned long *tbl;
+ u16 max;
+ void __iomem *dbr_bar_reg_iomem;
+ u64 unmapped_dbr;
+};
+
+struct bnxt_qplib_stats {
+ dma_addr_t dma_map;
+ void *dma;
+ u32 size;
+ u32 fw_id;
+};
+
+struct bnxt_qplib_vf_res {
+ u32 max_qp_per_vf;
+ u32 max_mrw_per_vf;
+ u32 max_srq_per_vf;
+ u32 max_cq_per_vf;
+ u32 max_gid_per_vf;
+};
+
+#define BNXT_QPLIB_MAX_QP_CTX_ENTRY_SIZE 448
+#define BNXT_QPLIB_MAX_SRQ_CTX_ENTRY_SIZE 64
+#define BNXT_QPLIB_MAX_CQ_CTX_ENTRY_SIZE 64
+#define BNXT_QPLIB_MAX_MRW_CTX_ENTRY_SIZE 128
+
+struct bnxt_qplib_ctx {
+ u32 qpc_count;
+ struct bnxt_qplib_hwq qpc_tbl;
+ u32 mrw_count;
+ struct bnxt_qplib_hwq mrw_tbl;
+ u32 srqc_count;
+ struct bnxt_qplib_hwq srqc_tbl;
+ u32 cq_count;
+ struct bnxt_qplib_hwq cq_tbl;
+ struct bnxt_qplib_hwq tim_tbl;
+#define MAX_TQM_ALLOC_REQ 32
+#define MAX_TQM_ALLOC_BLK_SIZE 8
+ u8 tqm_count[MAX_TQM_ALLOC_REQ];
+ struct bnxt_qplib_hwq tqm_pde;
+ u32 tqm_pde_level;
+ struct bnxt_qplib_hwq tqm_tbl[MAX_TQM_ALLOC_REQ];
+ struct bnxt_qplib_stats stats;
+ struct bnxt_qplib_vf_res vf_res;
+};
+
+struct bnxt_qplib_res {
+ struct pci_dev *pdev;
+ struct net_device *netdev;
+
+ struct bnxt_qplib_rcfw *rcfw;
+
+ struct bnxt_qplib_pd_tbl pd_tbl;
+ struct bnxt_qplib_sgid_tbl sgid_tbl;
+ struct bnxt_qplib_pkey_tbl pkey_tbl;
+ struct bnxt_qplib_dpi_tbl dpi_tbl;
+};
+
+struct bnxt_qplib_dev_attr;
+
+void bnxt_qplib_free_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq);
+int bnxt_qplib_alloc_init_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq,
+ struct scatterlist *sl, int nmap, u32 *elements,
+ u32 elements_per_page, u32 aux, u32 pg_size,
+ enum bnxt_qplib_hwq_type hwq_type);
+void bnxt_qplib_cleanup_res(struct bnxt_qplib_res *res);
+int bnxt_qplib_init_res(struct bnxt_qplib_res *res);
+void bnxt_qplib_free_res(struct bnxt_qplib_res *res);
+int bnxt_qplib_alloc_res(struct bnxt_qplib_res *res, struct pci_dev *pdev,
+ struct net_device *netdev,
+ struct bnxt_qplib_dev_attr *dev_attr);
+void bnxt_qplib_free_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_ctx *ctx);
+int bnxt_qplib_alloc_ctx(struct pci_dev *pdev,
+ struct bnxt_qplib_ctx *ctx,
+ bool virt_fn);
#endif /* __BNXT_QPLIB_RES_H__ */
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
index 667c8e1..77d448e 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
@@ -35,3 +35,166 @@
*
* Description: Slow Path Operators
*/
+
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/sched.h>
+#include <linux/pci.h>
+
+#include "roce_hsi.h"
+
+#include "qplib_res.h"
+#include "qplib_rcfw.h"
+#include "qplib_sp.h"
+/* Device */
+int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw,
+ struct bnxt_qplib_dev_attr *attr)
+{
+ struct cmdq_query_func req;
+ struct creq_query_func_resp *resp;
+ struct creq_query_func_resp_sb *sb;
+ u16 cmd_flags = 0;
+ u32 temp;
+ u8 *tqm_alloc;
+ int i;
+
+ RCFW_CMD_PREP(req, QUERY_FUNC, cmd_flags);
+
+ req.resp_size = sizeof(*sb) / BNXT_QPLIB_CMDQE_UNITS;
+ resp = (struct creq_query_func_resp *)
+ bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, (void **)&sb,
+ 0);
+ if (!resp) {
+ dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC send failed");
+ return -EINVAL;
+ }
+ if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) {
+ /* Cmd timed out */
+ dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC timed out");
+ return -ETIMEDOUT;
+ }
+ if (resp->status ||
+ le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) {
+ dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC failed ");
+ dev_err(&rcfw->pdev->dev,
+ "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x",
+ resp->status, le16_to_cpu(req.cookie),
+ le16_to_cpu(resp->cookie));
+ return -EINVAL;
+ }
+ /* Extract the context from the side buffer */
+ attr->max_qp = le32_to_cpu(sb->max_qp);
+ attr->max_qp_rd_atom =
+ sb->max_qp_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ?
+ BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_rd_atom;
+ attr->max_qp_init_rd_atom =
+ sb->max_qp_init_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ?
+ BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_init_rd_atom;
+ attr->max_qp_wqes = le16_to_cpu(sb->max_qp_wr);
+ attr->max_qp_sges = sb->max_sge;
+ attr->max_cq = le32_to_cpu(sb->max_cq);
+ attr->max_cq_wqes = le32_to_cpu(sb->max_cqe);
+ attr->max_cq_sges = attr->max_qp_sges;
+ attr->max_mr = le32_to_cpu(sb->max_mr);
+ attr->max_mw = le32_to_cpu(sb->max_mw);
+
+ attr->max_mr_size = le64_to_cpu(sb->max_mr_size);
+ attr->max_pd = 64 * 1024;
+ attr->max_raw_ethy_qp = le32_to_cpu(sb->max_raw_eth_qp);
+ attr->max_ah = le32_to_cpu(sb->max_ah);
+
+ attr->max_fmr = le32_to_cpu(sb->max_fmr);
+ attr->max_map_per_fmr = sb->max_map_per_fmr;
+
+ attr->max_srq = le16_to_cpu(sb->max_srq);
+ attr->max_srq_wqes = le32_to_cpu(sb->max_srq_wr) - 1;
+ attr->max_srq_sges = sb->max_srq_sge;
+ /* Bono only reports 1 PKEY for now, but it can support > 1 */
+ attr->max_pkey = le32_to_cpu(sb->max_pkeys);
+
+ attr->max_inline_data = le32_to_cpu(sb->max_inline_data);
+ attr->l2_db_size = (sb->l2_db_space_size + 1) * PAGE_SIZE;
+ attr->max_sgid = le32_to_cpu(sb->max_gid);
+
+ strlcpy(attr->fw_ver, "20.6.28.0", sizeof(attr->fw_ver));
+
+ for (i = 0; i < MAX_TQM_ALLOC_REQ / 4; i++) {
+ temp = le32_to_cpu(sb->tqm_alloc_reqs[i]);
+ tqm_alloc = (u8 *)&temp;
+ attr->tqm_alloc_reqs[i * 4] = *tqm_alloc;
+ attr->tqm_alloc_reqs[i * 4 + 1] = *(++tqm_alloc);
+ attr->tqm_alloc_reqs[i * 4 + 2] = *(++tqm_alloc);
+ attr->tqm_alloc_reqs[i * 4 + 3] = *(++tqm_alloc);
+ }
+ return 0;
+}
+
+int bnxt_qplib_del_pkey(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey,
+ bool update)
+{
+ int i, rc = 0;
+
+ if (!pkey_tbl) {
+ dev_err(&res->pdev->dev, "QPLIB: PKEY table not allocated");
+ return -EINVAL;
+ }
+
+ /* Do we need a pkey_lock here? */
+ if (!pkey_tbl->active) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: PKEY table has no active entries");
+ return -ENOMEM;
+ }
+ for (i = 0; i < pkey_tbl->max; i++) {
+ if (!memcmp(&pkey_tbl->tbl[i], pkey, sizeof(*pkey)))
+ break;
+ }
+ if (i == pkey_tbl->max) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: PKEY 0x%04x not found in the pkey table",
+ *pkey);
+ return -ENOMEM;
+ }
+ memset(&pkey_tbl->tbl[i], 0, sizeof(*pkey));
+ pkey_tbl->active--;
+
+ /* unlock */
+ return rc;
+}
+
+int bnxt_qplib_add_pkey(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey,
+ bool update)
+{
+ int i, free_idx, rc = 0;
+
+ if (!pkey_tbl) {
+ dev_err(&res->pdev->dev, "QPLIB: PKEY table not allocated");
+ return -EINVAL;
+ }
+
+ /* Do we need a pkey_lock here? */
+ if (pkey_tbl->active == pkey_tbl->max) {
+ dev_err(&res->pdev->dev, "QPLIB: PKEY table is full");
+ return -ENOMEM;
+ }
+ free_idx = pkey_tbl->max;
+ for (i = 0; i < pkey_tbl->max; i++) {
+ if (!memcmp(&pkey_tbl->tbl[i], pkey, sizeof(*pkey)))
+ return -EALREADY;
+ else if (!pkey_tbl->tbl[i] && free_idx == pkey_tbl->max)
+ free_idx = i;
+ }
+ if (free_idx == pkey_tbl->max) {
+ dev_err(&res->pdev->dev,
+ "QPLIB: PKEY table is FULL but count is not MAX??");
+ return -ENOMEM;
+ }
+ /* Add PKEY to the pkey_tbl */
+ memcpy(&pkey_tbl->tbl[free_idx], pkey, sizeof(*pkey));
+ pkey_tbl->active++;
+
+ /* unlock */
+ return rc;
+}
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.h b/drivers/infiniband/hw/bnxt_re/qplib_sp.h
index 90901c1..0b5adda 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_sp.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.h
@@ -40,4 +40,46 @@
#ifndef __BNXT_QPLIB_SP_H__
#define __BNXT_QPLIB_SP_H__
+struct bnxt_qplib_dev_attr {
+ char fw_ver[32];
+ u16 max_sgid;
+ u16 max_mrw;
+ u32 max_qp;
+#define BNXT_QPLIB_MAX_OUT_RD_ATOM 126
+ u32 max_qp_rd_atom;
+ u32 max_qp_init_rd_atom;
+ u32 max_qp_wqes;
+ u32 max_qp_sges;
+ u32 max_cq;
+ u32 max_cq_wqes;
+ u32 max_cq_sges;
+ u32 max_mr;
+ u64 max_mr_size;
+ u32 max_pd;
+ u32 max_mw;
+ u32 max_raw_ethy_qp;
+ u32 max_ah;
+ u32 max_fmr;
+ u32 max_map_per_fmr;
+ u32 max_srq;
+ u32 max_srq_wqes;
+ u32 max_srq_sges;
+ u32 max_pkey;
+ u32 max_inline_data;
+ u32 l2_db_size;
+ u8 tqm_alloc_reqs[MAX_TQM_ALLOC_REQ];
+};
+
+struct bnxt_qplib_gid {
+ u8 data[16];
+};
+
+int bnxt_qplib_del_pkey(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey,
+ bool update);
+int bnxt_qplib_add_pkey(struct bnxt_qplib_res *res,
+ struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey,
+ bool update);
+int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw,
+ struct bnxt_qplib_dev_attr *attr);
#endif /* __BNXT_QPLIB_SP_H__*/
--
2.5.5
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: Soft lockup in tc_classify
From: Shahar Klein @ 2016-12-21 11:25 UTC (permalink / raw)
To: Cong Wang
Cc: shahark, Daniel Borkmann, Or Gerlitz, Linux Netdev List,
Roi Dayan, David Miller, Jiri Pirko, John Fastabend,
Hadar Hen Zion
In-Reply-To: <CAM_iQpX3jL4URUQn5+-wV+3g2UKHWGBWYHr31wyZk2FQKVCzVA@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 1003 bytes --]
On 12/21/2016 9:03 AM, Cong Wang wrote:
> On Tue, Dec 20, 2016 at 10:44 PM, Shahar Klein <shahark@mellanox.com> wrote:
>>
>> Tried it with same results
>
> This piece is pretty interesting:
>
> [ 408.554689] DEBUGG:SK thread-2853[cpu-1] setting tp_created to 1
> tp=ffff94b5b02805a0 back=ffff94b9ea932060
> [ 408.574258] DEBUGG:SK thread-2853[cpu-1] add/change filter by:
> fl_get [cls_flower] tp=ffff94b5b02805a0 tp->next=ffff94b9ea932060
> [ 408.587849] DEBUGG:SK destroy ffff94b5b0280780 tcf_destroy:1905
> [ 408.595862] DEBUGG:SK thread-2845[cpu-1] add/change filter by:
> fl_get [cls_flower] tp=ffff94b5b02805a0 tp->next=ffff94b5b02805a0
>
> Looks like you added a debug printk inside tcf_destroy() too,
> which seems racy with filter creation, it should not happen since
> in both cases we take RTNL lock.
>
> Don't know if changing all RCU_INIT_POINTER in that file to
> rcu_assign_pointer could help anything or not. Mind to try?
>
Tried it with same results
>
> Thanks for debugging!
>
[-- Attachment #2: cls_api_no_RCU.c.diff --]
[-- Type: text/plain, Size: 2704 bytes --]
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 3fbba79..b8a66d8 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -305,7 +305,7 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n)
kfree(tp);
goto errout;
}
-
+ printk(KERN_ERR "DEBUGG:SK thread-%d[cpu-%d] setting tp_created to 1 tp=%p back=%p\n", current->pid, current->on_cpu, tp, rtnl_dereference(*back));
tp_created = 1;
} else if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], tp->ops->kind))
@@ -317,11 +317,13 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n)
if (n->nlmsg_type == RTM_DELTFILTER && t->tcm_handle == 0) {
struct tcf_proto *next = rtnl_dereference(tp->next);
- RCU_INIT_POINTER(*back, next);
+ printk(KERN_ERR "DEBUGG:SK delete filter by: %pf\n", tp->ops->get);
+
+ rcu_assign_pointer(*back, next);
tfilter_notify(net, skb, n, tp, fh,
RTM_DELTFILTER, false);
- tcf_destroy(tp, true);
+ tcf_destroy(tp);
err = 0;
goto errout;
}
@@ -331,25 +333,30 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n)
!(n->nlmsg_flags & NLM_F_CREATE))
goto errout;
} else {
+ bool last;
+
switch (n->nlmsg_type) {
case RTM_NEWTFILTER:
err = -EEXIST;
if (n->nlmsg_flags & NLM_F_EXCL) {
if (tp_created)
- tcf_destroy(tp, true);
+ tcf_destroy(tp);
goto errout;
}
break;
case RTM_DELTFILTER:
- err = tp->ops->delete(tp, fh);
+ printk(KERN_ERR "DEBUGG:SK %s:%d\n", __func__, __LINE__);
+ err = tp->ops->delete(tp, fh, &last);
if (err == 0) {
- struct tcf_proto *next = rtnl_dereference(tp->next);
-
tfilter_notify(net, skb, n, tp,
t->tcm_handle,
RTM_DELTFILTER, false);
- if (tcf_destroy(tp, false))
- RCU_INIT_POINTER(*back, next);
+ if (last) {
+ struct tcf_proto *next = rtnl_dereference(tp->next);
+
+ rcu_assign_pointer(*back, next);
+ tcf_destroy(tp);
+ }
}
goto errout;
case RTM_GETTFILTER:
@@ -366,13 +373,14 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n)
n->nlmsg_flags & NLM_F_CREATE ? TCA_ACT_NOREPLACE : TCA_ACT_REPLACE);
if (err == 0) {
if (tp_created) {
- RCU_INIT_POINTER(tp->next, rtnl_dereference(*back));
+ rcu_assign_pointer(tp->next, rtnl_dereference(*back));
rcu_assign_pointer(*back, tp);
+ printk(KERN_ERR "DEBUGG:SK thread-%d[cpu-%d] add/change filter by: %pf tp=%p tp->next=%p\n", current->pid, current->on_cpu, tp->ops->get, tp, tp->next);
}
tfilter_notify(net, skb, n, tp, fh, RTM_NEWTFILTER, false);
} else {
if (tp_created)
- tcf_destroy(tp, true);
+ tcf_destroy(tp);
}
errout:
^ permalink raw reply related
* [PATCH net-next 27/27] net: mvpp2: finally add the PPv2.2 compatible string
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
Now that the mvpp2 driver has been modified to accommodate the support
for PPv2.2, we can finally advertise this support by adding the
appropriate compatible string.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index d9368d6..2dc1003 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -7014,6 +7014,10 @@ static const struct of_device_id mvpp2_match[] = {
.compatible = "marvell,armada-375-pp2",
.data = (void *)MVPP21,
},
+ {
+ .compatible = "marvell,armada-7k-pp22",
+ .data = (void *)MVPP22,
+ },
{ }
};
MODULE_DEVICE_TABLE(of, mvpp2_match);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 24/27] net: mvpp2: rework RXQ interrupt group initialization for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit adjusts how the MVPP2_ISR_RXQ_GROUP_REG register is
configured, since it changed between PPv2.1 and PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 45 ++++++++++++++++++++++++++++++++----
1 file changed, 41 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 6c17c7d..d061a1a 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -188,7 +188,21 @@
/* Interrupt Cause and Mask registers */
#define MVPP2_ISR_RX_THRESHOLD_REG(rxq) (0x5200 + 4 * (rxq))
#define MVPP2_MAX_ISR_RX_THRESHOLD 0xfffff0
-#define MVPP2_ISR_RXQ_GROUP_REG(rxq) (0x5400 + 4 * (rxq))
+#define MVPP21_ISR_RXQ_GROUP_REG(rxq) (0x5400 + 4 * (rxq))
+
+#define MVPP22_ISR_RXQ_GROUP_INDEX_REG 0x5400
+#define MVPP22_ISR_RXQ_GROUP_INDEX_SUBGROUP_MASK 0xf
+#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_MASK 0x380
+#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_OFFSET 7
+
+#define MVPP22_ISR_RXQ_GROUP_INDEX_SUBGROUP_MASK 0xf
+#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_MASK 0x380
+
+#define MVPP22_ISR_RXQ_SUB_GROUP_CONFIG_REG 0x5404
+#define MVPP22_ISR_RXQ_SUB_GROUP_STARTQ_MASK 0x1f
+#define MVPP22_ISR_RXQ_SUB_GROUP_SIZE_MASK 0xf00
+#define MVPP22_ISR_RXQ_SUB_GROUP_SIZE_OFFSET 8
+
#define MVPP2_ISR_ENABLE_REG(port) (0x5420 + 4 * (port))
#define MVPP2_ISR_ENABLE_INTERRUPT(mask) ((mask) & 0xffff)
#define MVPP2_ISR_DISABLE_INTERRUPT(mask) (((mask) << 16) & 0xffff0000)
@@ -6407,7 +6421,18 @@ static int mvpp2_port_init(struct mvpp2_port *port)
}
/* Configure Rx queue group interrupt for this port */
- mvpp2_write(priv, MVPP2_ISR_RXQ_GROUP_REG(port->id), rxq_number);
+ if (priv->ip_version == MVPP21)
+ mvpp2_write(priv, MVPP21_ISR_RXQ_GROUP_REG(port->id),
+ rxq_number);
+ else {
+ u32 val;
+
+ val = (port->id << MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_OFFSET);
+ mvpp2_write(priv, MVPP22_ISR_RXQ_GROUP_INDEX_REG, val);
+
+ val = (rxq_number << MVPP22_ISR_RXQ_SUB_GROUP_SIZE_OFFSET);
+ mvpp2_write(priv, MVPP22_ISR_RXQ_SUB_GROUP_CONFIG_REG, val);
+ }
/* Create Rx descriptor rings */
for (queue = 0; queue < rxq_number; queue++) {
@@ -6792,8 +6817,20 @@ static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
mvpp2_rx_fifo_init(priv);
/* Reset Rx queue group interrupt configuration */
- for (i = 0; i < MVPP2_MAX_PORTS; i++)
- mvpp2_write(priv, MVPP2_ISR_RXQ_GROUP_REG(i), rxq_number);
+ for (i = 0; i < MVPP2_MAX_PORTS; i++) {
+ if (priv->ip_version == MVPP21)
+ mvpp2_write(priv, MVPP21_ISR_RXQ_GROUP_REG(i),
+ rxq_number);
+ else {
+ u32 val;
+
+ val = (i << MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_OFFSET);
+ mvpp2_write(priv, MVPP22_ISR_RXQ_GROUP_INDEX_REG, val);
+
+ val = (rxq_number << MVPP22_ISR_RXQ_SUB_GROUP_SIZE_OFFSET);
+ mvpp2_write(priv, MVPP22_ISR_RXQ_SUB_GROUP_CONFIG_REG, val);
+ }
+ }
if (priv->ip_version == MVPP21)
writel(MVPP2_EXT_GLOBAL_CTRL_DEFAULT,
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 23/27] net: mvpp2: add AXI bridge initialization for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
The PPv2.2 unit is connected to an AXI bus on Armada 7K/8K, so this
commit adds the necessary initialization of the AXI bridge.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 85 ++++++++++++++++++++++++++++++++++++
1 file changed, 85 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 9b5a7c2..6c17c7d 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -157,6 +157,34 @@
#define MVPP2_WIN_REMAP(w) (0x4040 + ((w) << 2))
#define MVPP2_BASE_ADDR_ENABLE 0x4060
+/* AXI Bridge Registers */
+#define MVPP22_AXI_BM_WR_ATTR_REG 0x4100
+#define MVPP22_AXI_BM_RD_ATTR_REG 0x4104
+#define MVPP22_AXI_AGGRQ_DESCR_RD_ATTR_REG 0x4110
+#define MVPP22_AXI_TXQ_DESCR_WR_ATTR_REG 0x4114
+#define MVPP22_AXI_TXQ_DESCR_RD_ATTR_REG 0x4118
+#define MVPP22_AXI_RXQ_DESCR_WR_ATTR_REG 0x411c
+#define MVPP22_AXI_RX_DATA_WR_ATTR_REG 0x4120
+#define MVPP22_AXI_TX_DATA_RD_ATTR_REG 0x4130
+#define MVPP22_AXI_RD_NORMAL_CODE_REG 0x4150
+#define MVPP22_AXI_RD_SNOOP_CODE_REG 0x4154
+#define MVPP22_AXI_WR_NORMAL_CODE_REG 0x4160
+#define MVPP22_AXI_WR_SNOOP_CODE_REG 0x4164
+
+/* Values for AXI Bridge registers */
+#define MVPP22_AXI_ATTR_CACHE_OFFS 0
+#define MVPP22_AXI_ATTR_DOMAIN_OFFS 12
+
+#define MVPP22_AXI_CODE_CACHE_OFFS 0
+#define MVPP22_AXI_CODE_DOMAIN_OFFS 4
+
+#define MVPP22_AXI_CODE_CACHE_NON_CACHE 0x3
+#define MVPP22_AXI_CODE_CACHE_WR_CACHE 0x7
+#define MVPP22_AXI_CODE_CACHE_RD_CACHE 0xb
+
+#define MVPP22_AXI_CODE_DOMAIN_OUTER_DOM 2
+#define MVPP22_AXI_CODE_DOMAIN_SYSTEM 3
+
/* Interrupt Cause and Mask registers */
#define MVPP2_ISR_RX_THRESHOLD_REG(rxq) (0x5200 + 4 * (rxq))
#define MVPP2_MAX_ISR_RX_THRESHOLD 0xfffff0
@@ -6657,6 +6685,60 @@ static void mvpp2_rx_fifo_init(struct mvpp2 *priv)
mvpp2_write(priv, MVPP2_RX_FIFO_INIT_REG, 0x1);
}
+static void mvpp2_axi_init(struct mvpp2 *priv)
+{
+ u32 val, rdval, wrval;
+
+ mvpp2_write(priv, MVPP22_BM_ADDR_HIGH_RLS_REG, 0x0);
+
+ /* AXI Bridge Configuration */
+
+ rdval = MVPP22_AXI_CODE_CACHE_RD_CACHE
+ << MVPP22_AXI_ATTR_CACHE_OFFS;
+ rdval |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
+ << MVPP22_AXI_ATTR_DOMAIN_OFFS;
+
+ wrval = MVPP22_AXI_CODE_CACHE_WR_CACHE
+ << MVPP22_AXI_ATTR_CACHE_OFFS;
+ wrval |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
+ << MVPP22_AXI_ATTR_DOMAIN_OFFS;
+
+ /* BM */
+ mvpp2_write(priv, MVPP22_AXI_BM_WR_ATTR_REG, wrval);
+ mvpp2_write(priv, MVPP22_AXI_BM_RD_ATTR_REG, rdval);
+
+ /* Descriptors */
+ mvpp2_write(priv, MVPP22_AXI_AGGRQ_DESCR_RD_ATTR_REG, rdval);
+ mvpp2_write(priv, MVPP22_AXI_TXQ_DESCR_WR_ATTR_REG, wrval);
+ mvpp2_write(priv, MVPP22_AXI_TXQ_DESCR_RD_ATTR_REG, rdval);
+ mvpp2_write(priv, MVPP22_AXI_RXQ_DESCR_WR_ATTR_REG, wrval);
+
+ /* Buffer Data */
+ mvpp2_write(priv, MVPP22_AXI_TX_DATA_RD_ATTR_REG, rdval);
+ mvpp2_write(priv, MVPP22_AXI_RX_DATA_WR_ATTR_REG, wrval);
+
+ val = MVPP22_AXI_CODE_CACHE_NON_CACHE
+ << MVPP22_AXI_CODE_CACHE_OFFS;
+ val |= MVPP22_AXI_CODE_DOMAIN_SYSTEM
+ << MVPP22_AXI_CODE_DOMAIN_OFFS;
+ mvpp2_write(priv, MVPP22_AXI_RD_NORMAL_CODE_REG, val);
+ mvpp2_write(priv, MVPP22_AXI_WR_NORMAL_CODE_REG, val);
+
+ val = MVPP22_AXI_CODE_CACHE_RD_CACHE
+ << MVPP22_AXI_CODE_CACHE_OFFS;
+ val |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
+ << MVPP22_AXI_CODE_DOMAIN_OFFS;
+
+ mvpp2_write(priv, MVPP22_AXI_RD_SNOOP_CODE_REG, val);
+
+ val = MVPP22_AXI_CODE_CACHE_WR_CACHE
+ << MVPP22_AXI_CODE_CACHE_OFFS;
+ val |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
+ << MVPP22_AXI_CODE_DOMAIN_OFFS;
+
+ mvpp2_write(priv, MVPP22_AXI_WR_SNOOP_CODE_REG, val);
+}
+
/* Initialize network controller common part HW */
static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
{
@@ -6676,6 +6758,9 @@ static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
if (dram_target_info)
mvpp2_conf_mbus_windows(dram_target_info, priv);
+ if (priv->ip_version == MVPP22)
+ mvpp2_axi_init(priv);
+
/* Disable HW PHY polling */
if (priv->ip_version == MVPP21) {
val = readl(priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 25/27] net: mvpp2: adapt rxq distribution to PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
In PPv2.1, we have a maximum of 8 RXQs per port, with a default of 4
RXQs per port, and we were assigning RXQs 0->3 to the first port, 4->7
to the second port, 8->11 to the third port, etc.
In PPv2.2, we have a maximum of 32 RXQs per port, and we must allocate
RXQs from the range of 32 RXQs available for each port. So port 0 must
use RXQs in the range 0->31, port 1 in the range 32->63, etc.
This commit adapts the mvpp2 to this difference between PPv2.1 and
PPv2.2:
- The constant definition MVPP2_MAX_RXQ is replaced by a new field
'max_port_rxqs' in 'struct mvpp2', which stores the maximum number of
RXQs per port. This field is initialized during ->probe() depending
on the IP version.
- MVPP2_RXQ_TOTAL_NUM is removed, and instead we calculate the total
number of RXQs by multiplying the number of ports by the maximum of
RXQs per port. This was anyway used in only one place.
- In mvpp2_port_probe(), the calculation of port->first_rxq is adjusted
to cope with the different allocation strategy between PPv2.1 and
PPv2.2. Due to this change, the 'next_first_rxq' argument of this
function is no longer needed and is removed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 35 +++++++++++++++++++----------------
1 file changed, 19 insertions(+), 16 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index d061a1a..c389590 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -399,15 +399,9 @@
/* Maximum number of TXQs used by single port */
#define MVPP2_MAX_TXQ 8
-/* Maximum number of RXQs used by single port */
-#define MVPP2_MAX_RXQ 8
-
/* Dfault number of RXQs in use */
#define MVPP2_DEFAULT_RXQ 4
-/* Total number of RXQs available to all ports */
-#define MVPP2_RXQ_TOTAL_NUM (MVPP2_MAX_PORTS * MVPP2_MAX_RXQ)
-
/* Max number of Rx descriptors */
#define MVPP2_MAX_RXD 128
@@ -728,6 +722,9 @@ struct mvpp2 {
/* HW IP Version */
enum { MVPP21, MVPP22 } ip_version;
+
+ /* Maximum number of RXQs per port */
+ unsigned int max_port_rxqs;
};
struct mvpp2_pcpu_stats {
@@ -6355,7 +6352,8 @@ static int mvpp2_port_init(struct mvpp2_port *port)
struct mvpp2_txq_pcpu *txq_pcpu;
int queue, cpu, err;
- if (port->first_rxq + rxq_number > MVPP2_RXQ_TOTAL_NUM)
+ if (port->first_rxq + rxq_number >
+ MVPP2_MAX_PORTS * priv->max_port_rxqs)
return -EINVAL;
/* Disable port */
@@ -6474,8 +6472,7 @@ static int mvpp2_port_init(struct mvpp2_port *port)
/* Ports initialization */
static int mvpp2_port_probe(struct platform_device *pdev,
struct device_node *port_node,
- struct mvpp2 *priv,
- int *next_first_rxq)
+ struct mvpp2 *priv)
{
struct device_node *phy_node;
struct mvpp2_port *port;
@@ -6533,7 +6530,11 @@ static int mvpp2_port_probe(struct platform_device *pdev,
port->priv = priv;
port->id = id;
- port->first_rxq = *next_first_rxq;
+ if (priv->ip_version == MVPP21)
+ port->first_rxq = port->id * rxq_number;
+ else
+ port->first_rxq = port->id * priv->max_port_rxqs;
+
port->phy_node = phy_node;
port->phy_interface = phy_mode;
@@ -6625,8 +6626,6 @@ static int mvpp2_port_probe(struct platform_device *pdev,
}
netdev_info(dev, "Using %s mac address %pM\n", mac_from, dev->dev_addr);
- /* Increment the first Rx queue number to be used by the next port */
- *next_first_rxq += rxq_number;
priv->port_list[id] = port;
return 0;
@@ -6772,7 +6771,7 @@ static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
u32 val;
/* Checks for hardware constraints */
- if (rxq_number % 4 || (rxq_number > MVPP2_MAX_RXQ) ||
+ if (rxq_number % 4 || (rxq_number > priv->max_port_rxqs) ||
(txq_number > MVPP2_MAX_TXQ)) {
dev_err(&pdev->dev, "invalid queue size parameter\n");
return -EINVAL;
@@ -6861,7 +6860,7 @@ static int mvpp2_probe(struct platform_device *pdev)
struct device_node *port_node;
struct mvpp2 *priv;
struct resource *res;
- int port_count, first_rxq, cpu;
+ int port_count, cpu;
int err;
priv = devm_kzalloc(&pdev->dev, sizeof(struct mvpp2), GFP_KERNEL);
@@ -6896,6 +6895,11 @@ static int mvpp2_probe(struct platform_device *pdev)
priv->cpu_base[cpu] = priv->base + cpu * addr_space_sz;
}
+ if (priv->ip_version == MVPP21)
+ priv->max_port_rxqs = 8;
+ else
+ priv->max_port_rxqs = 32;
+
priv->pp_clk = devm_clk_get(&pdev->dev, "pp_clk");
if (IS_ERR(priv->pp_clk))
return PTR_ERR(priv->pp_clk);
@@ -6938,9 +6942,8 @@ static int mvpp2_probe(struct platform_device *pdev)
}
/* Initialize ports */
- first_rxq = 0;
for_each_available_child_of_node(dn, port_node) {
- err = mvpp2_port_probe(pdev, port_node, priv, &first_rxq);
+ err = mvpp2_port_probe(pdev, port_node, priv);
if (err < 0)
goto err_gop_clk;
}
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 26/27] net: mvpp2: add support for an additional clock needed for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
The PPv2.2 variant of the network controller needs an additional
clock, the "MG clock" in order for the IP block to operate
properly. This commit adds support for this additional clock to the
driver, reworking as needed the error handling path.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 25 +++++++++++++++++++++----
1 file changed, 21 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index c389590..d9368d6 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -702,6 +702,7 @@ struct mvpp2 {
/* Common clocks */
struct clk *pp_clk;
struct clk *gop_clk;
+ struct clk *mg_clk;
/* List of pointers to port structures */
struct mvpp2_port **port_list;
@@ -6916,6 +6917,18 @@ static int mvpp2_probe(struct platform_device *pdev)
if (err < 0)
goto err_pp_clk;
+ if (priv->ip_version == MVPP22) {
+ priv->mg_clk = devm_clk_get(&pdev->dev, "mg_clk");
+ if (IS_ERR(priv->mg_clk)) {
+ err = PTR_ERR(priv->mg_clk);
+ goto err_gop_clk;
+ }
+
+ err = clk_prepare_enable(priv->mg_clk);
+ if (err < 0)
+ goto err_gop_clk;
+ }
+
/* Get system's tclk rate */
priv->tclk = clk_get_rate(priv->pp_clk);
@@ -6923,14 +6936,14 @@ static int mvpp2_probe(struct platform_device *pdev)
err = mvpp2_init(pdev, priv);
if (err < 0) {
dev_err(&pdev->dev, "failed to initialize controller\n");
- goto err_gop_clk;
+ goto err_mg_clk;
}
port_count = of_get_available_child_count(dn);
if (port_count == 0) {
dev_err(&pdev->dev, "no ports enabled\n");
err = -ENODEV;
- goto err_gop_clk;
+ goto err_mg_clk;
}
priv->port_list = devm_kcalloc(&pdev->dev, port_count,
@@ -6938,19 +6951,22 @@ static int mvpp2_probe(struct platform_device *pdev)
GFP_KERNEL);
if (!priv->port_list) {
err = -ENOMEM;
- goto err_gop_clk;
+ goto err_mg_clk;
}
/* Initialize ports */
for_each_available_child_of_node(dn, port_node) {
err = mvpp2_port_probe(pdev, port_node, priv);
if (err < 0)
- goto err_gop_clk;
+ goto err_mg_clk;
}
platform_set_drvdata(pdev, priv);
return 0;
+err_mg_clk:
+ if (priv->ip_version == MVPP22)
+ clk_disable_unprepare(priv->mg_clk);
err_gop_clk:
clk_disable_unprepare(priv->gop_clk);
err_pp_clk:
@@ -6986,6 +7002,7 @@ static int mvpp2_remove(struct platform_device *pdev)
aggr_txq->descs_phys);
}
+ clk_disable_unprepare(priv->mg_clk);
clk_disable_unprepare(priv->pp_clk);
clk_disable_unprepare(priv->gop_clk);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 21/27] net: mvpp2: handle register mapping and access for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit adjusts the mvpp2 driver register mapping and access logic
to support PPv2.2, to handle a number of differences.
Due to how the registers are laid out in memory, the Device Tree binding
for the "reg" property is different:
- On PPv2.1, we had a first area for the common registers, and then one
area per port.
- On PPv2.2, we have a first area for the common registers, and a
second area for all the per-ports registers.
In addition, on PPv2.2, the area for the common registers is split into
so-called "address spaces" of 64 KB each. They allow to access the same
registers, but from different CPUs. Hence the introduction of cpu_base[]
in 'struct mvpp2', and the modification of the mvpp2_write() and
mvpp2_read() register accessors. For PPv2.1, the compatibility is
preserved by using an "address space" size of 0.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 78 +++++++++++++++++++++++++++++-------
1 file changed, 64 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index eb9fe86..4163b0e 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -304,6 +304,9 @@
#define MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(v) (((v) << 6) & \
MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK)
+#define MVPP22_PORT_BASE 0x30e00
+#define MVPP22_PORT_OFFSET 0x1000
+
#define MVPP2_CAUSE_TXQ_SENT_DESC_ALL_MASK 0xff
/* Descriptor ring Macros */
@@ -631,6 +634,11 @@ enum mvpp2_prs_l3_cast {
*/
#define MVPP2_BM_SHORT_PKT_SIZE MVPP2_RX_MAX_PKT_SIZE(512)
+#define MVPP21_ADDR_SPACE_SZ 0
+#define MVPP22_ADDR_SPACE_SZ SZ_64K
+
+#define MVPP2_MAX_CPUS 4
+
enum mvpp2_bm_type {
MVPP2_BM_FREE,
MVPP2_BM_SWF_LONG,
@@ -644,6 +652,13 @@ struct mvpp2 {
/* Shared registers' base addresses */
void __iomem *base;
void __iomem *lms_base;
+ void __iomem *iface_base;
+
+ /* On PPv2.2, each CPU can access the base register through a
+ * separate address space, each 64 KB apart from each
+ * other.
+ */
+ void __iomem *cpu_base[MVPP2_MAX_CPUS];
/* Common clocks */
struct clk *pp_clk;
@@ -1021,12 +1036,21 @@ static int txq_number = MVPP2_MAX_TXQ;
static void mvpp2_write(struct mvpp2 *priv, u32 offset, u32 data)
{
- writel(data, priv->base + offset);
+ int cpu = get_cpu();
+
+ writel(data, priv->cpu_base[cpu] + offset);
+ put_cpu();
}
static u32 mvpp2_read(struct mvpp2 *priv, u32 offset)
{
- return readl(priv->base + offset);
+ int cpu = get_cpu();
+ u32 val;
+
+ val = readl(priv->cpu_base[cpu] + offset);
+ put_cpu();
+
+ return val;
}
static dma_addr_t mvpp2_txdesc_phys_addr_get(struct mvpp2_port *port,
@@ -6408,7 +6432,6 @@ static int mvpp2_port_probe(struct platform_device *pdev,
u32 id;
int features;
int phy_mode;
- int priv_common_regs_num = 2;
int err, i, cpu;
dev = alloc_etherdev_mqs(sizeof(struct mvpp2_port), txq_number,
@@ -6458,12 +6481,24 @@ static int mvpp2_port_probe(struct platform_device *pdev,
port->phy_node = phy_node;
port->phy_interface = phy_mode;
- res = platform_get_resource(pdev, IORESOURCE_MEM,
- priv_common_regs_num + id);
- port->base = devm_ioremap_resource(&pdev->dev, res);
- if (IS_ERR(port->base)) {
- err = PTR_ERR(port->base);
- goto err_free_irq;
+ if (priv->ip_version == MVPP21) {
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 2 + id);
+ port->base = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(port->base)) {
+ err = PTR_ERR(port->base);
+ goto err_free_irq;
+ }
+ } else {
+ u32 gop_id;
+
+ if (of_property_read_u32(port_node, "gop-port-id", &gop_id)) {
+ err = -EINVAL;
+ dev_err(&pdev->dev, "missing gop-port-id value\n");
+ goto err_free_irq;
+ }
+
+ port->base = priv->iface_base + MVPP22_PORT_BASE +
+ gop_id * MVPP22_PORT_OFFSET;
}
/* Alloc per-cpu stats */
@@ -6691,7 +6726,7 @@ static int mvpp2_probe(struct platform_device *pdev)
struct device_node *port_node;
struct mvpp2 *priv;
struct resource *res;
- int port_count, first_rxq;
+ int port_count, first_rxq, cpu;
int err;
priv = devm_kzalloc(&pdev->dev, sizeof(struct mvpp2), GFP_KERNEL);
@@ -6706,10 +6741,25 @@ static int mvpp2_probe(struct platform_device *pdev)
if (IS_ERR(priv->base))
return PTR_ERR(priv->base);
- res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
- priv->lms_base = devm_ioremap_resource(&pdev->dev, res);
- if (IS_ERR(priv->lms_base))
- return PTR_ERR(priv->lms_base);
+ if (priv->ip_version == MVPP21) {
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+ priv->lms_base = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(priv->lms_base))
+ return PTR_ERR(priv->lms_base);
+ } else {
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+ priv->iface_base = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(priv->iface_base))
+ return PTR_ERR(priv->iface_base);
+ }
+
+ for_each_present_cpu(cpu) {
+ u32 addr_space_sz;
+
+ addr_space_sz = (priv->ip_version == MVPP21 ?
+ MVPP21_ADDR_SPACE_SZ : MVPP22_ADDR_SPACE_SZ);
+ priv->cpu_base[cpu] = priv->base + cpu * addr_space_sz;
+ }
priv->pp_clk = devm_clk_get(&pdev->dev, "pp_clk");
if (IS_ERR(priv->pp_clk))
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 20/27] net: mvpp2: adjust mvpp2_{rxq,txq}_init for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
In PPv2.2, the MVPP2_RXQ_DESC_ADDR_REG and MVPP2_TXQ_DESC_ADDR_REG
registers have a slightly different layout, because they need to contain
a 64-bit address for the RX and TX descriptor arrays. This commit
adjusts those functions accordingly.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 26 +++++++++++++++++++++-----
1 file changed, 21 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 1394ed9..eb9fe86 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -102,6 +102,7 @@
/* Descriptor Manager Top Registers */
#define MVPP2_RXQ_NUM_REG 0x2040
#define MVPP2_RXQ_DESC_ADDR_REG 0x2044
+#define MVPP22_DESC_ADDR_OFFS 8
#define MVPP2_RXQ_DESC_SIZE_REG 0x2048
#define MVPP2_RXQ_DESC_SIZE_MASK 0x3ff0
#define MVPP2_RXQ_STATUS_UPDATE_REG(rxq) (0x3000 + 4 * (rxq))
@@ -143,6 +144,7 @@
#define MVPP2_TXQ_RSVD_CLR_REG 0x20b8
#define MVPP2_TXQ_RSVD_CLR_OFFSET 16
#define MVPP2_AGGR_TXQ_DESC_ADDR_REG(cpu) (0x2100 + 4 * (cpu))
+#define MVPP22_AGGR_TXQ_DESC_ADDR_OFFS 8
#define MVPP2_AGGR_TXQ_DESC_SIZE_REG(cpu) (0x2140 + 4 * (cpu))
#define MVPP2_AGGR_TXQ_DESC_SIZE_MASK 0x3ff0
#define MVPP2_AGGR_TXQ_STATUS_REG(cpu) (0x2180 + 4 * (cpu))
@@ -4769,6 +4771,8 @@ static int mvpp2_aggr_txq_init(struct platform_device *pdev,
int desc_num, int cpu,
struct mvpp2 *priv)
{
+ u32 txq_phys;
+
/* Allocate memory for TX descriptors */
aggr_txq->descs = dma_alloc_coherent(&pdev->dev,
desc_num * MVPP2_DESC_ALIGNED_SIZE,
@@ -4782,10 +4786,16 @@ static int mvpp2_aggr_txq_init(struct platform_device *pdev,
aggr_txq->next_desc_to_proc = mvpp2_read(priv,
MVPP2_AGGR_TXQ_INDEX_REG(cpu));
- /* Set Tx descriptors queue starting address */
- /* indirect access */
- mvpp2_write(priv, MVPP2_AGGR_TXQ_DESC_ADDR_REG(cpu),
- aggr_txq->descs_phys);
+ /* Set Tx descriptors queue starting address indirect
+ * access
+ */
+ if (priv->ip_version == MVPP21)
+ txq_phys = aggr_txq->descs_phys;
+ else
+ txq_phys = aggr_txq->descs_phys >>
+ MVPP22_AGGR_TXQ_DESC_ADDR_OFFS;
+
+ mvpp2_write(priv, MVPP2_AGGR_TXQ_DESC_ADDR_REG(cpu), txq_phys);
mvpp2_write(priv, MVPP2_AGGR_TXQ_DESC_SIZE_REG(cpu), desc_num);
return 0;
@@ -4796,6 +4806,8 @@ static int mvpp2_rxq_init(struct mvpp2_port *port,
struct mvpp2_rx_queue *rxq)
{
+ u32 rxq_phys;
+
rxq->size = port->rx_ring_size;
/* Allocate memory for RX descriptors */
@@ -4812,7 +4824,11 @@ static int mvpp2_rxq_init(struct mvpp2_port *port,
/* Set Rx descriptors queue starting address - indirect access */
mvpp2_write(port->priv, MVPP2_RXQ_NUM_REG, rxq->id);
- mvpp2_write(port->priv, MVPP2_RXQ_DESC_ADDR_REG, rxq->descs_phys);
+ if (port->priv->ip_version == MVPP21)
+ rxq_phys = rxq->descs_phys;
+ else
+ rxq_phys = rxq->descs_phys >> MVPP22_DESC_ADDR_OFFS;
+ mvpp2_write(port->priv, MVPP2_RXQ_DESC_ADDR_REG, rxq_phys);
mvpp2_write(port->priv, MVPP2_RXQ_DESC_SIZE_REG, rxq->size);
mvpp2_write(port->priv, MVPP2_RXQ_INDEX_REG, 0);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 22/27] net: mvpp2: handle misc PPv2.1/PPv2.2 differences
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit handles a few miscellaneous differences between PPv2.1 and
PPv2.2 in different areas, where code done for PPv2.1 doesn't apply for
PPv2.2 or needs to be adjusted (getting the MAC address, disabling PHY
polling, etc.).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 31 ++++++++++++++++++++++---------
1 file changed, 22 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 4163b0e..9b5a7c2 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -304,6 +304,9 @@
#define MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(v) (((v) << 6) & \
MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK)
+#define MVPP22_SMI_MISC_CFG_REG 0x2a204
+#define MVPP22_SMI_POLLING_EN BIT(10)
+
#define MVPP22_PORT_BASE 0x30e00
#define MVPP22_PORT_OFFSET 0x1000
@@ -5846,7 +5849,7 @@ static int mvpp2_check_ringparam_valid(struct net_device *dev,
return 0;
}
-static void mvpp2_get_mac_address(struct mvpp2_port *port, unsigned char *addr)
+static void mvpp21_get_mac_address(struct mvpp2_port *port, unsigned char *addr)
{
u32 mac_addr_l, mac_addr_m, mac_addr_h;
@@ -6294,7 +6297,7 @@ static const struct ethtool_ops mvpp2_eth_tool_ops = {
/* Driver initialization */
-static void mvpp2_port_power_up(struct mvpp2_port *port)
+static void mvpp21_port_power_up(struct mvpp2_port *port)
{
mvpp2_port_mii_set(port);
mvpp2_port_periodic_xon_disable(port);
@@ -6513,7 +6516,8 @@ static int mvpp2_port_probe(struct platform_device *pdev,
mac_from = "device tree";
ether_addr_copy(dev->dev_addr, dt_mac_addr);
} else {
- mvpp2_get_mac_address(port, hw_mac_addr);
+ if (priv->ip_version == MVPP21)
+ mvpp21_get_mac_address(port, hw_mac_addr);
if (is_valid_ether_addr(hw_mac_addr)) {
mac_from = "hardware";
ether_addr_copy(dev->dev_addr, hw_mac_addr);
@@ -6533,7 +6537,9 @@ static int mvpp2_port_probe(struct platform_device *pdev,
dev_err(&pdev->dev, "failed to init port %d\n", id);
goto err_free_stats;
}
- mvpp2_port_power_up(port);
+
+ if (priv->ip_version == MVPP21)
+ mvpp21_port_power_up(port);
port->pcpu = alloc_percpu(struct mvpp2_port_pcpu);
if (!port->pcpu) {
@@ -6671,9 +6677,15 @@ static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
mvpp2_conf_mbus_windows(dram_target_info, priv);
/* Disable HW PHY polling */
- val = readl(priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
- val |= MVPP2_PHY_AN_STOP_SMI0_MASK;
- writel(val, priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
+ if (priv->ip_version == MVPP21) {
+ val = readl(priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
+ val |= MVPP2_PHY_AN_STOP_SMI0_MASK;
+ writel(val, priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
+ } else {
+ val = readl(priv->iface_base + MVPP22_SMI_MISC_CFG_REG);
+ val &= ~MVPP22_SMI_POLLING_EN;
+ writel(val, priv->iface_base + MVPP22_SMI_MISC_CFG_REG);
+ }
/* Allocate and initialize aggregated TXQs */
priv->aggr_txqs = devm_kcalloc(&pdev->dev, num_present_cpus(),
@@ -6698,8 +6710,9 @@ static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
for (i = 0; i < MVPP2_MAX_PORTS; i++)
mvpp2_write(priv, MVPP2_ISR_RXQ_GROUP_REG(i), rxq_number);
- writel(MVPP2_EXT_GLOBAL_CTRL_DEFAULT,
- priv->lms_base + MVPP2_MNG_EXTENDED_GLOBAL_CTRL_REG);
+ if (priv->ip_version == MVPP21)
+ writel(MVPP2_EXT_GLOBAL_CTRL_DEFAULT,
+ priv->lms_base + MVPP2_MNG_EXTENDED_GLOBAL_CTRL_REG);
/* Allow cache snoop when transmiting packets */
mvpp2_write(priv, MVPP2_TX_SNOOP_REG, 0x1);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 18/27] net: mvpp2: adapt the mvpp2_rxq_*_pool_set functions to PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
The MVPP2_RXQ_CONFIG_REG register has a slightly different layout
between PPv2.1 and PPv2.2, so this commit adapts the functions modifying
this register to accommodate for both the PPv2.1 and PPv2.2 cases.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 32 ++++++++++++++++++++------------
1 file changed, 20 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 772c384..dd44065 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -50,9 +50,11 @@
#define MVPP2_SNOOP_PKT_SIZE_MASK 0x1ff
#define MVPP2_SNOOP_BUF_HDR_MASK BIT(9)
#define MVPP2_RXQ_POOL_SHORT_OFFS 20
-#define MVPP2_RXQ_POOL_SHORT_MASK 0x700000
+#define MVPP21_RXQ_POOL_SHORT_MASK 0x700000
+#define MVPP22_RXQ_POOL_SHORT_MASK 0xf00000
#define MVPP2_RXQ_POOL_LONG_OFFS 24
-#define MVPP2_RXQ_POOL_LONG_MASK 0x7000000
+#define MVPP21_RXQ_POOL_LONG_MASK 0x7000000
+#define MVPP22_RXQ_POOL_LONG_MASK 0xf000000
#define MVPP2_RXQ_PACKET_OFFSET_OFFS 28
#define MVPP2_RXQ_PACKET_OFFSET_MASK 0x70000000
#define MVPP2_RXQ_DISABLE_MASK BIT(31)
@@ -3762,17 +3764,20 @@ static int mvpp2_bm_init(struct platform_device *pdev, struct mvpp2 *priv)
static void mvpp2_rxq_long_pool_set(struct mvpp2_port *port,
int lrxq, int long_pool)
{
- u32 val;
+ u32 val, mask;
int prxq;
/* Get queue physical ID */
prxq = port->rxqs[lrxq]->id;
- val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
- val &= ~MVPP2_RXQ_POOL_LONG_MASK;
- val |= ((long_pool << MVPP2_RXQ_POOL_LONG_OFFS) &
- MVPP2_RXQ_POOL_LONG_MASK);
+ if (port->priv->ip_version == MVPP21)
+ mask = MVPP21_RXQ_POOL_LONG_MASK;
+ else
+ mask = MVPP22_RXQ_POOL_LONG_MASK;
+ val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
+ val &= ~mask;
+ val |= (long_pool << MVPP2_RXQ_POOL_LONG_OFFS) & mask;
mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}
@@ -3780,17 +3785,20 @@ static void mvpp2_rxq_long_pool_set(struct mvpp2_port *port,
static void mvpp2_rxq_short_pool_set(struct mvpp2_port *port,
int lrxq, int short_pool)
{
- u32 val;
+ u32 val, mask;
int prxq;
/* Get queue physical ID */
prxq = port->rxqs[lrxq]->id;
- val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
- val &= ~MVPP2_RXQ_POOL_SHORT_MASK;
- val |= ((short_pool << MVPP2_RXQ_POOL_SHORT_OFFS) &
- MVPP2_RXQ_POOL_SHORT_MASK);
+ if (port->priv->ip_version == MVPP21)
+ mask = MVPP21_RXQ_POOL_SHORT_MASK;
+ else
+ mask = MVPP22_RXQ_POOL_SHORT_MASK;
+ val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
+ val &= ~mask;
+ val |= (short_pool << MVPP2_RXQ_POOL_SHORT_OFFS) & mask;
mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 19/27] net: mvpp2: adapt mvpp2_defaults_set() to PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit modifies the mvpp2_defaults_set() function to not do the
loopback and FIFO threshold initialization, which are not needed for
PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 22 ++++++++++++----------
1 file changed, 12 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index dd44065..1394ed9 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -4205,16 +4205,18 @@ static void mvpp2_defaults_set(struct mvpp2_port *port)
{
int tx_port_num, val, queue, ptxq, lrxq;
- /* Configure port to loopback if needed */
- if (port->flags & MVPP2_F_LOOPBACK)
- mvpp2_port_loopback_set(port);
-
- /* Update TX FIFO MIN Threshold */
- val = readl(port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
- val &= ~MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK;
- /* Min. TX threshold must be less than minimal packet length */
- val |= MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(64 - 4 - 2);
- writel(val, port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
+ if (port->priv->ip_version == MVPP21) {
+ /* Configure port to loopback if needed */
+ if (port->flags & MVPP2_F_LOOPBACK)
+ mvpp2_port_loopback_set(port);
+
+ /* Update TX FIFO MIN Threshold */
+ val = readl(port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
+ val &= ~MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK;
+ /* Min. TX threshold must be less than minimal packet length */
+ val |= MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(64 - 4 - 2);
+ writel(val, port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
+ }
/* Disable Legacy WRR, Disable EJP, Release from reset */
tx_port_num = mvpp2_egress_port(port);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 16/27] net: mvpp2: introduce PPv2.2 HW descriptors and adapt accessors
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit adds the definition of the PPv2.2 HW descriptors, adjusts
the mvpp2_tx_desc and mvpp2_rx_desc structures accordingly, and adapts
the accessors to work on both PPv2.1 and PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 109 +++++++++++++++++++++++++++++++----
1 file changed, 98 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 1086b36..41d7bed 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -773,18 +773,42 @@ struct mvpp21_rx_desc {
u32 reserved8;
};
+/* HW TX descriptor for PPv2.2 */
+struct mvpp22_tx_desc {
+ u32 command;
+ u8 packet_offset;
+ u8 phys_txq;
+ u16 data_size;
+ u64 reserved1;
+ u64 buf_phys_addr_ptp;
+ u64 buf_cookie_misc;
+};
+
+/* HW RX descriptor for PPv2.2 */
+struct mvpp22_rx_desc {
+ u32 status;
+ u16 reserved1;
+ u16 data_size;
+ u32 reserved2;
+ u32 reserved3;
+ u64 buf_phys_addr_key_hash;
+ u64 buf_cookie_misc;
+};
+
/* Opaque type used by the driver to manipulate the HW TX and RX
* descriptors
*/
struct mvpp2_tx_desc {
union {
struct mvpp21_tx_desc pp21;
+ struct mvpp22_tx_desc pp22;
};
};
struct mvpp2_rx_desc {
union {
struct mvpp21_rx_desc pp21;
+ struct mvpp22_rx_desc pp22;
};
};
@@ -991,72 +1015,135 @@ static u32 mvpp2_read(struct mvpp2 *priv, u32 offset)
static dma_addr_t mvpp2_txdesc_phys_addr_get(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc)
{
- return tx_desc->pp21.buf_phys_addr;
+ if (port->priv->ip_version == MVPP21)
+ return tx_desc->pp21.buf_phys_addr;
+ else
+ return tx_desc->pp22.buf_phys_addr_ptp & DMA_BIT_MASK(40);
}
static void mvpp2_txdesc_phys_addr_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
dma_addr_t phys_addr)
{
- tx_desc->pp21.buf_phys_addr = phys_addr;
+ if (port->priv->ip_version == MVPP21) {
+ tx_desc->pp21.buf_phys_addr = phys_addr;
+ } else {
+ u64 val = (u64)phys_addr;
+
+ tx_desc->pp22.buf_phys_addr_ptp &= ~DMA_BIT_MASK(40);
+ tx_desc->pp22.buf_phys_addr_ptp |= val;
+ }
}
static size_t mvpp2_txdesc_size_get(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc)
{
- return tx_desc->pp21.data_size;
+ if (port->priv->ip_version == MVPP21)
+ return tx_desc->pp21.data_size;
+ else
+ return tx_desc->pp22.data_size;
}
static void mvpp2_txdesc_size_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
size_t size)
{
- tx_desc->pp21.data_size = size;
+ if (port->priv->ip_version == MVPP21)
+ tx_desc->pp21.data_size = size;
+ else
+ tx_desc->pp22.data_size = size;
}
static void mvpp2_txdesc_txq_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int txq)
{
- tx_desc->pp21.phys_txq = txq;
+ if (port->priv->ip_version == MVPP21)
+ tx_desc->pp21.phys_txq = txq;
+ else
+ tx_desc->pp22.phys_txq = txq;
}
static void mvpp2_txdesc_cmd_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int command)
{
- tx_desc->pp21.command = command;
+ if (port->priv->ip_version == MVPP21)
+ tx_desc->pp21.command = command;
+ else
+ tx_desc->pp22.command = command;
}
static void mvpp2_txdesc_offset_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int offset)
{
- tx_desc->pp21.packet_offset = offset;
+ if (port->priv->ip_version == MVPP21)
+ tx_desc->pp21.packet_offset = offset;
+ else
+ tx_desc->pp22.packet_offset = offset;
}
static dma_addr_t mvpp2_rxdesc_phys_addr_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->pp21.buf_phys_addr;
+ if (port->priv->ip_version == MVPP21)
+ return rx_desc->pp21.buf_phys_addr;
+ else
+ return rx_desc->pp22.buf_phys_addr_key_hash & DMA_BIT_MASK(40);
}
static unsigned long mvpp2_rxdesc_virt_addr_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->pp21.buf_cookie;
+ /* PPv2.1 can only be used on 32 bits architectures, and there
+ * are 32 bits in buf_cookie which are enough to store the
+ * full virtual address, so things are easy.
+ */
+ if (port->priv->ip_version == MVPP21) {
+ return rx_desc->pp21.buf_cookie;
+ } else {
+ /* On PPv2.2, the situation is more complicated,
+ * because there is only 40 bits to store the virtual
+ * address, which is not sufficient. So on 64 bits
+ * systems, we use phys_to_virt() to get the virtual
+ * address from the physical address, which is fine
+ * because the kernel linear mapping includes the
+ * entire 40 bits physical address space. On 32 bits
+ * systems however, we can't use phys_to_virt(), but
+ * since virtual addresses are 32 bits only, there is
+ * enough space in the RX descriptor for the full
+ * virtual address.
+ */
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+ dma_addr_t dma_addr =
+ rx_desc->pp22.buf_phys_addr_key_hash & DMA_BIT_MASK(40);
+ phys_addr_t phys_addr =
+ dma_to_phys(port->dev->dev.parent, dma_addr);
+
+ return (unsigned long)phys_to_virt(phys_addr);
+#else
+ return rx_desc->pp22.buf_cookie_misc & DMA_BIT_MASK(40);
+#endif
+ }
}
static size_t mvpp2_rxdesc_size_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->pp21.data_size;
+ if (port->priv->ip_version == MVPP21)
+ return rx_desc->pp21.data_size;
+ else
+ return rx_desc->pp22.data_size;
}
static u32 mvpp2_rxdesc_status_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->pp21.status;
+ if (port->priv->ip_version == MVPP21)
+ return rx_desc->pp21.status;
+ else
+ return rx_desc->pp22.status;
}
static void mvpp2_txq_inc_get(struct mvpp2_txq_pcpu *txq_pcpu)
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 17/27] net: mvpp2: adjust the allocation/free of BM pools for PPv2.2
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit adjusts the allocation and freeing of BM pools to support
PPv2.2. This involves:
- Checking that the number of buffer pointers is a multiple of 16, as
required by the hardware.
- Adjusting the size of the DMA coherent area allocated for buffer
pointers. Indeed, PPv2.2 needs space for 2 pointers of 64-bits per
buffer, as opposed to 2 pointers of 32-bits per buffer in
PPv2.1. The size in bytes is now stored in a new field of the
mvpp2_bm_pool structure.
- On PPv2.2, the 32 high order bits of the BM pointer area physical
address must be programmed in the MVPP2_BM_HIGH_BASE_REG register.
- On PPv2.2, getting the physical and virtual address of each buffer
requires reading the MVPP2_BM_ADDR_HIGH_ALLOC to get the high order
bits of those addresses. A new utility function
mvpp2_bm_bufs_get_addrs() is introduced to handle this.
- On PPv2.2, releasing a buffer requires writing the high order 32 bits
of the physical address to MVPP2_BM_PHY_VIRT_HIGH_RLS_REG. We no
longer need to write the virtual address to MVPP2_BM_VIRT_RLS_REG.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 97 ++++++++++++++++++++++++++++++------
1 file changed, 82 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 41d7bed..772c384 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -208,17 +208,28 @@
#define MVPP2_BM_BPPE_FULL_MASK BIT(3)
#define MVPP2_BM_AVAILABLE_BP_LOW_MASK BIT(4)
#define MVPP2_BM_INTR_MASK_REG(pool) (0x6280 + ((pool) * 4))
+#define MVPP2_BM_HIGH_BASE_REG 0x6310
+#define MVPP2_BM_HIGH_BASE_MASK 0xff
#define MVPP2_BM_PHY_ALLOC_REG(pool) (0x6400 + ((pool) * 4))
#define MVPP2_BM_PHY_ALLOC_GRNTD_MASK BIT(0)
#define MVPP2_BM_VIRT_ALLOC_REG 0x6440
+#define MVPP2_BM_ADDR_HIGH_ALLOC 0x6444
+#define MVPP2_BM_ADDR_HIGH_PHYS_MASK 0xff
+#define MVPP2_BM_ADDR_HIGH_VIRT_MASK 0xff00
+#define MVPP2_BM_ADDR_HIGH_VIRT_SHIFT 8
#define MVPP2_BM_PHY_RLS_REG(pool) (0x6480 + ((pool) * 4))
#define MVPP2_BM_PHY_RLS_MC_BUFF_MASK BIT(0)
#define MVPP2_BM_PHY_RLS_PRIO_EN_MASK BIT(1)
#define MVPP2_BM_PHY_RLS_GRNTD_MASK BIT(2)
#define MVPP2_BM_VIRT_RLS_REG 0x64c0
-#define MVPP2_BM_MC_RLS_REG 0x64c4
+#define MVPP21_BM_MC_RLS_REG 0x64c4
#define MVPP2_BM_MC_ID_MASK 0xfff
#define MVPP2_BM_FORCE_RELEASE_MASK BIT(12)
+#define MVPP22_BM_ADDR_HIGH_RLS_REG 0x64c4
+#define MVPP22_BM_ADDR_HIGH_PHYS_RLS_MASK 0xff
+#define MVPP22_BM_ADDR_HIGH_VIRT_RLS_MASK 0xff00
+#define MVPP22_BM_ADDR_HIGH_VIRT_RLS_SHIFT 8
+#define MVPP22_BM_MC_RLS_REG 0x64d4
/* TX Scheduler registers */
#define MVPP2_TXP_SCHED_PORT_INDEX_REG 0x8000
@@ -957,6 +968,8 @@ struct mvpp2_bm_pool {
/* Buffer Pointers Pool External (BPPE) size */
int size;
+ /* BPPE size in bytes */
+ int size_bytes;
/* Number of buffers for this pool */
int buf_num;
/* Pool buffer size */
@@ -3558,11 +3571,23 @@ static int mvpp2_bm_pool_create(struct platform_device *pdev,
struct mvpp2 *priv,
struct mvpp2_bm_pool *bm_pool, int size)
{
- int size_bytes;
u32 val;
- size_bytes = sizeof(u32) * size;
- bm_pool->virt_addr = dma_alloc_coherent(&pdev->dev, size_bytes,
+ /* Number of buffer pointers must be a multiple of 16, as per
+ * hardware constraints
+ */
+ if (!IS_ALIGNED(size, 16))
+ return -EINVAL;
+
+ /* PPv2.1 needs 8 bytes per buffer pointer, PPv2.2 needs 16
+ * bytes per buffer pointer
+ */
+ if (priv->ip_version == MVPP21)
+ bm_pool->size_bytes = 2 * sizeof(u32) * size;
+ else
+ bm_pool->size_bytes = 2 * sizeof(u64) * size;
+
+ bm_pool->virt_addr = dma_alloc_coherent(&pdev->dev, bm_pool->size_bytes,
&bm_pool->phys_addr,
GFP_KERNEL);
if (!bm_pool->virt_addr)
@@ -3570,15 +3595,24 @@ static int mvpp2_bm_pool_create(struct platform_device *pdev,
if (!IS_ALIGNED((unsigned long)bm_pool->virt_addr,
MVPP2_BM_POOL_PTR_ALIGN)) {
- dma_free_coherent(&pdev->dev, size_bytes, bm_pool->virt_addr,
- bm_pool->phys_addr);
+ dma_free_coherent(&pdev->dev, bm_pool->size_bytes,
+ bm_pool->virt_addr, bm_pool->phys_addr);
dev_err(&pdev->dev, "BM pool %d is not %d bytes aligned\n",
bm_pool->id, MVPP2_BM_POOL_PTR_ALIGN);
return -ENOMEM;
}
mvpp2_write(priv, MVPP2_BM_POOL_BASE_REG(bm_pool->id),
- bm_pool->phys_addr);
+ lower_32_bits(bm_pool->phys_addr));
+ /* On PPv2.2, program the high order bits of the base address */
+ if (priv->ip_version == MVPP22) {
+ if (sizeof(dma_addr_t) == 8)
+ val = upper_32_bits(bm_pool->phys_addr) &
+ MVPP2_BM_HIGH_BASE_MASK;
+ else
+ val = 0;
+ mvpp2_write(priv, MVPP2_BM_HIGH_BASE_REG, val);
+ }
mvpp2_write(priv, MVPP2_BM_POOL_SIZE_REG(bm_pool->id), size);
val = mvpp2_read(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id));
@@ -3606,6 +3640,27 @@ static void mvpp2_bm_pool_bufsize_set(struct mvpp2 *priv,
mvpp2_write(priv, MVPP2_POOL_BUF_SIZE_REG(bm_pool->id), val);
}
+static void mvpp2_bm_bufs_get_addrs(struct device *dev, struct mvpp2 *priv,
+ struct mvpp2_bm_pool *bm_pool,
+ dma_addr_t *paddr, unsigned long *vaddr)
+{
+ *paddr = mvpp2_read(priv, MVPP2_BM_PHY_ALLOC_REG(bm_pool->id));
+ *vaddr = mvpp2_read(priv, MVPP2_BM_VIRT_ALLOC_REG);
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+ if (priv->ip_version == MVPP22) {
+ u32 val;
+ u32 paddr_highbits;
+
+ val = mvpp2_read(priv, MVPP2_BM_ADDR_HIGH_ALLOC);
+ paddr_highbits = (val & MVPP2_BM_ADDR_HIGH_PHYS_MASK);
+
+ *paddr |= (dma_addr_t)paddr_highbits << 32;
+ *vaddr = (unsigned long)phys_to_virt(dma_to_phys(dev, *paddr));
+ }
+#endif
+}
+
/* Free all buffers from the pool */
static void mvpp2_bm_bufs_free(struct device *dev, struct mvpp2 *priv,
struct mvpp2_bm_pool *bm_pool)
@@ -3616,10 +3671,8 @@ static void mvpp2_bm_bufs_free(struct device *dev, struct mvpp2 *priv,
dma_addr_t buf_phys_addr;
unsigned long vaddr;
- /* Get buffer virtual address (indirect access) */
- buf_phys_addr = mvpp2_read(priv,
- MVPP2_BM_PHY_ALLOC_REG(bm_pool->id));
- vaddr = mvpp2_read(priv, MVPP2_BM_VIRT_ALLOC_REG);
+ mvpp2_bm_bufs_get_addrs(dev, priv, bm_pool,
+ &buf_phys_addr, &vaddr);
dma_unmap_single(dev, buf_phys_addr,
bm_pool->buf_size, DMA_FROM_DEVICE);
@@ -3651,7 +3704,7 @@ static int mvpp2_bm_pool_destroy(struct platform_device *pdev,
val |= MVPP2_BM_STOP_MASK;
mvpp2_write(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id), val);
- dma_free_coherent(&pdev->dev, sizeof(u32) * bm_pool->size,
+ dma_free_coherent(&pdev->dev, bm_pool->size_bytes,
bm_pool->virt_addr,
bm_pool->phys_addr);
return 0;
@@ -3787,8 +3840,19 @@ static inline void mvpp2_bm_pool_put(struct mvpp2_port *port, int pool,
dma_addr_t buf_phys_addr,
unsigned long buf_virt_addr)
{
- mvpp2_write(port->priv, MVPP2_BM_VIRT_RLS_REG, buf_virt_addr);
- mvpp2_write(port->priv, MVPP2_BM_PHY_RLS_REG(pool), buf_phys_addr);
+#if defined(CONFIG_ARCH_DMA_ADDR_T_64BIT)
+ u32 val;
+
+ val = upper_32_bits(buf_phys_addr) & MVPP22_BM_ADDR_HIGH_PHYS_RLS_MASK;
+ val |= (upper_32_bits(buf_virt_addr) &
+ MVPP22_BM_ADDR_HIGH_VIRT_RLS_MASK)
+ << MVPP22_BM_ADDR_HIGH_VIRT_RLS_SHIFT;
+ mvpp2_write(port->priv, MVPP22_BM_ADDR_HIGH_RLS_REG, val);
+#endif
+ mvpp2_write(port->priv, MVPP2_BM_VIRT_RLS_REG,
+ lower_32_bits(buf_virt_addr));
+ mvpp2_write(port->priv, MVPP2_BM_PHY_RLS_REG(pool),
+ lower_32_bits(buf_phys_addr));
}
/* Release multicast buffer */
@@ -3800,7 +3864,10 @@ static void mvpp2_bm_pool_mc_put(struct mvpp2_port *port, int pool,
u32 val = 0;
val |= (mc_id & MVPP2_BM_MC_ID_MASK);
- mvpp2_write(port->priv, MVPP2_BM_MC_RLS_REG, val);
+ if (port->priv->ip_version == MVPP21)
+ mvpp2_write(port->priv, MVPP21_BM_MC_RLS_REG, val);
+ else
+ mvpp2_write(port->priv, MVPP22_BM_MC_RLS_REG, val);
mvpp2_bm_pool_put(port, pool,
buf_phys_addr | MVPP2_BM_PHY_RLS_MC_BUFF_MASK,
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 13/27] net: mvpp2: add and use accessors for TX/RX descriptors
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
The PPv2.2 IP has a different TX and RX descriptor layout compared to
PPv2.1. In order to prepare for the introduction of PPv2.2 support in
mvpp2, this commit adds accessors for the different fields of the TX
and RX descriptors, and changes the code to use them.
For now, the mvpp2_port argument passed to the accessors is not used,
but it will be used in follow-up to update the descriptor according to
the version of the IP being used.
Apart from the mechanical changes to use the newly introduced
accessors, a few other changes, needed to use the accessors, are made:
- The mvpp2_txq_inc_put() function now takes a mvpp2_port as first
argument, as it is needed to use the accessors.
- Similarly, the mvpp2_bm_cookie_build() gains a mvpp2_port first
argument, for the same reason.
- In mvpp2_rx_error(), instead of accessing the RX descriptor in each
case of the switch, we introduce a local variable to store the
packet size.
- Similarly, in mvpp2_buff_hdr_rx(), we introduce a local "cookie"
variable to store the RX descriptor cookie, rather than accessing
it from the descriptor each time.
- In mvpp2_tx_frag_process() and mvpp2_tx() instead of accessing the
packet size from the TX descriptor, we use the actual value
available in the function, which is used to set the TX descriptor
packet size a few lines before.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 187 +++++++++++++++++++++++++----------
1 file changed, 137 insertions(+), 50 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 55ebfb3..e35871f 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -967,6 +967,77 @@ static u32 mvpp2_read(struct mvpp2 *priv, u32 offset)
return readl(priv->base + offset);
}
+static dma_addr_t mvpp2_txdesc_phys_addr_get(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc)
+{
+ return tx_desc->buf_phys_addr;
+}
+
+static void mvpp2_txdesc_phys_addr_set(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc,
+ dma_addr_t phys_addr)
+{
+ tx_desc->buf_phys_addr = phys_addr;
+}
+
+static size_t mvpp2_txdesc_size_get(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc)
+{
+ return tx_desc->data_size;
+}
+
+static void mvpp2_txdesc_size_set(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc,
+ size_t size)
+{
+ tx_desc->data_size = size;
+}
+
+static void mvpp2_txdesc_txq_set(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc,
+ unsigned int txq)
+{
+ tx_desc->phys_txq = txq;
+}
+
+static void mvpp2_txdesc_cmd_set(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc,
+ unsigned int command)
+{
+ tx_desc->command = command;
+}
+
+static void mvpp2_txdesc_offset_set(struct mvpp2_port *port,
+ struct mvpp2_tx_desc *tx_desc,
+ unsigned int offset)
+{
+ tx_desc->packet_offset = offset;
+}
+
+static dma_addr_t mvpp2_rxdesc_phys_addr_get(struct mvpp2_port *port,
+ struct mvpp2_rx_desc *rx_desc)
+{
+ return rx_desc->buf_phys_addr;
+}
+
+static unsigned long mvpp2_rxdesc_virt_addr_get(struct mvpp2_port *port,
+ struct mvpp2_rx_desc *rx_desc)
+{
+ return rx_desc->buf_cookie;
+}
+
+static size_t mvpp2_rxdesc_size_get(struct mvpp2_port *port,
+ struct mvpp2_rx_desc *rx_desc)
+{
+ return rx_desc->data_size;
+}
+
+static u32 mvpp2_rxdesc_status_get(struct mvpp2_port *port,
+ struct mvpp2_rx_desc *rx_desc)
+{
+ return rx_desc->status;
+}
+
static void mvpp2_txq_inc_get(struct mvpp2_txq_pcpu *txq_pcpu)
{
txq_pcpu->txq_get_index++;
@@ -974,15 +1045,16 @@ static void mvpp2_txq_inc_get(struct mvpp2_txq_pcpu *txq_pcpu)
txq_pcpu->txq_get_index = 0;
}
-static void mvpp2_txq_inc_put(struct mvpp2_txq_pcpu *txq_pcpu,
+static void mvpp2_txq_inc_put(struct mvpp2_port *port,
+ struct mvpp2_txq_pcpu *txq_pcpu,
struct sk_buff *skb,
struct mvpp2_tx_desc *tx_desc)
{
struct mvpp2_txq_pcpu_buf *tx_buf =
txq_pcpu->buffs + txq_pcpu->txq_put_index;
tx_buf->skb = skb;
- tx_buf->size = tx_desc->data_size;
- tx_buf->phys = tx_desc->buf_phys_addr;
+ tx_buf->size = mvpp2_txdesc_size_get(port, tx_desc);
+ tx_buf->phys = mvpp2_txdesc_phys_addr_get(port, tx_desc);
txq_pcpu->txq_put_index++;
if (txq_pcpu->txq_put_index == txq_pcpu->size)
txq_pcpu->txq_put_index = 0;
@@ -4147,11 +4219,15 @@ static void mvpp2_rxq_offset_set(struct mvpp2_port *port,
}
/* Obtain BM cookie information from descriptor */
-static u32 mvpp2_bm_cookie_build(struct mvpp2_rx_desc *rx_desc)
+static u32 mvpp2_bm_cookie_build(struct mvpp2_port *port,
+ struct mvpp2_rx_desc *rx_desc)
{
- int pool = (rx_desc->status & MVPP2_RXD_BM_POOL_ID_MASK) >>
- MVPP2_RXD_BM_POOL_ID_OFFS;
int cpu = smp_processor_id();
+ int pool;
+
+ pool = (mvpp2_rxdesc_status_get(port, rx_desc) &
+ MVPP2_RXD_BM_POOL_ID_MASK) >>
+ MVPP2_RXD_BM_POOL_ID_OFFS;
return ((pool & 0xFF) << MVPP2_BM_COOKIE_POOL_OFFS) |
((cpu & 0xFF) << MVPP2_BM_COOKIE_CPU_OFFS);
@@ -4580,10 +4656,11 @@ static void mvpp2_rxq_drop_pkts(struct mvpp2_port *port,
for (i = 0; i < rx_received; i++) {
struct mvpp2_rx_desc *rx_desc = mvpp2_rxq_next_desc_get(rxq);
- u32 bm = mvpp2_bm_cookie_build(rx_desc);
+ u32 bm = mvpp2_bm_cookie_build(port, rx_desc);
- mvpp2_pool_refill(port, bm, rx_desc->buf_phys_addr,
- rx_desc->buf_cookie);
+ mvpp2_pool_refill(port, bm,
+ mvpp2_rxdesc_phys_addr_get(port, rx_desc),
+ mvpp2_rxdesc_virt_addr_get(port, rx_desc));
}
mvpp2_rxq_status_update(port, rxq->id, rx_received, rx_received);
}
@@ -4972,20 +5049,21 @@ static enum hrtimer_restart mvpp2_hr_timer_cb(struct hrtimer *timer)
static void mvpp2_rx_error(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- u32 status = rx_desc->status;
+ u32 status = mvpp2_rxdesc_status_get(port, rx_desc);
+ size_t sz = mvpp2_rxdesc_size_get(port, rx_desc);
switch (status & MVPP2_RXD_ERR_CODE_MASK) {
case MVPP2_RXD_ERR_CRC:
- netdev_err(port->dev, "bad rx status %08x (crc error), size=%d\n",
- status, rx_desc->data_size);
+ netdev_err(port->dev, "bad rx status %08x (crc error), size=%zu\n",
+ status, sz);
break;
case MVPP2_RXD_ERR_OVERRUN:
- netdev_err(port->dev, "bad rx status %08x (overrun error), size=%d\n",
- status, rx_desc->data_size);
+ netdev_err(port->dev, "bad rx status %08x (overrun error), size=%zu\n",
+ status, sz);
break;
case MVPP2_RXD_ERR_RESOURCE:
- netdev_err(port->dev, "bad rx status %08x (resource error), size=%d\n",
- status, rx_desc->data_size);
+ netdev_err(port->dev, "bad rx status %08x (resource error), size=%zu\n",
+ status, sz);
break;
}
}
@@ -5061,7 +5139,7 @@ static void mvpp2_buff_hdr_rx(struct mvpp2_port *port,
{
struct mvpp2_buff_hdr *buff_hdr;
struct sk_buff *skb;
- u32 rx_status = rx_desc->status;
+ u32 rx_status = mvpp2_rxdesc_status_get(port, rx_desc);
dma_addr_t buff_phys_addr;
unsigned long buff_virt_addr;
dma_addr_t buff_phys_addr_next;
@@ -5071,8 +5149,8 @@ static void mvpp2_buff_hdr_rx(struct mvpp2_port *port,
pool_id = (rx_status & MVPP2_RXD_BM_POOL_ID_MASK) >>
MVPP2_RXD_BM_POOL_ID_OFFS;
- buff_phys_addr = rx_desc->buf_phys_addr;
- buff_virt_addr = rx_desc->buf_cookie;
+ buff_phys_addr = mvpp2_rxdesc_phys_addr_get(port, rx_desc);
+ buff_virt_addr = mvpp2_rxdesc_virt_addr_get(port, rx_desc);
do {
skb = (struct sk_buff *)buff_virt_addr;
@@ -5119,12 +5197,13 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
void *data;
rx_done++;
- rx_status = rx_desc->status;
- rx_bytes = rx_desc->data_size - MVPP2_MH_SIZE;
- phys_addr = rx_desc->buf_phys_addr;
- data = (void *)rx_desc->buf_cookie;
+ rx_status = mvpp2_rxdesc_status_get(port, rx_desc);
+ rx_bytes = mvpp2_rxdesc_size_get(port, rx_desc);
+ rx_bytes -= MVPP2_MH_SIZE;
+ phys_addr = mvpp2_rxdesc_phys_addr_get(port, rx_desc);
+ data = (void *)mvpp2_rxdesc_virt_addr_get(port, rx_desc);
- bm = mvpp2_bm_cookie_build(rx_desc);
+ bm = mvpp2_bm_cookie_build(port, rx_desc);
pool = mvpp2_bm_cookie_pool_get(bm);
bm_pool = &port->priv->bm_pools[pool];
/* Check if buffer header is used */
@@ -5143,9 +5222,8 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
dev->stats.rx_errors++;
mvpp2_rx_error(port, rx_desc);
/* Return the buffer to the pool */
-
- mvpp2_pool_refill(port, bm, rx_desc->buf_phys_addr,
- rx_desc->buf_cookie);
+ mvpp2_pool_refill(port, bm, phys_addr,
+ (unsigned long)data);
continue;
}
@@ -5197,11 +5275,15 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
}
static inline void
-tx_desc_unmap_put(struct device *dev, struct mvpp2_tx_queue *txq,
+tx_desc_unmap_put(struct mvpp2_port *port, struct mvpp2_tx_queue *txq,
struct mvpp2_tx_desc *desc)
{
- dma_unmap_single(dev, desc->buf_phys_addr,
- desc->data_size, DMA_TO_DEVICE);
+ dma_addr_t buf_phys_addr =
+ mvpp2_txdesc_phys_addr_get(port, desc);
+ size_t buf_sz =
+ mvpp2_txdesc_size_get(port, desc);
+ dma_unmap_single(port->dev->dev.parent, buf_phys_addr,
+ buf_sz, DMA_TO_DEVICE);
mvpp2_txq_desc_put(txq);
}
@@ -5220,28 +5302,31 @@ static int mvpp2_tx_frag_process(struct mvpp2_port *port, struct sk_buff *skb,
void *addr = page_address(frag->page.p) + frag->page_offset;
tx_desc = mvpp2_txq_next_desc_get(aggr_txq);
- tx_desc->phys_txq = txq->id;
- tx_desc->data_size = frag->size;
+ mvpp2_txdesc_txq_set(port, tx_desc, txq->id);
+ mvpp2_txdesc_size_set(port, tx_desc, frag->size);
buf_phys_addr = dma_map_single(port->dev->dev.parent, addr,
- tx_desc->data_size,
+ frag->size,
DMA_TO_DEVICE);
if (dma_mapping_error(port->dev->dev.parent, buf_phys_addr)) {
mvpp2_txq_desc_put(txq);
goto error;
}
- tx_desc->packet_offset = buf_phys_addr & MVPP2_TX_DESC_ALIGN;
- tx_desc->buf_phys_addr = buf_phys_addr & (~MVPP2_TX_DESC_ALIGN);
+ mvpp2_txdesc_phys_addr_set(port, tx_desc,
+ buf_phys_addr & MVPP2_TX_DESC_ALIGN);
+ mvpp2_txdesc_offset_set(port, tx_desc,
+ buf_phys_addr & (~MVPP2_TX_DESC_ALIGN));
if (i == (skb_shinfo(skb)->nr_frags - 1)) {
/* Last descriptor */
- tx_desc->command = MVPP2_TXD_L_DESC;
- mvpp2_txq_inc_put(txq_pcpu, skb, tx_desc);
+ mvpp2_txdesc_cmd_set(port, tx_desc,
+ MVPP2_TXD_L_DESC);
+ mvpp2_txq_inc_put(port, txq_pcpu, skb, tx_desc);
} else {
/* Descriptor in the middle: Not First, Not Last */
- tx_desc->command = 0;
- mvpp2_txq_inc_put(txq_pcpu, NULL, tx_desc);
+ mvpp2_txdesc_cmd_set(port, tx_desc, 0);
+ mvpp2_txq_inc_put(port, txq_pcpu, NULL, tx_desc);
}
}
@@ -5253,7 +5338,7 @@ static int mvpp2_tx_frag_process(struct mvpp2_port *port, struct sk_buff *skb,
*/
for (i = i - 1; i >= 0; i--) {
tx_desc = txq->descs + i;
- tx_desc_unmap_put(port->dev->dev.parent, txq, tx_desc);
+ tx_desc_unmap_put(port, txq, tx_desc);
}
return -ENOMEM;
@@ -5288,35 +5373,37 @@ static int mvpp2_tx(struct sk_buff *skb, struct net_device *dev)
/* Get a descriptor for the first part of the packet */
tx_desc = mvpp2_txq_next_desc_get(aggr_txq);
- tx_desc->phys_txq = txq->id;
- tx_desc->data_size = skb_headlen(skb);
+ mvpp2_txdesc_txq_set(port, tx_desc, txq->id);
+ mvpp2_txdesc_size_set(port, tx_desc, skb_headlen(skb));
buf_phys_addr = dma_map_single(dev->dev.parent, skb->data,
- tx_desc->data_size, DMA_TO_DEVICE);
+ skb_headlen(skb), DMA_TO_DEVICE);
if (unlikely(dma_mapping_error(dev->dev.parent, buf_phys_addr))) {
mvpp2_txq_desc_put(txq);
frags = 0;
goto out;
}
- tx_desc->packet_offset = buf_phys_addr & MVPP2_TX_DESC_ALIGN;
- tx_desc->buf_phys_addr = buf_phys_addr & ~MVPP2_TX_DESC_ALIGN;
+ mvpp2_txdesc_offset_set(port, tx_desc,
+ buf_phys_addr & MVPP2_TX_DESC_ALIGN);
+ mvpp2_txdesc_phys_addr_set(port, tx_desc,
+ buf_phys_addr & ~MVPP2_TX_DESC_ALIGN);
tx_cmd = mvpp2_skb_tx_csum(port, skb);
if (frags == 1) {
/* First and Last descriptor */
tx_cmd |= MVPP2_TXD_F_DESC | MVPP2_TXD_L_DESC;
- tx_desc->command = tx_cmd;
- mvpp2_txq_inc_put(txq_pcpu, skb, tx_desc);
+ mvpp2_txdesc_cmd_set(port, tx_desc, tx_cmd);
+ mvpp2_txq_inc_put(port, txq_pcpu, skb, tx_desc);
} else {
/* First but not Last */
tx_cmd |= MVPP2_TXD_F_DESC | MVPP2_TXD_PADDING_DISABLE;
- tx_desc->command = tx_cmd;
- mvpp2_txq_inc_put(txq_pcpu, NULL, tx_desc);
+ mvpp2_txdesc_cmd_set(port, tx_desc, tx_cmd);
+ mvpp2_txq_inc_put(port, txq_pcpu, NULL, tx_desc);
/* Continue with other skb fragments */
if (mvpp2_tx_frag_process(port, skb, aggr_txq, txq)) {
- tx_desc_unmap_put(port->dev->dev.parent, txq, tx_desc);
+ tx_desc_unmap_put(port, txq, tx_desc);
frags = 0;
goto out;
}
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 12/27] net: mvpp2: enable building on 64-bit platforms
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
The mvpp2 is going to be extended to support the Marvell Armada 7K/8K
platform, which is ARM64. As a preparation to this work, this commit
enables building the mvpp2 driver on ARM64, by:
- Adjusting the Kconfig dependency
- Fixing the types used in the driver so that they are 32/64-bits
compliant. We use dma_addr_t for DMA addresses, and unsigned long
for virtual addresses.
It is worth mentioning that after this commit, the driver is for now
still only used on 32-bits platforms.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/Kconfig | 2 +-
drivers/net/ethernet/marvell/mvpp2.c | 29 +++++++++++++++++------------
2 files changed, 18 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index 2664827..9ea757a 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -77,7 +77,7 @@ config MVNETA_BM
config MVPP2
tristate "Marvell Armada 375 network interface support"
- depends on MACH_ARMADA_375
+ depends on ARCH_MVEBU
select MVMDIO
---help---
This driver supports the network interface units in the
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 43ddc65..55ebfb3 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -3388,7 +3388,8 @@ static int mvpp2_bm_pool_create(struct platform_device *pdev,
if (!bm_pool->virt_addr)
return -ENOMEM;
- if (!IS_ALIGNED((u32)bm_pool->virt_addr, MVPP2_BM_POOL_PTR_ALIGN)) {
+ if (!IS_ALIGNED((unsigned long)bm_pool->virt_addr,
+ MVPP2_BM_POOL_PTR_ALIGN)) {
dma_free_coherent(&pdev->dev, size_bytes, bm_pool->virt_addr,
bm_pool->phys_addr);
dev_err(&pdev->dev, "BM pool %d is not %d bytes aligned\n",
@@ -3433,7 +3434,7 @@ static void mvpp2_bm_bufs_free(struct device *dev, struct mvpp2 *priv,
for (i = 0; i < bm_pool->buf_num; i++) {
dma_addr_t buf_phys_addr;
- u32 vaddr;
+ unsigned long vaddr;
/* Get buffer virtual address (indirect access) */
buf_phys_addr = mvpp2_read(priv,
@@ -3596,14 +3597,15 @@ static inline u32 mvpp2_bm_cookie_pool_set(u32 cookie, int pool)
}
/* Get pool number from a BM cookie */
-static inline int mvpp2_bm_cookie_pool_get(u32 cookie)
+static inline int mvpp2_bm_cookie_pool_get(unsigned long cookie)
{
return (cookie >> MVPP2_BM_COOKIE_POOL_OFFS) & 0xFF;
}
/* Release buffer to BM */
static inline void mvpp2_bm_pool_put(struct mvpp2_port *port, int pool,
- u32 buf_phys_addr, u32 buf_virt_addr)
+ dma_addr_t buf_phys_addr,
+ unsigned long buf_virt_addr)
{
mvpp2_write(port->priv, MVPP2_BM_VIRT_RLS_REG, buf_virt_addr);
mvpp2_write(port->priv, MVPP2_BM_PHY_RLS_REG(pool), buf_phys_addr);
@@ -3611,7 +3613,8 @@ static inline void mvpp2_bm_pool_put(struct mvpp2_port *port, int pool,
/* Release multicast buffer */
static void mvpp2_bm_pool_mc_put(struct mvpp2_port *port, int pool,
- u32 buf_phys_addr, u32 buf_virt_addr,
+ dma_addr_t buf_phys_addr,
+ unsigned long buf_virt_addr,
int mc_id)
{
u32 val = 0;
@@ -3626,7 +3629,8 @@ static void mvpp2_bm_pool_mc_put(struct mvpp2_port *port, int pool,
/* Refill BM pool */
static void mvpp2_pool_refill(struct mvpp2_port *port, u32 bm,
- u32 phys_addr, u32 cookie)
+ dma_addr_t phys_addr,
+ unsigned long cookie)
{
int pool = mvpp2_bm_cookie_pool_get(bm);
@@ -3657,7 +3661,8 @@ static int mvpp2_bm_bufs_add(struct mvpp2_port *port,
if (!buf)
break;
- mvpp2_bm_pool_put(port, bm_pool->id, (u32)phys_addr, (u32)buf);
+ mvpp2_bm_pool_put(port, bm_pool->id, phys_addr,
+ (unsigned long)buf);
}
/* Update BM driver with number of buffers added to pool */
@@ -5015,7 +5020,7 @@ static int mvpp2_rx_refill(struct mvpp2_port *port,
if (!buf)
return -ENOMEM;
- mvpp2_pool_refill(port, bm, (u32)phys_addr, (u32)buf);
+ mvpp2_pool_refill(port, bm, phys_addr, (unsigned long)buf);
return 0;
}
@@ -5057,10 +5062,10 @@ static void mvpp2_buff_hdr_rx(struct mvpp2_port *port,
struct mvpp2_buff_hdr *buff_hdr;
struct sk_buff *skb;
u32 rx_status = rx_desc->status;
- u32 buff_phys_addr;
- u32 buff_virt_addr;
- u32 buff_phys_addr_next;
- u32 buff_virt_addr_next;
+ dma_addr_t buff_phys_addr;
+ unsigned long buff_virt_addr;
+ dma_addr_t buff_phys_addr_next;
+ unsigned long buff_virt_addr_next;
int mc_id;
int pool_id;
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 15/27] net: mvpp2: introduce an intermediate union for the TX/RX descriptors
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
Since the format of the HW descriptors is different between PPv2.1 and
PPv2.2, this commit introduces an intermediate union, with for now
only the PPv2.1 descriptors. The bulk of the driver code only
manipulates opaque mvpp2_tx_desc and mvpp2_rx_desc pointers, and the
descriptors can only be accessed and modified through the accessor
functions. A follow-up commit will add the descriptor definitions for
PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 43 +++++++++++++++++++++++++-----------
1 file changed, 30 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index af00405..1086b36 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -745,7 +745,8 @@ struct mvpp2_port {
#define MVPP2_RXD_L3_IP6 BIT(30)
#define MVPP2_RXD_BUF_HDR BIT(31)
-struct mvpp2_tx_desc {
+/* HW TX descriptor for PPv2.1 */
+struct mvpp21_tx_desc {
u32 command; /* Options used by HW for packet transmitting.*/
u8 packet_offset; /* the offset from the buffer beginning */
u8 phys_txq; /* destination queue ID */
@@ -756,7 +757,8 @@ struct mvpp2_tx_desc {
u32 reserved2; /* reserved (for future use) */
};
-struct mvpp2_rx_desc {
+/* HW RX descriptor for PPv2.1 */
+struct mvpp21_rx_desc {
u32 status; /* info about received packet */
u16 reserved1; /* parser_info (for future use, PnC) */
u16 data_size; /* size of received packet in bytes */
@@ -771,6 +773,21 @@ struct mvpp2_rx_desc {
u32 reserved8;
};
+/* Opaque type used by the driver to manipulate the HW TX and RX
+ * descriptors
+ */
+struct mvpp2_tx_desc {
+ union {
+ struct mvpp21_tx_desc pp21;
+ };
+};
+
+struct mvpp2_rx_desc {
+ union {
+ struct mvpp21_rx_desc pp21;
+ };
+};
+
struct mvpp2_txq_pcpu_buf {
/* Transmitted SKB */
struct sk_buff *skb;
@@ -974,72 +991,72 @@ static u32 mvpp2_read(struct mvpp2 *priv, u32 offset)
static dma_addr_t mvpp2_txdesc_phys_addr_get(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc)
{
- return tx_desc->buf_phys_addr;
+ return tx_desc->pp21.buf_phys_addr;
}
static void mvpp2_txdesc_phys_addr_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
dma_addr_t phys_addr)
{
- tx_desc->buf_phys_addr = phys_addr;
+ tx_desc->pp21.buf_phys_addr = phys_addr;
}
static size_t mvpp2_txdesc_size_get(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc)
{
- return tx_desc->data_size;
+ return tx_desc->pp21.data_size;
}
static void mvpp2_txdesc_size_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
size_t size)
{
- tx_desc->data_size = size;
+ tx_desc->pp21.data_size = size;
}
static void mvpp2_txdesc_txq_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int txq)
{
- tx_desc->phys_txq = txq;
+ tx_desc->pp21.phys_txq = txq;
}
static void mvpp2_txdesc_cmd_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int command)
{
- tx_desc->command = command;
+ tx_desc->pp21.command = command;
}
static void mvpp2_txdesc_offset_set(struct mvpp2_port *port,
struct mvpp2_tx_desc *tx_desc,
unsigned int offset)
{
- tx_desc->packet_offset = offset;
+ tx_desc->pp21.packet_offset = offset;
}
static dma_addr_t mvpp2_rxdesc_phys_addr_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->buf_phys_addr;
+ return rx_desc->pp21.buf_phys_addr;
}
static unsigned long mvpp2_rxdesc_virt_addr_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->buf_cookie;
+ return rx_desc->pp21.buf_cookie;
}
static size_t mvpp2_rxdesc_size_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->data_size;
+ return rx_desc->pp21.data_size;
}
static u32 mvpp2_rxdesc_status_get(struct mvpp2_port *port,
struct mvpp2_rx_desc *rx_desc)
{
- return rx_desc->status;
+ return rx_desc->pp21.status;
}
static void mvpp2_txq_inc_get(struct mvpp2_txq_pcpu *txq_pcpu)
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 14/27] net: mvpp2: add ip_version field in "struct mvpp2"
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
In preparation to the introduction for the support of PPv2.2 in the
mvpp2 driver, this commit adds an ip_version field to the struct
mvpp2, and uses the .data field of the DT match table to fill it in.
Having the MVPP21 and MVPP22 definitions available will allow to start
adding the necessary conditional code to support PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 12 +++++++++++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index e35871f..af00405 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -25,6 +25,7 @@
#include <linux/of_mdio.h>
#include <linux/of_net.h>
#include <linux/of_address.h>
+#include <linux/of_device.h>
#include <linux/phy.h>
#include <linux/clk.h>
#include <linux/hrtimer.h>
@@ -649,6 +650,9 @@ struct mvpp2 {
/* Tclk value */
u32 tclk;
+
+ /* HW IP Version */
+ enum { MVPP21, MVPP22 } ip_version;
};
struct mvpp2_pcpu_stats {
@@ -6497,6 +6501,9 @@ static int mvpp2_probe(struct platform_device *pdev)
if (!priv)
return -ENOMEM;
+ priv->ip_version =
+ (unsigned long)of_device_get_match_data(&pdev->dev);
+
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
priv->base = devm_ioremap_resource(&pdev->dev, res);
if (IS_ERR(priv->base))
@@ -6601,7 +6608,10 @@ static int mvpp2_remove(struct platform_device *pdev)
}
static const struct of_device_id mvpp2_match[] = {
- { .compatible = "marvell,armada-375-pp2" },
+ {
+ .compatible = "marvell,armada-375-pp2",
+ .data = (void *)MVPP21,
+ },
{ }
};
MODULE_DEVICE_TABLE(of, mvpp2_match);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 11/27] net: mvpp2: switch to build_skb() in the RX path
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit adapts the mvpp2 RX path to use the build_skb() method. Not
only build_skb() is now the recommended mechanism, but it also
simplifies the addition of support for the PPv2.2 variant.
Indeed, without build_skb(), we have to keep track for each RX
descriptor of the physical address of the packet buffer, and the virtual
address of the SKB. However, in PPv2.2 running on 64 bits platform,
there is not enough space in the descriptor to store the virtual address
of the SKB. So having to take care only of the address of the packet
buffer, and building the SKB upon reception helps in supporting PPv2.2.
The implementation is fairly straightforward:
- mvpp2_skb_alloc() is renamed to mvpp2_buf_alloc() and no longer
allocates a SKB. Instead, it allocates a buffer using the new
mvpp2_frag_alloc() function, with enough space for the data and SKB.
- The initialization of the RX buffers in mvpp2_bm_bufs_add() as well
as the refill of the RX buffers in mvpp2_rx_refill() is adjusted
accordingly.
- Finally, the mvpp2_rx() is modified to use build_skb().
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 77 +++++++++++++++++++++++++-----------
1 file changed, 55 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index a713e7d..43ddc65 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -918,6 +918,7 @@ struct mvpp2_bm_pool {
int buf_size;
/* Packet size */
int pkt_size;
+ int frag_size;
/* BPPE virtual base address */
u32 *virt_addr;
@@ -3354,6 +3355,22 @@ static void mvpp2_cls_oversize_rxq_set(struct mvpp2_port *port)
mvpp2_write(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG, val);
}
+static void *mvpp2_frag_alloc(const struct mvpp2_bm_pool *pool)
+{
+ if (likely(pool->frag_size <= PAGE_SIZE))
+ return netdev_alloc_frag(pool->frag_size);
+ else
+ return kmalloc(pool->frag_size, GFP_ATOMIC);
+}
+
+static void mvpp2_frag_free(const struct mvpp2_bm_pool *pool, void *data)
+{
+ if (likely(pool->frag_size <= PAGE_SIZE))
+ skb_free_frag(data);
+ else
+ kfree(data);
+}
+
/* Buffer Manager configuration routines */
/* Create pool */
@@ -3428,7 +3445,8 @@ static void mvpp2_bm_bufs_free(struct device *dev, struct mvpp2 *priv,
if (!vaddr)
break;
- dev_kfree_skb_any((struct sk_buff *)vaddr);
+
+ mvpp2_frag_free(bm_pool, (void *)vaddr);
}
/* Update BM driver with number of buffers removed from pool */
@@ -3542,29 +3560,28 @@ static void mvpp2_rxq_short_pool_set(struct mvpp2_port *port,
mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}
-/* Allocate skb for BM pool */
-static struct sk_buff *mvpp2_skb_alloc(struct mvpp2_port *port,
- struct mvpp2_bm_pool *bm_pool,
- dma_addr_t *buf_phys_addr,
- gfp_t gfp_mask)
+static void *mvpp2_buf_alloc(struct mvpp2_port *port,
+ struct mvpp2_bm_pool *bm_pool,
+ dma_addr_t *buf_phys_addr,
+ gfp_t gfp_mask)
{
- struct sk_buff *skb;
dma_addr_t phys_addr;
+ void *data;
- skb = __dev_alloc_skb(bm_pool->pkt_size, gfp_mask);
- if (!skb)
+ data = mvpp2_frag_alloc(bm_pool);
+ if (!data)
return NULL;
- phys_addr = dma_map_single(port->dev->dev.parent, skb->head,
+ phys_addr = dma_map_single(port->dev->dev.parent, data,
MVPP2_RX_BUF_SIZE(bm_pool->pkt_size),
DMA_FROM_DEVICE);
if (unlikely(dma_mapping_error(port->dev->dev.parent, phys_addr))) {
- dev_kfree_skb_any(skb);
+ mvpp2_frag_free(bm_pool, data);
return NULL;
}
*buf_phys_addr = phys_addr;
- return skb;
+ return data;
}
/* Set pool number in a BM cookie */
@@ -3620,9 +3637,9 @@ static void mvpp2_pool_refill(struct mvpp2_port *port, u32 bm,
static int mvpp2_bm_bufs_add(struct mvpp2_port *port,
struct mvpp2_bm_pool *bm_pool, int buf_num)
{
- struct sk_buff *skb;
int i, buf_size, total_size;
dma_addr_t phys_addr;
+ void *buf;
buf_size = MVPP2_RX_BUF_SIZE(bm_pool->pkt_size);
total_size = MVPP2_RX_TOTAL_SIZE(buf_size);
@@ -3636,11 +3653,11 @@ static int mvpp2_bm_bufs_add(struct mvpp2_port *port,
}
for (i = 0; i < buf_num; i++) {
- skb = mvpp2_skb_alloc(port, bm_pool, &phys_addr, GFP_KERNEL);
- if (!skb)
+ buf = mvpp2_buf_alloc(port, bm_pool, &phys_addr, GFP_KERNEL);
+ if (!buf)
break;
- mvpp2_bm_pool_put(port, bm_pool->id, (u32)phys_addr, (u32)skb);
+ mvpp2_bm_pool_put(port, bm_pool->id, (u32)phys_addr, (u32)buf);
}
/* Update BM driver with number of buffers added to pool */
@@ -3696,6 +3713,9 @@ mvpp2_bm_pool_use(struct mvpp2_port *port, int pool, enum mvpp2_bm_type type,
port->priv, new_pool);
new_pool->pkt_size = pkt_size;
+ new_pool->frag_size =
+ SKB_DATA_ALIGN(MVPP2_RX_BUF_SIZE(pkt_size)) +
+ MVPP2_SKB_SHINFO_SIZE;
/* Allocate buffers for this pool */
num = mvpp2_bm_bufs_add(port, new_pool, pkts_num);
@@ -4987,15 +5007,15 @@ static void mvpp2_rx_csum(struct mvpp2_port *port, u32 status,
static int mvpp2_rx_refill(struct mvpp2_port *port,
struct mvpp2_bm_pool *bm_pool, u32 bm)
{
- struct sk_buff *skb;
dma_addr_t phys_addr;
+ void *buf;
/* No recycle or too many buffers are in use, so allocate a new skb */
- skb = mvpp2_skb_alloc(port, bm_pool, &phys_addr, GFP_ATOMIC);
- if (!skb)
+ buf = mvpp2_buf_alloc(port, bm_pool, &phys_addr, GFP_ATOMIC);
+ if (!buf)
return -ENOMEM;
- mvpp2_pool_refill(port, bm, (u32)phys_addr, (u32)skb);
+ mvpp2_pool_refill(port, bm, (u32)phys_addr, (u32)buf);
return 0;
}
@@ -5087,14 +5107,17 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
struct mvpp2_rx_desc *rx_desc = mvpp2_rxq_next_desc_get(rxq);
struct mvpp2_bm_pool *bm_pool;
struct sk_buff *skb;
+ unsigned int frag_size;
dma_addr_t phys_addr;
u32 bm, rx_status;
int pool, rx_bytes, err;
+ void *data;
rx_done++;
rx_status = rx_desc->status;
rx_bytes = rx_desc->data_size - MVPP2_MH_SIZE;
phys_addr = rx_desc->buf_phys_addr;
+ data = (void *)rx_desc->buf_cookie;
bm = mvpp2_bm_cookie_build(rx_desc);
pool = mvpp2_bm_cookie_pool_get(bm);
@@ -5115,12 +5138,22 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
dev->stats.rx_errors++;
mvpp2_rx_error(port, rx_desc);
/* Return the buffer to the pool */
+
mvpp2_pool_refill(port, bm, rx_desc->buf_phys_addr,
rx_desc->buf_cookie);
continue;
}
- skb = (struct sk_buff *)rx_desc->buf_cookie;
+ if (bm_pool->frag_size > PAGE_SIZE)
+ frag_size = 0;
+ else
+ frag_size = bm_pool->frag_size;
+
+ skb = build_skb(data, frag_size);
+ if (!skb) {
+ netdev_warn(port->dev, "skb build failed\n");
+ goto err_drop_frame;
+ }
err = mvpp2_rx_refill(port, bm_pool, bm);
if (err) {
@@ -5134,7 +5167,7 @@ static int mvpp2_rx(struct mvpp2_port *port, int rx_todo,
rcvd_pkts++;
rcvd_bytes += rx_bytes;
- skb_reserve(skb, MVPP2_MH_SIZE);
+ skb_reserve(skb, MVPP2_MH_SIZE + NET_SKB_PAD);
skb_put(skb, rx_bytes);
skb->protocol = eth_type_trans(skb, dev);
mvpp2_rx_csum(port, rx_status, skb);
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 05/27] net: mvpp2: remove unused 'tx_skb' field of 'struct mvpp2_tx_queue'
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
This commit remove a field of 'struct mvpp2_tx_queue' that is not used
anywhere.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 84eb5fb..fe854b3 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -823,9 +823,6 @@ struct mvpp2_tx_queue {
/* Per-CPU control of physical Tx queues */
struct mvpp2_txq_pcpu __percpu *pcpu;
- /* Array of transmitted skb */
- struct sk_buff **tx_skb;
-
u32 done_pkts_coal;
/* Virtual address of thex Tx DMA descriptors array */
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 08/27] net: mvpp2: remove unused register definitions
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 4 ----
1 file changed, 4 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 98edcb6..e11eb6f 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -253,10 +253,6 @@
#define MVPP2_SRC_ADDR_HIGH 0x28
#define MVPP2_PHY_AN_CFG0_REG 0x34
#define MVPP2_PHY_AN_STOP_SMI0_MASK BIT(7)
-#define MVPP2_MIB_COUNTERS_BASE(port) (0x1000 + ((port) >> 1) * \
- 0x400 + (port) * 0x400)
-#define MVPP2_MIB_LATE_COLLISION 0x7c
-#define MVPP2_ISR_SUM_MASK_REG 0x220c
#define MVPP2_MNG_EXTENDED_GLOBAL_CTRL_REG 0x305c
#define MVPP2_EXT_GLOBAL_CTRL_DEFAULT 0x27
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 03/27] net: mvpp2: handle too large value in mvpp2_rx_time_coal_set()
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
When configuring the MVPP2_ISR_RX_THRESHOLD_REG with the RX coalescing
time threshold, we do not check for the maximum allowed value supported
by the driver, which means we might overflow and use a bogus value. This
commit adds a check for this situation, and if a value higher than what
is supported by the hardware is provided, then we use the maximum value
supported by the hardware.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 4c9e3ea..12edefe 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -154,6 +154,7 @@
/* Interrupt Cause and Mask registers */
#define MVPP2_ISR_RX_THRESHOLD_REG(rxq) (0x5200 + 4 * (rxq))
+#define MVPP2_MAX_ISR_RX_THRESHOLD 0xfffff0
#define MVPP2_ISR_RXQ_GROUP_REG(rxq) (0x5400 + 4 * (rxq))
#define MVPP2_ISR_ENABLE_REG(port) (0x5420 + 4 * (port))
#define MVPP2_ISR_ENABLE_INTERRUPT(mask) ((mask) & 0xffff)
@@ -4397,6 +4398,12 @@ static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
u32 val;
val = (port->priv->tclk / USEC_PER_SEC) * usec;
+
+ if (val > MVPP2_MAX_ISR_RX_THRESHOLD) {
+ val = MVPP2_MAX_ISR_RX_THRESHOLD;
+ usec = (val * USEC_PER_SEC) / port->priv->tclk;
+ }
+
mvpp2_write(port->priv, MVPP2_ISR_RX_THRESHOLD_REG(rxq->id), val);
rxq->time_coal = usec;
--
2.7.4
^ permalink raw reply related
* [PATCH net-next 02/27] net: mvpp2: handle too large value handling in mvpp2_rx_pkts_coal_set()
From: Thomas Petazzoni @ 2016-12-21 11:16 UTC (permalink / raw)
To: David S. Miller, netdev
Cc: Nadav Haklai, Hanna Hawa, Yehuda Yitschak, Jason Cooper,
Andrew Lunn, Sebastian Hesselbarth, Gregory Clement,
linux-arm-kernel, Stefan Chulski, Marcin Wojtas, Thomas Petazzoni
In-Reply-To: <1482318994-23488-1-git-send-email-thomas.petazzoni@free-electrons.com>
Currently, mvpp2_rx_pkts_coal_set() does the following to avoid setting
a too large value for the RX coalescing by packet number:
val = (pkts & MVPP2_OCCUPIED_THRESH_MASK);
This means that if you set a value that is slightly higher the the
maximum number of packets, you in fact get a very low value. It makes a
lot more sense to simply check if the value is too high, and if it's too
high, limit it to the maximum possible value.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/net/ethernet/marvell/mvpp2.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index 930c816..4c9e3ea 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -4381,11 +4381,11 @@ static void mvpp2_txp_max_tx_size_set(struct mvpp2_port *port)
static void mvpp2_rx_pkts_coal_set(struct mvpp2_port *port,
struct mvpp2_rx_queue *rxq, u32 pkts)
{
- u32 val;
+ if (pkts > MVPP2_OCCUPIED_THRESH_MASK)
+ pkts = MVPP2_OCCUPIED_THRESH_MASK;
- val = (pkts & MVPP2_OCCUPIED_THRESH_MASK);
mvpp2_write(port->priv, MVPP2_RXQ_NUM_REG, rxq->id);
- mvpp2_write(port->priv, MVPP2_RXQ_THRESH_REG, val);
+ mvpp2_write(port->priv, MVPP2_RXQ_THRESH_REG, pkts);
rxq->pkts_coal = pkts;
}
--
2.7.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox