* [PATCH 5/6] i40iw: Assign MSS only when it is a new MTU
From: Henry Orosco @ 2016-12-06 21:49 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA,
e1000-rdma-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, Mustafa Ismail,
Henry Orosco
In-Reply-To: <20161206214935.41584-1-henry.orosco-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
From: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Currently we are changing the MSS regardless of whether
there is a change or not in MTU. Fix to make the
assignment of MSS dependent on an MTU change.
Signed-off-by: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Henry Orosco <henry.orosco-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
drivers/infiniband/hw/i40iw/i40iw_main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c
index 85d8fa6..cf9d288 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_main.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_main.c
@@ -1724,6 +1724,8 @@ static void i40iw_l2param_change(struct i40e_info *ldev, struct i40e_client *cli
for (i = 0; i < I40E_CLIENT_MAX_USER_PRIORITY; i++)
l2params->qs_handle_list[i] = params->qos.prio_qos[i].qs_handle;
+ l2params->mss = (params->mtu) ? params->mtu - I40IW_MTU_TO_MSS : iwdev->mss;
+
INIT_WORK(&work->work, i40iw_l2params_worker);
queue_work(iwdev->param_wq, &work->work);
}
--
2.8.3
--
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 6/6] i40iw: Fix incorrect check for error
From: Henry Orosco @ 2016-12-06 21:49 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA,
e1000-rdma-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, Mustafa Ismail,
Henry Orosco
In-Reply-To: <20161206214935.41584-1-henry.orosco-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
From: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
In i40iw_ieq_handle_partial() the check for !status is incorrect.
Signed-off-by: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Henry Orosco <henry.orosco-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
drivers/infiniband/hw/i40iw/i40iw_puda.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.c b/drivers/infiniband/hw/i40iw/i40iw_puda.c
index 7541b0d..c3d28ba 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_puda.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_puda.c
@@ -1132,7 +1132,7 @@ static enum i40iw_status_code i40iw_ieq_handle_partial(struct i40iw_puda_rsrc *i
list_add(&buf->list, &pbufl);
status = i40iw_ieq_create_pbufl(pfpdu, rxlist, &pbufl, buf, fpdu_len);
- if (!status)
+ if (status)
goto error;
txbuf = i40iw_puda_get_bufpool(ieq);
--
2.8.3
--
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: Enabling peer to peer device transactions for PCIe devices
From: Dan Williams @ 2016-12-06 22:02 UTC (permalink / raw)
To: Logan Gunthorpe
Cc: Haggai Eran, John.Bridgman-5C7GfCeVMHo@public.gmane.org,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org,
Felix.Kuehling-5C7GfCeVMHo@public.gmane.org,
serguei.sagalovitch-5C7GfCeVMHo@public.gmane.org,
Paul.Blinzer-5C7GfCeVMHo@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
dri-devel-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org,
Stephen Bates, Jason Gunthorpe,
ben.sander-5C7GfCeVMHo@public.gmane.org,
Suravee.Suthikulpanit-5C7GfCeVMHo@public.gmane.org,
linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Alexander.Deucher-5C7GfCeVMHo@public.gmane.org, Max Gurtovoy,
christian.koenig-5C7GfCeVMHo@public.gmane.org
In-Reply-To: <bedaa7a2-e42d-da83-5c2b-9d639b0397c5-OTvnGxWRz7hWk0Htik3J/w@public.gmane.org>
On Tue, Dec 6, 2016 at 1:47 PM, Logan Gunthorpe <logang-OTvnGxWRz7hWk0Htik3J/w@public.gmane.org> wrote:
> Hey,
>
>> Okay, so clearly this needs a kernel side NVMe specific allocator
>> and locking so users don't step on each other..
>
> Yup, ideally. That's why device dax isn't ideal for this application: it
> doesn't provide any way to prevent users from stepping on each other.
On this particular point I'm in the process of posting patches that
allow device-dax sub-division, so you could carve up a bar into
multiple devices of various sizes.
^ permalink raw reply
* RE: [PATCH rdma-next 01/10] IB/core: Add raw packet protocol
From: Hefty, Sean @ 2016-12-06 22:13 UTC (permalink / raw)
To: Jason Gunthorpe, Or Gerlitz
Cc: Doug Ledford, Steve Wise,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Liran Liss,
Matan Barak, Leon Romanovsky
In-Reply-To: <20161206213938.GC647-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
> I've already proposed disallowing multiprotocol struct ib_devices.
My preference is to discontinue attempts at associating a protocol with the device. A device could implement a dozen protocols in software. Transports belong to QPs or cm ids, not devices. Each rdma_cm_id should be associated with a specific cm/transport directly, rather than indirectly selecting one based on the bound <device, port>.
If an app wants a specific transport type for a QP, why doesn't it just try to open one and see if the call fails?
--
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
* [PATCH] i40iw: Reorganize structures to align with HW capabilities
From: Henry Orosco @ 2016-12-06 22:16 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA,
e1000-rdma-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, Henry Orosco
Some resources are incorrectly organized and at odds with
HW capabilities. Specifically, ILQ, IEQ, QPs, MSS, QOS
and statistics belong in a VSI.
Signed-off-by: Faisal Latif <faisal.latif-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Henry Orosco <henry.orosco-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
drivers/infiniband/hw/i40iw/i40iw.h | 7 +-
drivers/infiniband/hw/i40iw/i40iw_cm.c | 36 +-
drivers/infiniband/hw/i40iw/i40iw_ctrl.c | 527 +++++++++++++++------------
drivers/infiniband/hw/i40iw/i40iw_d.h | 21 +-
drivers/infiniband/hw/i40iw/i40iw_hw.c | 4 +-
drivers/infiniband/hw/i40iw/i40iw_main.c | 53 ++-
drivers/infiniband/hw/i40iw/i40iw_osdep.h | 6 +-
drivers/infiniband/hw/i40iw/i40iw_p.h | 23 +-
drivers/infiniband/hw/i40iw/i40iw_puda.c | 268 ++++++++------
drivers/infiniband/hw/i40iw/i40iw_puda.h | 20 +-
drivers/infiniband/hw/i40iw/i40iw_type.h | 79 ++--
drivers/infiniband/hw/i40iw/i40iw_utils.c | 150 +++++++-
drivers/infiniband/hw/i40iw/i40iw_verbs.c | 17 +-
drivers/infiniband/hw/i40iw/i40iw_virtchnl.c | 29 +-
14 files changed, 775 insertions(+), 465 deletions(-)
diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h
index 2aab85b..da2eb5a 100644
--- a/drivers/infiniband/hw/i40iw/i40iw.h
+++ b/drivers/infiniband/hw/i40iw/i40iw.h
@@ -236,6 +236,7 @@ struct i40iw_device {
struct net_device *netdev;
wait_queue_head_t vchnl_waitq;
struct i40iw_sc_dev sc_dev;
+ struct i40iw_sc_vsi vsi;
struct i40iw_handler *hdl;
struct i40e_info *ldev;
struct i40e_client *client;
@@ -289,7 +290,6 @@ struct i40iw_device {
u32 sd_type;
struct workqueue_struct *param_wq;
atomic_t params_busy;
- u32 mss;
enum init_completion_state init_state;
u16 mac_ip_table_idx;
atomic_t vchnl_msgs;
@@ -525,6 +525,7 @@ enum i40iw_status_code i40iw_handle_cqp_op(struct i40iw_device *iwdev,
enum i40iw_status_code i40iw_add_mac_addr(struct i40iw_device *iwdev,
u8 *mac_addr, u8 *mac_index);
int i40iw_modify_qp(struct ib_qp *, struct ib_qp_attr *, int, struct ib_udata *);
+void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq);
void i40iw_rem_pdusecount(struct i40iw_pd *iwpd, struct i40iw_device *iwdev);
void i40iw_add_pdusecount(struct i40iw_pd *iwpd);
@@ -542,8 +543,8 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev,
enum i40iw_quad_hash_manage_type mtype,
void *cmnode,
bool wait);
-void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf);
-void i40iw_free_sqbuf(struct i40iw_sc_dev *dev, void *bufp);
+void i40iw_receive_ilq(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *rbuf);
+void i40iw_free_sqbuf(struct i40iw_sc_vsi *vsi, void *bufp);
void i40iw_free_qp_resources(struct i40iw_device *iwdev,
struct i40iw_qp *iwqp,
u32 qp_num);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c
index a217d2f..e4820be 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c
@@ -68,13 +68,13 @@ static void i40iw_disconnect_worker(struct work_struct *work);
/**
* i40iw_free_sqbuf - put back puda buffer if refcount = 0
- * @dev: FPK device
+ * @vsi: pointer to vsi structure
* @buf: puda buffer to free
*/
-void i40iw_free_sqbuf(struct i40iw_sc_dev *dev, void *bufp)
+void i40iw_free_sqbuf(struct i40iw_sc_vsi *vsi, void *bufp)
{
struct i40iw_puda_buf *buf = (struct i40iw_puda_buf *)bufp;
- struct i40iw_puda_rsrc *ilq = dev->ilq;
+ struct i40iw_puda_rsrc *ilq = vsi->ilq;
if (!atomic_dec_return(&buf->refcount))
i40iw_puda_ret_bufpool(ilq, buf);
@@ -337,13 +337,13 @@ static struct i40iw_cm_event *i40iw_create_event(struct i40iw_cm_node *cm_node,
*/
static void i40iw_free_retrans_entry(struct i40iw_cm_node *cm_node)
{
- struct i40iw_sc_dev *dev = cm_node->dev;
+ struct i40iw_device *iwdev = cm_node->iwdev;
struct i40iw_timer_entry *send_entry;
send_entry = cm_node->send_entry;
if (send_entry) {
cm_node->send_entry = NULL;
- i40iw_free_sqbuf(dev, (void *)send_entry->sqbuf);
+ i40iw_free_sqbuf(&iwdev->vsi, (void *)send_entry->sqbuf);
kfree(send_entry);
atomic_dec(&cm_node->ref_count);
}
@@ -377,7 +377,7 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node,
u8 flags)
{
struct i40iw_puda_buf *sqbuf;
- struct i40iw_sc_dev *dev = cm_node->dev;
+ struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi;
u8 *buf;
struct tcphdr *tcph;
@@ -391,7 +391,7 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node,
u32 hdr_len = 0;
u16 vtag;
- sqbuf = i40iw_puda_get_bufpool(dev->ilq);
+ sqbuf = i40iw_puda_get_bufpool(vsi->ilq);
if (!sqbuf)
return NULL;
buf = sqbuf->mem.va;
@@ -1059,7 +1059,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node,
int send_retrans,
int close_when_complete)
{
- struct i40iw_sc_dev *dev = cm_node->dev;
+ struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi;
struct i40iw_cm_core *cm_core = cm_node->cm_core;
struct i40iw_timer_entry *new_send;
int ret = 0;
@@ -1068,7 +1068,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node,
new_send = kzalloc(sizeof(*new_send), GFP_ATOMIC);
if (!new_send) {
- i40iw_free_sqbuf(cm_node->dev, (void *)sqbuf);
+ i40iw_free_sqbuf(vsi, (void *)sqbuf);
return -ENOMEM;
}
new_send->retrycount = I40IW_DEFAULT_RETRYS;
@@ -1083,7 +1083,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node,
new_send->timetosend += (HZ / 10);
if (cm_node->close_entry) {
kfree(new_send);
- i40iw_free_sqbuf(cm_node->dev, (void *)sqbuf);
+ i40iw_free_sqbuf(vsi, (void *)sqbuf);
i40iw_pr_err("already close entry\n");
return -EINVAL;
}
@@ -1098,7 +1098,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node,
new_send->timetosend = jiffies + I40IW_RETRY_TIMEOUT;
atomic_inc(&sqbuf->refcount);
- i40iw_puda_send_buf(dev->ilq, sqbuf);
+ i40iw_puda_send_buf(vsi->ilq, sqbuf);
if (!send_retrans) {
i40iw_cleanup_retrans_entry(cm_node);
if (close_when_complete)
@@ -1195,6 +1195,7 @@ static void i40iw_cm_timer_tick(unsigned long pass)
struct i40iw_cm_node *cm_node;
struct i40iw_timer_entry *send_entry, *close_entry;
struct list_head *list_core_temp;
+ struct i40iw_sc_vsi *vsi;
struct list_head *list_node;
struct i40iw_cm_core *cm_core = (struct i40iw_cm_core *)pass;
u32 settimer = 0;
@@ -1270,9 +1271,10 @@ static void i40iw_cm_timer_tick(unsigned long pass)
cm_node->cm_core->stats_pkt_retrans++;
spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
+ vsi = &cm_node->iwdev->vsi;
dev = cm_node->dev;
atomic_inc(&send_entry->sqbuf->refcount);
- i40iw_puda_send_buf(dev->ilq, send_entry->sqbuf);
+ i40iw_puda_send_buf(vsi->ilq, send_entry->sqbuf);
spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
if (send_entry->send_retrans) {
send_entry->retranscount--;
@@ -1373,10 +1375,11 @@ int i40iw_send_syn(struct i40iw_cm_node *cm_node, u32 sendack)
static void i40iw_send_ack(struct i40iw_cm_node *cm_node)
{
struct i40iw_puda_buf *sqbuf;
+ struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi;
sqbuf = i40iw_form_cm_frame(cm_node, NULL, NULL, NULL, SET_ACK);
if (sqbuf)
- i40iw_puda_send_buf(cm_node->dev->ilq, sqbuf);
+ i40iw_puda_send_buf(vsi->ilq, sqbuf);
else
i40iw_pr_err("no sqbuf\n");
}
@@ -2179,7 +2182,7 @@ static struct i40iw_cm_node *i40iw_make_cm_node(
I40IW_CM_DEFAULT_RCV_WND_SCALED >> I40IW_CM_DEFAULT_RCV_WND_SCALE;
ts = current_kernel_time();
cm_node->tcp_cntxt.loc_seq_num = ts.tv_nsec;
- cm_node->tcp_cntxt.mss = iwdev->mss;
+ cm_node->tcp_cntxt.mss = iwdev->vsi.mss;
cm_node->iwdev = iwdev;
cm_node->dev = &iwdev->sc_dev;
@@ -3059,10 +3062,10 @@ static int i40iw_cm_close(struct i40iw_cm_node *cm_node)
/**
* i40iw_receive_ilq - recv an ETHERNET packet, and process it
* through CM
- * @dev: FPK dev struct
+ * @vsi: pointer to the vsi structure
* @rbuf: receive buffer
*/
-void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf)
+void i40iw_receive_ilq(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *rbuf)
{
struct i40iw_cm_node *cm_node;
struct i40iw_cm_listener *listener;
@@ -3070,6 +3073,7 @@ void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf)
struct ipv6hdr *ip6h;
struct tcphdr *tcph;
struct i40iw_cm_info cm_info;
+ struct i40iw_sc_dev *dev = vsi->dev;
struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev;
struct i40iw_cm_core *cm_core = &iwdev->cm_core;
struct vlan_ethhdr *ethh;
diff --git a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c
index a135037..392f783 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c
@@ -103,6 +103,7 @@ static enum i40iw_status_code i40iw_cqp_poll_registers(
if (newtail != tail) {
/* SUCCESS */
I40IW_RING_MOVE_TAIL(cqp->sq_ring);
+ cqp->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS]++;
return 0;
}
udelay(I40IW_SLEEP_COUNT);
@@ -276,11 +277,12 @@ static struct i40iw_sc_qp *i40iw_get_qp(struct list_head *head, struct i40iw_sc_
/**
* i40iw_change_l2params - given the new l2 parameters, change all qp
- * @dev: IWARP device pointer
+ * @vsi: pointer to the vsi structure
* @l2params: New paramaters from l2
*/
-void i40iw_change_l2params(struct i40iw_sc_dev *dev, struct i40iw_l2params *l2params)
+void i40iw_change_l2params(struct i40iw_sc_vsi *vsi, struct i40iw_l2params *l2params)
{
+ struct i40iw_sc_dev *dev = vsi->dev;
struct i40iw_sc_qp *qp = NULL;
bool qs_handle_change = false;
bool mss_change = false;
@@ -288,20 +290,20 @@ void i40iw_change_l2params(struct i40iw_sc_dev *dev, struct i40iw_l2params *l2pa
u16 qs_handle;
int i;
- if (dev->mss != l2params->mss) {
+ if (vsi->mss != l2params->mss) {
mss_change = true;
- dev->mss = l2params->mss;
+ vsi->mss = l2params->mss;
}
i40iw_fill_qos_list(l2params->qs_handle_list);
for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) {
qs_handle = l2params->qs_handle_list[i];
- if (dev->qos[i].qs_handle != qs_handle)
+ if (vsi->qos[i].qs_handle != qs_handle)
qs_handle_change = true;
else if (!mss_change)
continue; /* no MSS nor qs handle change */
- spin_lock_irqsave(&dev->qos[i].lock, flags);
- qp = i40iw_get_qp(&dev->qos[i].qplist, qp);
+ spin_lock_irqsave(&vsi->qos[i].lock, flags);
+ qp = i40iw_get_qp(&vsi->qos[i].qplist, qp);
while (qp) {
if (mss_change)
i40iw_qp_mss_modify(dev, qp);
@@ -310,43 +312,45 @@ void i40iw_change_l2params(struct i40iw_sc_dev *dev, struct i40iw_l2params *l2pa
/* issue cqp suspend command */
i40iw_qp_suspend_resume(dev, qp, true);
}
- qp = i40iw_get_qp(&dev->qos[i].qplist, qp);
+ qp = i40iw_get_qp(&vsi->qos[i].qplist, qp);
}
- spin_unlock_irqrestore(&dev->qos[i].lock, flags);
- dev->qos[i].qs_handle = qs_handle;
+ spin_unlock_irqrestore(&vsi->qos[i].lock, flags);
+ vsi->qos[i].qs_handle = qs_handle;
}
}
/**
* i40iw_qp_rem_qos - remove qp from qos lists during destroy qp
- * @dev: IWARP device pointer
* @qp: qp to be removed from qos
*/
-static void i40iw_qp_rem_qos(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
+static void i40iw_qp_rem_qos(struct i40iw_sc_qp *qp)
{
+ struct i40iw_sc_vsi *vsi = qp->vsi;
unsigned long flags;
if (!qp->on_qoslist)
return;
- spin_lock_irqsave(&dev->qos[qp->user_pri].lock, flags);
+ spin_lock_irqsave(&vsi->qos[qp->user_pri].lock, flags);
list_del(&qp->list);
- spin_unlock_irqrestore(&dev->qos[qp->user_pri].lock, flags);
+ spin_unlock_irqrestore(&vsi->qos[qp->user_pri].lock, flags);
}
/**
* i40iw_qp_add_qos - called during setctx fot qp to be added to qos
- * @dev: IWARP device pointer
* @qp: qp to be added to qos
*/
-void i40iw_qp_add_qos(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
+void i40iw_qp_add_qos(struct i40iw_sc_qp *qp)
{
+ struct i40iw_sc_vsi *vsi = qp->vsi;
unsigned long flags;
- spin_lock_irqsave(&dev->qos[qp->user_pri].lock, flags);
- qp->qs_handle = dev->qos[qp->user_pri].qs_handle;
- list_add(&qp->list, &dev->qos[qp->user_pri].qplist);
+ if (qp->on_qoslist)
+ return;
+ spin_lock_irqsave(&vsi->qos[qp->user_pri].lock, flags);
+ qp->qs_handle = vsi->qos[qp->user_pri].qs_handle;
+ list_add(&qp->list, &vsi->qos[qp->user_pri].qplist);
qp->on_qoslist = true;
- spin_unlock_irqrestore(&dev->qos[qp->user_pri].lock, flags);
+ spin_unlock_irqrestore(&vsi->qos[qp->user_pri].lock, flags);
}
/**
@@ -419,6 +423,9 @@ static enum i40iw_status_code i40iw_sc_cqp_init(struct i40iw_sc_cqp *cqp,
info->dev->cqp = cqp;
I40IW_RING_INIT(cqp->sq_ring, cqp->sq_size);
+ cqp->dev->cqp_cmd_stats[OP_REQUESTED_COMMANDS] = 0;
+ cqp->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS] = 0;
+
i40iw_debug(cqp->dev, I40IW_DEBUG_WQE,
"%s: sq_size[%04d] hw_sq_size[%04d] sq_base[%p] sq_pa[%llxh] cqp[%p] polarity[x%04X]\n",
__func__, cqp->sq_size, cqp->hw_sq_size,
@@ -546,6 +553,7 @@ u64 *i40iw_sc_cqp_get_next_send_wqe(struct i40iw_sc_cqp *cqp, u64 scratch)
return NULL;
}
I40IW_ATOMIC_RING_MOVE_HEAD(cqp->sq_ring, wqe_idx, ret_code);
+ cqp->dev->cqp_cmd_stats[OP_REQUESTED_COMMANDS]++;
if (ret_code)
return NULL;
if (!wqe_idx)
@@ -681,6 +689,8 @@ static enum i40iw_status_code i40iw_sc_ccq_get_cqe_info(
I40IW_RING_GETCURRENT_HEAD(ccq->cq_uk.cq_ring));
wmb(); /* write shadow area before tail */
I40IW_RING_MOVE_TAIL(cqp->sq_ring);
+ ccq->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS]++;
+
return ret_code;
}
@@ -1173,6 +1183,7 @@ static enum i40iw_status_code i40iw_sc_manage_qhash_table_entry(
u64 qw1 = 0;
u64 qw2 = 0;
u64 temp;
+ struct i40iw_sc_vsi *vsi = info->vsi;
wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch);
if (!wqe)
@@ -1204,7 +1215,7 @@ static enum i40iw_status_code i40iw_sc_manage_qhash_table_entry(
LS_64(info->dest_ip[2], I40IW_CQPSQ_QHASH_ADDR2) |
LS_64(info->dest_ip[3], I40IW_CQPSQ_QHASH_ADDR3));
}
- qw2 = LS_64(cqp->dev->qos[info->user_pri].qs_handle, I40IW_CQPSQ_QHASH_QS_HANDLE);
+ qw2 = LS_64(vsi->qos[info->user_pri].qs_handle, I40IW_CQPSQ_QHASH_QS_HANDLE);
if (info->vlan_valid)
qw2 |= LS_64(info->vlan_id, I40IW_CQPSQ_QHASH_VLANID);
set_64bit_val(wqe, 16, qw2);
@@ -2225,6 +2236,7 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp,
u32 offset;
qp->dev = info->pd->dev;
+ qp->vsi = info->vsi;
qp->sq_pa = info->sq_pa;
qp->rq_pa = info->rq_pa;
qp->hw_host_ctx_pa = info->host_ctx_pa;
@@ -2273,7 +2285,7 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp,
qp->rq_tph_en = info->rq_tph_en;
qp->rcv_tph_en = info->rcv_tph_en;
qp->xmit_tph_en = info->xmit_tph_en;
- qp->qs_handle = qp->pd->dev->qos[qp->user_pri].qs_handle;
+ qp->qs_handle = qp->vsi->qos[qp->user_pri].qs_handle;
qp->exception_lan_queue = qp->pd->dev->exception_lan_queue;
return 0;
@@ -2418,7 +2430,7 @@ static enum i40iw_status_code i40iw_sc_qp_destroy(
struct i40iw_sc_cqp *cqp;
u64 header;
- i40iw_qp_rem_qos(qp->pd->dev, qp);
+ i40iw_qp_rem_qos(qp);
cqp = qp->pd->dev->cqp;
wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch);
if (!wqe)
@@ -2566,13 +2578,17 @@ static enum i40iw_status_code i40iw_sc_qp_setctx(
{
struct i40iwarp_offload_info *iw;
struct i40iw_tcp_offload_info *tcp;
+ struct i40iw_sc_vsi *vsi;
+ struct i40iw_sc_dev *dev;
u64 qw0, qw3, qw7 = 0;
iw = info->iwarp_info;
tcp = info->tcp_info;
+ vsi = qp->vsi;
+ dev = qp->dev;
if (info->add_to_qoslist) {
qp->user_pri = info->user_pri;
- i40iw_qp_add_qos(qp->pd->dev, qp);
+ i40iw_qp_add_qos(qp);
i40iw_debug(qp->dev, I40IW_DEBUG_DCB, "%s qp[%d] UP[%d] qset[%d]\n",
__func__, qp->qp_uk.qp_id, qp->user_pri, qp->qs_handle);
}
@@ -2616,7 +2632,10 @@ static enum i40iw_status_code i40iw_sc_qp_setctx(
LS_64(iw->rdmap_ver, I40IWQPC_RDMAP_VER);
qw7 |= LS_64(iw->pd_id, I40IWQPC_PDIDX);
- set_64bit_val(qp_ctx, 144, qp->q2_pa);
+ set_64bit_val(qp_ctx,
+ 144,
+ LS_64(qp->q2_pa, I40IWQPC_Q2ADDR) |
+ LS_64(vsi->fcn_id, I40IWQPC_STAT_INDEX));
set_64bit_val(qp_ctx,
152,
LS_64(iw->last_byte_sent, I40IWQPC_LASTBYTESENT));
@@ -2631,6 +2650,9 @@ static enum i40iw_status_code i40iw_sc_qp_setctx(
LS_64(iw->bind_en, I40IWQPC_BINDEN) |
LS_64(iw->fast_reg_en, I40IWQPC_FASTREGEN) |
LS_64(iw->priv_mode_en, I40IWQPC_PRIVEN) |
+ LS_64((((vsi->stats_fcn_id_alloc) &&
+ (dev->is_pf) && (vsi->fcn_id >= I40IW_FIRST_NON_PF_STAT)) ? 1 : 0),
+ I40IWQPC_USESTATSINSTANCE) |
LS_64(1, I40IWQPC_IWARPMODE) |
LS_64(iw->rcv_mark_en, I40IWQPC_RCVMARKERS) |
LS_64(iw->align_hdrs, I40IWQPC_ALIGNHDRS) |
@@ -4447,286 +4469,370 @@ void i40iw_terminate_received(struct i40iw_sc_qp *qp, struct i40iw_aeqe_info *in
}
/**
- * i40iw_hw_stat_init - Initiliaze HW stats table
- * @devstat: pestat struct
+ * i40iw_sc_vsi_init - Initialize virtual device
+ * @vsi: pointer to the vsi structure
+ * @info: parameters to initialize vsi
+ **/
+void i40iw_sc_vsi_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_init_info *info)
+{
+ int i;
+
+ vsi->dev = info->dev;
+ vsi->back_vsi = info->back_vsi;
+ vsi->mss = info->params->mss;
+ i40iw_fill_qos_list(info->params->qs_handle_list);
+
+ for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) {
+ vsi->qos[i].qs_handle =
+ info->params->qs_handle_list[i];
+ i40iw_debug(vsi->dev, I40IW_DEBUG_DCB, "qset[%d]: %d\n", i, vsi->qos[i].qs_handle);
+ spin_lock_init(&vsi->qos[i].lock);
+ INIT_LIST_HEAD(&vsi->qos[i].qplist);
+ }
+}
+
+/**
+ * i40iw_hw_stats_init - Initiliaze HW stats table
+ * @stats: pestat struct
* @fcn_idx: PCI fn id
- * @hw: PF i40iw_hw structure.
* @is_pf: Is it a PF?
*
- * Populate the HW stat table with register offset addr for each
- * stat. And start the perioidic stats timer.
+ * Populate the HW stats table with register offset addr for each
+ * stats. And start the perioidic stats timer.
*/
-static void i40iw_hw_stat_init(struct i40iw_dev_pestat *devstat,
- u8 fcn_idx,
- struct i40iw_hw *hw, bool is_pf)
+void i40iw_hw_stats_init(struct i40iw_vsi_pestat *stats, u8 fcn_idx, bool is_pf)
{
- u32 stat_reg_offset;
- u32 stat_index;
- struct i40iw_dev_hw_stat_offsets *stat_table =
- &devstat->hw_stat_offsets;
- struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats;
-
- devstat->hw = hw;
+ u32 stats_reg_offset;
+ u32 stats_index;
+ struct i40iw_dev_hw_stats_offsets *stats_table =
+ &stats->hw_stats_offsets;
+ struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats;
if (is_pf) {
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] =
I40E_GLPES_PFIP4RXDISCARD(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] =
I40E_GLPES_PFIP4RXTRUNC(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] =
I40E_GLPES_PFIP4TXNOROUTE(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] =
I40E_GLPES_PFIP6RXDISCARD(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] =
I40E_GLPES_PFIP6RXTRUNC(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] =
I40E_GLPES_PFIP6TXNOROUTE(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] =
I40E_GLPES_PFTCPRTXSEG(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] =
I40E_GLPES_PFTCPRXOPTERR(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] =
I40E_GLPES_PFTCPRXPROTOERR(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] =
I40E_GLPES_PFIP4RXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] =
I40E_GLPES_PFIP4RXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] =
I40E_GLPES_PFIP4RXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] =
I40E_GLPES_PFIP4RXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] =
I40E_GLPES_PFIP4TXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] =
I40E_GLPES_PFIP4TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] =
I40E_GLPES_PFIP4TXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] =
I40E_GLPES_PFIP4TXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] =
I40E_GLPES_PFIP6RXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] =
I40E_GLPES_PFIP6RXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] =
I40E_GLPES_PFIP6RXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] =
I40E_GLPES_PFIP6RXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] =
I40E_GLPES_PFIP6TXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
I40E_GLPES_PFIP6TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
I40E_GLPES_PFIP6TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] =
I40E_GLPES_PFIP6TXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] =
I40E_GLPES_PFTCPRXSEGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] =
I40E_GLPES_PFTCPTXSEGLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] =
I40E_GLPES_PFRDMARXRDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] =
I40E_GLPES_PFRDMARXSNDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] =
I40E_GLPES_PFRDMARXWRSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] =
I40E_GLPES_PFRDMATXRDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] =
I40E_GLPES_PFRDMATXSNDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] =
I40E_GLPES_PFRDMATXWRSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] =
I40E_GLPES_PFRDMAVBNDLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] =
I40E_GLPES_PFRDMAVINVLO(fcn_idx);
} else {
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] =
I40E_GLPES_VFIP4RXDISCARD(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] =
I40E_GLPES_VFIP4RXTRUNC(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] =
I40E_GLPES_VFIP4TXNOROUTE(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] =
I40E_GLPES_VFIP6RXDISCARD(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] =
I40E_GLPES_VFIP6RXTRUNC(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] =
I40E_GLPES_VFIP6TXNOROUTE(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] =
I40E_GLPES_VFTCPRTXSEG(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] =
I40E_GLPES_VFTCPRXOPTERR(fcn_idx);
- stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] =
+ stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] =
I40E_GLPES_VFTCPRXPROTOERR(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] =
I40E_GLPES_VFIP4RXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] =
I40E_GLPES_VFIP4RXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] =
I40E_GLPES_VFIP4RXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] =
I40E_GLPES_VFIP4RXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] =
I40E_GLPES_VFIP4TXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] =
I40E_GLPES_VFIP4TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] =
I40E_GLPES_VFIP4TXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] =
I40E_GLPES_VFIP4TXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] =
I40E_GLPES_VFIP6RXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] =
I40E_GLPES_VFIP6RXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] =
I40E_GLPES_VFIP6RXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] =
I40E_GLPES_VFIP6RXMCPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] =
I40E_GLPES_VFIP6TXOCTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
I40E_GLPES_VFIP6TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] =
I40E_GLPES_VFIP6TXPKTSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] =
I40E_GLPES_VFIP6TXFRAGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] =
I40E_GLPES_VFTCPRXSEGSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] =
I40E_GLPES_VFTCPTXSEGLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] =
I40E_GLPES_VFRDMARXRDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] =
I40E_GLPES_VFRDMARXSNDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] =
I40E_GLPES_VFRDMARXWRSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] =
I40E_GLPES_VFRDMATXRDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] =
I40E_GLPES_VFRDMATXSNDSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] =
I40E_GLPES_VFRDMATXWRSLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] =
I40E_GLPES_VFRDMAVBNDLO(fcn_idx);
- stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] =
+ stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] =
I40E_GLPES_VFRDMAVINVLO(fcn_idx);
}
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64;
- stat_index++) {
- stat_reg_offset = stat_table->stat_offset_64[stat_index];
- last_rd_stats->stat_value_64[stat_index] =
- readq(devstat->hw->hw_addr + stat_reg_offset);
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64;
+ stats_index++) {
+ stats_reg_offset = stats_table->stats_offset_64[stats_index];
+ last_rd_stats->stats_value_64[stats_index] =
+ readq(stats->hw->hw_addr + stats_reg_offset);
}
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32;
- stat_index++) {
- stat_reg_offset = stat_table->stat_offset_32[stat_index];
- last_rd_stats->stat_value_32[stat_index] =
- i40iw_rd32(devstat->hw, stat_reg_offset);
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32;
+ stats_index++) {
+ stats_reg_offset = stats_table->stats_offset_32[stats_index];
+ last_rd_stats->stats_value_32[stats_index] =
+ i40iw_rd32(stats->hw, stats_reg_offset);
}
}
/**
- * i40iw_hw_stat_read_32 - Read 32-bit HW stat counters and accommodates for roll-overs.
- * @devstat: pestat struct
- * @index: index in HW stat table which contains offset reg-addr
- * @value: hw stat value
+ * i40iw_hw_stats_read_32 - Read 32-bit HW stats counters and accommodates for roll-overs.
+ * @stat: pestat struct
+ * @index: index in HW stats table which contains offset reg-addr
+ * @value: hw stats value
*/
-static void i40iw_hw_stat_read_32(struct i40iw_dev_pestat *devstat,
- enum i40iw_hw_stat_index_32b index,
- u64 *value)
+void i40iw_hw_stats_read_32(struct i40iw_vsi_pestat *stats,
+ enum i40iw_hw_stats_index_32b index,
+ u64 *value)
{
- struct i40iw_dev_hw_stat_offsets *stat_table =
- &devstat->hw_stat_offsets;
- struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats;
- struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats;
- u64 new_stat_value = 0;
- u32 stat_reg_offset = stat_table->stat_offset_32[index];
-
- new_stat_value = i40iw_rd32(devstat->hw, stat_reg_offset);
+ struct i40iw_dev_hw_stats_offsets *stats_table =
+ &stats->hw_stats_offsets;
+ struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats;
+ struct i40iw_dev_hw_stats *hw_stats = &stats->hw_stats;
+ u64 new_stats_value = 0;
+ u32 stats_reg_offset = stats_table->stats_offset_32[index];
+
+ new_stats_value = i40iw_rd32(stats->hw, stats_reg_offset);
/*roll-over case */
- if (new_stat_value < last_rd_stats->stat_value_32[index])
- hw_stats->stat_value_32[index] += new_stat_value;
+ if (new_stats_value < last_rd_stats->stats_value_32[index])
+ hw_stats->stats_value_32[index] += new_stats_value;
else
- hw_stats->stat_value_32[index] +=
- new_stat_value - last_rd_stats->stat_value_32[index];
- last_rd_stats->stat_value_32[index] = new_stat_value;
- *value = hw_stats->stat_value_32[index];
+ hw_stats->stats_value_32[index] +=
+ new_stats_value - last_rd_stats->stats_value_32[index];
+ last_rd_stats->stats_value_32[index] = new_stats_value;
+ *value = hw_stats->stats_value_32[index];
}
/**
- * i40iw_hw_stat_read_64 - Read HW stat counters (greater than 32-bit) and accommodates for roll-overs.
- * @devstat: pestat struct
- * @index: index in HW stat table which contains offset reg-addr
- * @value: hw stat value
+ * i40iw_hw_stats_read_64 - Read HW stats counters (greater than 32-bit) and accommodates for roll-overs.
+ * @stats: pestat struct
+ * @index: index in HW stats table which contains offset reg-addr
+ * @value: hw stats value
*/
-static void i40iw_hw_stat_read_64(struct i40iw_dev_pestat *devstat,
- enum i40iw_hw_stat_index_64b index,
- u64 *value)
+void i40iw_hw_stats_read_64(struct i40iw_vsi_pestat *stats,
+ enum i40iw_hw_stats_index_64b index,
+ u64 *value)
{
- struct i40iw_dev_hw_stat_offsets *stat_table =
- &devstat->hw_stat_offsets;
- struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats;
- struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats;
- u64 new_stat_value = 0;
- u32 stat_reg_offset = stat_table->stat_offset_64[index];
-
- new_stat_value = readq(devstat->hw->hw_addr + stat_reg_offset);
+ struct i40iw_dev_hw_stats_offsets *stats_table =
+ &stats->hw_stats_offsets;
+ struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats;
+ struct i40iw_dev_hw_stats *hw_stats = &stats->hw_stats;
+ u64 new_stats_value = 0;
+ u32 stats_reg_offset = stats_table->stats_offset_64[index];
+
+ new_stats_value = readq(stats->hw->hw_addr + stats_reg_offset);
/*roll-over case */
- if (new_stat_value < last_rd_stats->stat_value_64[index])
- hw_stats->stat_value_64[index] += new_stat_value;
+ if (new_stats_value < last_rd_stats->stats_value_64[index])
+ hw_stats->stats_value_64[index] += new_stats_value;
else
- hw_stats->stat_value_64[index] +=
- new_stat_value - last_rd_stats->stat_value_64[index];
- last_rd_stats->stat_value_64[index] = new_stat_value;
- *value = hw_stats->stat_value_64[index];
+ hw_stats->stats_value_64[index] +=
+ new_stats_value - last_rd_stats->stats_value_64[index];
+ last_rd_stats->stats_value_64[index] = new_stats_value;
+ *value = hw_stats->stats_value_64[index];
}
/**
- * i40iw_hw_stat_read_all - read all HW stat counters
- * @devstat: pestat struct
- * @stat_values: hw stats structure
+ * i40iw_hw_stats_read_all - read all HW stat counters
+ * @stats: pestat struct
+ * @stats_values: hw stats structure
*
* Read all the HW stat counters and populates hw_stats structure
- * of passed-in dev's pestat as well as copy created in stat_values.
+ * of passed-in vsi's pestat as well as copy created in stat_values.
*/
-static void i40iw_hw_stat_read_all(struct i40iw_dev_pestat *devstat,
- struct i40iw_dev_hw_stats *stat_values)
+void i40iw_hw_stats_read_all(struct i40iw_vsi_pestat *stats,
+ struct i40iw_dev_hw_stats *stats_values)
{
- u32 stat_index;
-
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32;
- stat_index++)
- i40iw_hw_stat_read_32(devstat, stat_index,
- &stat_values->stat_value_32[stat_index]);
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64;
- stat_index++)
- i40iw_hw_stat_read_64(devstat, stat_index,
- &stat_values->stat_value_64[stat_index]);
+ u32 stats_index;
+ unsigned long flags;
+
+ spin_lock_irqsave(&stats->lock, flags);
+
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32;
+ stats_index++)
+ i40iw_hw_stats_read_32(stats, stats_index,
+ &stats_values->stats_value_32[stats_index]);
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64;
+ stats_index++)
+ i40iw_hw_stats_read_64(stats, stats_index,
+ &stats_values->stats_value_64[stats_index]);
+ spin_unlock_irqrestore(&stats->lock, flags);
}
/**
- * i40iw_hw_stat_refresh_all - Update all HW stat structs
- * @devstat: pestat struct
- * @stat_values: hw stats structure
+ * i40iw_hw_stats_refresh_all - Update all HW stats structs
+ * @stats: pestat struct
*
- * Read all the HW stat counters to refresh values in hw_stats structure
+ * Read all the HW stats counters to refresh values in hw_stats structure
* of passed-in dev's pestat
*/
-static void i40iw_hw_stat_refresh_all(struct i40iw_dev_pestat *devstat)
+void i40iw_hw_stats_refresh_all(struct i40iw_vsi_pestat *stats)
+{
+ u64 stats_value;
+ u32 stats_index;
+ unsigned long flags;
+
+ spin_lock_irqsave(&stats->lock, flags);
+
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32;
+ stats_index++)
+ i40iw_hw_stats_read_32(stats, stats_index, &stats_value);
+ for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64;
+ stats_index++)
+ i40iw_hw_stats_read_64(stats, stats_index, &stats_value);
+ spin_unlock_irqrestore(&stats->lock, flags);
+}
+
+/**
+ * i40iw_get_fcn_id - Return the function id
+ * @dev: pointer to the device
+ */
+static u8 i40iw_get_fcn_id(struct i40iw_sc_dev *dev)
+{
+ u8 fcn_id = I40IW_INVALID_FCN_ID;
+ u8 i;
+
+ for (i = I40IW_FIRST_NON_PF_STAT; i < I40IW_MAX_STATS_COUNT; i++)
+ if (!dev->fcn_id_array[i]) {
+ fcn_id = i;
+ dev->fcn_id_array[i] = true;
+ break;
+ }
+ return fcn_id;
+}
+
+/**
+ * i40iw_vsi_stats_init - Initialize the vsi statistics
+ * @vsi: pointer to the vsi structure
+ * @info: The info structure used for initialization
+ */
+enum i40iw_status_code i40iw_vsi_stats_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_stats_info *info)
{
- u64 stat_value;
- u32 stat_index;
-
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32;
- stat_index++)
- i40iw_hw_stat_read_32(devstat, stat_index, &stat_value);
- for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64;
- stat_index++)
- i40iw_hw_stat_read_64(devstat, stat_index, &stat_value);
+ u8 fcn_id = info->fcn_id;
+
+ if (info->alloc_fcn_id)
+ fcn_id = i40iw_get_fcn_id(vsi->dev);
+
+ if (fcn_id == I40IW_INVALID_FCN_ID)
+ return I40IW_ERR_NOT_READY;
+
+ vsi->pestat = info->pestat;
+ vsi->pestat->hw = vsi->dev->hw;
+
+ if (info->stats_initialize) {
+ i40iw_hw_stats_init(vsi->pestat, fcn_id, true);
+ spin_lock_init(&vsi->pestat->lock);
+ i40iw_hw_stats_start_timer(vsi);
+ }
+ vsi->stats_fcn_id_alloc = info->alloc_fcn_id;
+ vsi->fcn_id = fcn_id;
+ return I40IW_SUCCESS;
+}
+
+/**
+ * i40iw_vsi_stats_free - Free the vsi stats
+ * @vsi: pointer to the vsi structure
+ */
+void i40iw_vsi_stats_free(struct i40iw_sc_vsi *vsi)
+{
+ u8 fcn_id = vsi->fcn_id;
+
+ if ((vsi->stats_fcn_id_alloc) && (fcn_id != I40IW_INVALID_FCN_ID))
+ vsi->dev->fcn_id_array[fcn_id] = false;
+ i40iw_hw_stats_stop_timer(vsi);
}
static struct i40iw_cqp_ops iw_cqp_ops = {
@@ -4837,23 +4943,6 @@ static struct i40iw_hmc_ops iw_hmc_ops = {
NULL
};
-static const struct i40iw_device_pestat_ops iw_device_pestat_ops = {
- i40iw_hw_stat_init,
- i40iw_hw_stat_read_32,
- i40iw_hw_stat_read_64,
- i40iw_hw_stat_read_all,
- i40iw_hw_stat_refresh_all
-};
-
-/**
- * i40iw_device_init_pestat - Initialize the pestat structure
- * @dev: pestat struct
- */
-void i40iw_device_init_pestat(struct i40iw_dev_pestat *devstat)
-{
- devstat->ops = iw_device_pestat_ops;
-}
-
/**
* i40iw_device_init - Initialize IWARP device
* @dev: IWARP device pointer
@@ -4867,7 +4956,6 @@ enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev,
u16 hmc_fcn = 0;
enum i40iw_status_code ret_code = 0;
u8 db_size;
- int i;
spin_lock_init(&dev->cqp_lock);
INIT_LIST_HEAD(&dev->cqp_cmd_head); /* for the cqp commands backlog. */
@@ -4876,15 +4964,7 @@ enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev,
dev->debug_mask = info->debug_mask;
- i40iw_device_init_pestat(&dev->dev_pestat);
dev->hmc_fn_id = info->hmc_fn_id;
- i40iw_fill_qos_list(info->l2params.qs_handle_list);
- for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) {
- dev->qos[i].qs_handle = info->l2params.qs_handle_list[i];
- i40iw_debug(dev, I40IW_DEBUG_DCB, "qset[%d]: %d\n", i, dev->qos[i].qs_handle);
- spin_lock_init(&dev->qos[i].lock);
- INIT_LIST_HEAD(&dev->qos[i].qplist);
- }
dev->exception_lan_queue = info->exception_lan_queue;
dev->is_pf = info->is_pf;
@@ -4897,15 +4977,10 @@ enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev,
dev->hw = info->hw;
dev->hw->hw_addr = info->bar0;
- val = i40iw_rd32(dev->hw, I40E_GLPCI_DREVID);
- dev->hw_rev = (u8)RS_32(val, I40E_GLPCI_DREVID_DEFAULT_REVID);
-
if (dev->is_pf) {
- dev->dev_pestat.ops.iw_hw_stat_init(&dev->dev_pestat,
- dev->hmc_fn_id, dev->hw, true);
- spin_lock_init(&dev->dev_pestat.stats_lock);
- /*start the periodic stats_timer */
- i40iw_hw_stats_start_timer(dev);
+ val = i40iw_rd32(dev->hw, I40E_GLPCI_DREVID);
+ dev->hw_rev = (u8)RS_32(val, I40E_GLPCI_DREVID_DEFAULT_REVID);
+
val = i40iw_rd32(dev->hw, I40E_GLPCI_LBARCTRL);
db_size = (u8)RS_32(val, I40E_GLPCI_LBARCTRL_PE_DB_SIZE);
if ((db_size != I40IW_PE_DB_SIZE_4M) &&
diff --git a/drivers/infiniband/hw/i40iw/i40iw_d.h b/drivers/infiniband/hw/i40iw/i40iw_d.h
index 1bd4bad..a39ac12 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_d.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_d.h
@@ -69,6 +69,9 @@
#define I40IW_STAG_TYPE_NONSHARED 1
#define I40IW_MAX_USER_PRIORITY 8
+#define I40IW_MAX_STATS_COUNT 16
+#define I40IW_FIRST_NON_PF_STAT 4
+
#define LS_64_1(val, bits) ((u64)(uintptr_t)val << bits)
#define RS_64_1(val, bits) ((u64)(uintptr_t)val >> bits)
@@ -1203,8 +1206,11 @@
#define I40IWQPC_RXCQNUM_SHIFT 32
#define I40IWQPC_RXCQNUM_MASK (0x1ffffULL << I40IWQPC_RXCQNUM_SHIFT)
-#define I40IWQPC_Q2ADDR_SHIFT I40IW_CQPHC_QPCTX_SHIFT
-#define I40IWQPC_Q2ADDR_MASK I40IW_CQPHC_QPCTX_MASK
+#define I40IWQPC_STAT_INDEX_SHIFT 0
+#define I40IWQPC_STAT_INDEX_MASK (0x1fULL << I40IWQPC_STAT_INDEX_SHIFT)
+
+#define I40IWQPC_Q2ADDR_SHIFT 0
+#define I40IWQPC_Q2ADDR_MASK (0xffffffffffffff00ULL << I40IWQPC_Q2ADDR_SHIFT)
#define I40IWQPC_LASTBYTESENT_SHIFT 0
#define I40IWQPC_LASTBYTESENT_MASK (0xffUL << I40IWQPC_LASTBYTESENT_SHIFT)
@@ -1236,11 +1242,8 @@
#define I40IWQPC_PRIVEN_SHIFT 25
#define I40IWQPC_PRIVEN_MASK (1UL << I40IWQPC_PRIVEN_SHIFT)
-#define I40IWQPC_LSMMPRESENT_SHIFT 26
-#define I40IWQPC_LSMMPRESENT_MASK (1UL << I40IWQPC_LSMMPRESENT_SHIFT)
-
-#define I40IWQPC_ADJUSTFORLSMM_SHIFT 27
-#define I40IWQPC_ADJUSTFORLSMM_MASK (1UL << I40IWQPC_ADJUSTFORLSMM_SHIFT)
+#define I40IWQPC_USESTATSINSTANCE_SHIFT 26
+#define I40IWQPC_USESTATSINSTANCE_MASK (1UL << I40IWQPC_USESTATSINSTANCE_SHIFT)
#define I40IWQPC_IWARPMODE_SHIFT 28
#define I40IWQPC_IWARPMODE_MASK (1UL << I40IWQPC_IWARPMODE_SHIFT)
@@ -1717,6 +1720,8 @@ enum i40iw_alignment {
#define OP_MANAGE_VF_PBLE_BP 28
#define OP_QUERY_FPM_VALUES 29
#define OP_COMMIT_FPM_VALUES 30
-#define OP_SIZE_CQP_STAT_ARRAY 31
+#define OP_REQUESTED_COMMANDS 31
+#define OP_COMPLETED_COMMANDS 32
+#define OP_SIZE_CQP_STAT_ARRAY 33
#endif
diff --git a/drivers/infiniband/hw/i40iw/i40iw_hw.c b/drivers/infiniband/hw/i40iw/i40iw_hw.c
index 4394a67..476867a 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_hw.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_hw.c
@@ -542,6 +542,7 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev,
{
struct i40iw_qhash_table_info *info;
struct i40iw_sc_dev *dev = &iwdev->sc_dev;
+ struct i40iw_sc_vsi *vsi = &iwdev->vsi;
enum i40iw_status_code status;
struct i40iw_cqp *iwcqp = &iwdev->cqp;
struct i40iw_cqp_request *cqp_request;
@@ -554,6 +555,7 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev,
info = &cqp_info->in.u.manage_qhash_table_entry.info;
memset(info, 0, sizeof(*info));
+ info->vsi = &iwdev->vsi;
info->manage = mtype;
info->entry_type = etype;
if (cminfo->vlan_id != 0xFFFF) {
@@ -566,7 +568,7 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev,
info->ipv4_valid = cminfo->ipv4;
info->user_pri = cminfo->user_pri;
ether_addr_copy(info->mac_addr, iwdev->netdev->dev_addr);
- info->qp_num = cpu_to_le32(dev->ilq->qp_id);
+ info->qp_num = cpu_to_le32(vsi->ilq->qp_id);
info->dest_port = cpu_to_le16(cminfo->loc_port);
info->dest_ip[0] = cpu_to_le32(cminfo->loc_addr[0]);
info->dest_ip[1] = cpu_to_le32(cminfo->loc_addr[1]);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c
index cf9d288..2bdb8b0 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_main.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_main.c
@@ -932,6 +932,7 @@ static enum i40iw_status_code i40iw_initialize_ilq(struct i40iw_device *iwdev)
struct i40iw_puda_rsrc_info info;
enum i40iw_status_code status;
+ memset(&info, 0, sizeof(info));
info.type = I40IW_PUDA_RSRC_TYPE_ILQ;
info.cq_id = 1;
info.qp_id = 0;
@@ -941,10 +942,9 @@ static enum i40iw_status_code i40iw_initialize_ilq(struct i40iw_device *iwdev)
info.rq_size = 8192;
info.buf_size = 1024;
info.tx_buf_cnt = 16384;
- info.mss = iwdev->sc_dev.mss;
info.receive = i40iw_receive_ilq;
info.xmit_complete = i40iw_free_sqbuf;
- status = i40iw_puda_create_rsrc(&iwdev->sc_dev, &info);
+ status = i40iw_puda_create_rsrc(&iwdev->vsi, &info);
if (status)
i40iw_pr_err("ilq create fail\n");
return status;
@@ -961,6 +961,7 @@ static enum i40iw_status_code i40iw_initialize_ieq(struct i40iw_device *iwdev)
struct i40iw_puda_rsrc_info info;
enum i40iw_status_code status;
+ memset(&info, 0, sizeof(info));
info.type = I40IW_PUDA_RSRC_TYPE_IEQ;
info.cq_id = 2;
info.qp_id = iwdev->sc_dev.exception_lan_queue;
@@ -969,9 +970,8 @@ static enum i40iw_status_code i40iw_initialize_ieq(struct i40iw_device *iwdev)
info.sq_size = 8192;
info.rq_size = 8192;
info.buf_size = 2048;
- info.mss = iwdev->sc_dev.mss;
info.tx_buf_cnt = 16384;
- status = i40iw_puda_create_rsrc(&iwdev->sc_dev, &info);
+ status = i40iw_puda_create_rsrc(&iwdev->vsi, &info);
if (status)
i40iw_pr_err("ieq create fail\n");
return status;
@@ -1296,12 +1296,16 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev,
enum i40iw_status_code status;
struct i40iw_sc_dev *dev = &iwdev->sc_dev;
struct i40iw_device_init_info info;
+ struct i40iw_vsi_init_info vsi_info;
struct i40iw_dma_mem mem;
+ struct i40iw_l2params l2params;
u32 size;
+ struct i40iw_vsi_stats_info stats_info;
u16 last_qset = I40IW_NO_QSET;
u16 qset;
u32 i;
+ memset(&l2params, 0, sizeof(l2params));
memset(&info, 0, sizeof(info));
size = sizeof(struct i40iw_hmc_pble_rsrc) + sizeof(struct i40iw_hmc_info) +
(sizeof(struct i40iw_hmc_obj_info) * I40IW_HMC_IW_MAX);
@@ -1330,16 +1334,17 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev,
info.bar0 = ldev->hw_addr;
info.hw = &iwdev->hw;
info.debug_mask = debug;
- info.l2params.mss =
+ l2params.mss =
(ldev->params.mtu) ? ldev->params.mtu - I40IW_MTU_TO_MSS : I40IW_DEFAULT_MSS;
for (i = 0; i < I40E_CLIENT_MAX_USER_PRIORITY; i++) {
qset = ldev->params.qos.prio_qos[i].qs_handle;
- info.l2params.qs_handle_list[i] = qset;
+ l2params.qs_handle_list[i] = qset;
if (last_qset == I40IW_NO_QSET)
last_qset = qset;
else if ((qset != last_qset) && (qset != I40IW_NO_QSET))
iwdev->dcb = true;
}
+ i40iw_pr_info("DCB is set/clear = %d\n", iwdev->dcb);
info.exception_lan_queue = 1;
info.vchnl_send = i40iw_virtchnl_send;
status = i40iw_device_init(&iwdev->sc_dev, &info);
@@ -1348,6 +1353,20 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev,
kfree(iwdev->hmc_info_mem);
iwdev->hmc_info_mem = NULL;
}
+ memset(&vsi_info, 0, sizeof(vsi_info));
+ vsi_info.dev = &iwdev->sc_dev;
+ vsi_info.back_vsi = (void *)iwdev;
+ vsi_info.params = &l2params;
+ i40iw_sc_vsi_init(&iwdev->vsi, &vsi_info);
+
+ if (dev->is_pf) {
+ memset(&stats_info, 0, sizeof(stats_info));
+ stats_info.fcn_id = ldev->fid;
+ stats_info.pestat = kzalloc(sizeof(*stats_info.pestat), GFP_KERNEL);
+ stats_info.stats_initialize = true;
+ if (stats_info.pestat)
+ i40iw_vsi_stats_init(&iwdev->vsi, &stats_info);
+ }
return status;
}
@@ -1457,10 +1476,10 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset)
i40iw_destroy_aeq(iwdev, reset);
/* fallthrough */
case IEQ_CREATED:
- i40iw_puda_dele_resources(dev, I40IW_PUDA_RSRC_TYPE_IEQ, reset);
+ i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_IEQ, reset);
/* fallthrough */
case ILQ_CREATED:
- i40iw_puda_dele_resources(dev, I40IW_PUDA_RSRC_TYPE_ILQ, reset);
+ i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_ILQ, reset);
/* fallthrough */
case CCQ_CREATED:
i40iw_destroy_ccq(iwdev, reset);
@@ -1476,9 +1495,10 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset)
/* fallthrough */
case INITIAL_STATE:
i40iw_cleanup_cm_core(&iwdev->cm_core);
- if (dev->is_pf)
- i40iw_hw_stats_del_timer(dev);
-
+ if (iwdev->vsi.pestat) {
+ i40iw_vsi_stats_free(&iwdev->vsi);
+ kfree(iwdev->vsi.pestat);
+ }
i40iw_del_init_mem(iwdev);
break;
case INVALID_STATE:
@@ -1523,7 +1543,6 @@ static enum i40iw_status_code i40iw_setup_init_state(struct i40iw_handler *hdl,
iwdev->max_enabled_vfs = iwdev->max_rdma_vfs;
iwdev->netdev = ldev->netdev;
hdl->client = client;
- iwdev->mss = (!ldev->params.mtu) ? I40IW_DEFAULT_MSS : ldev->params.mtu - I40IW_MTU_TO_MSS;
if (!ldev->ftype)
iwdev->db_start = pci_resource_start(ldev->pcidev, 0) + I40IW_DB_ADDR_OFFSET;
else
@@ -1683,7 +1702,7 @@ static void i40iw_l2params_worker(struct work_struct *work)
container_of(work, struct l2params_work, work);
struct i40iw_device *iwdev = dwork->iwdev;
- i40iw_change_l2params(&iwdev->sc_dev, &dwork->l2params);
+ i40iw_change_l2params(&iwdev->vsi, &dwork->l2params);
atomic_dec(&iwdev->params_busy);
kfree(work);
}
@@ -1724,7 +1743,7 @@ static void i40iw_l2param_change(struct i40e_info *ldev, struct i40e_client *cli
for (i = 0; i < I40E_CLIENT_MAX_USER_PRIORITY; i++)
l2params->qs_handle_list[i] = params->qos.prio_qos[i].qs_handle;
- l2params->mss = (params->mtu) ? params->mtu - I40IW_MTU_TO_MSS : iwdev->mss;
+ l2params->mss = (params->mtu) ? params->mtu - I40IW_MTU_TO_MSS : iwdev->vsi.mss;
INIT_WORK(&work->work, i40iw_l2params_worker);
queue_work(iwdev->param_wq, &work->work);
@@ -1773,21 +1792,23 @@ static void i40iw_vf_reset(struct i40e_info *ldev, struct i40e_client *client, u
struct i40iw_vfdev *tmp_vfdev;
unsigned int i;
unsigned long flags;
+ struct i40iw_device *iwdev;
hdl = i40iw_find_i40e_handler(ldev);
if (!hdl)
return;
dev = &hdl->device.sc_dev;
+ iwdev = (struct i40iw_device *)dev->back_dev;
for (i = 0; i < I40IW_MAX_PE_ENABLED_VF_COUNT; i++) {
if (!dev->vf_dev[i] || (dev->vf_dev[i]->vf_id != vf_id))
continue;
/* free all resources allocated on behalf of vf */
tmp_vfdev = dev->vf_dev[i];
- spin_lock_irqsave(&dev->dev_pestat.stats_lock, flags);
+ spin_lock_irqsave(&iwdev->vsi.pestat->lock, flags);
dev->vf_dev[i] = NULL;
- spin_unlock_irqrestore(&dev->dev_pestat.stats_lock, flags);
+ spin_unlock_irqrestore(&iwdev->vsi.pestat->lock, flags);
i40iw_del_hmc_objects(dev, &tmp_vfdev->hmc_info, false, false);
/* remove vf hmc function */
memset(&hmc_fcn_info, 0, sizeof(hmc_fcn_info));
diff --git a/drivers/infiniband/hw/i40iw/i40iw_osdep.h b/drivers/infiniband/hw/i40iw/i40iw_osdep.h
index a6b18cd..aa66c1c 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_osdep.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_osdep.h
@@ -209,9 +209,9 @@ void i40iw_terminate_del_timer(struct i40iw_sc_qp *qp);
enum i40iw_status_code i40iw_hw_manage_vf_pble_bp(struct i40iw_device *iwdev,
struct i40iw_manage_vf_pble_info *info,
bool wait);
-struct i40iw_dev_pestat;
-void i40iw_hw_stats_start_timer(struct i40iw_sc_dev *);
-void i40iw_hw_stats_del_timer(struct i40iw_sc_dev *);
+struct i40iw_sc_vsi;
+void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi);
+void i40iw_hw_stats_stop_timer(struct i40iw_sc_vsi *vsi);
#define i40iw_mmiowb() mmiowb()
void i40iw_wr32(struct i40iw_hw *hw, u32 reg, u32 value);
u32 i40iw_rd32(struct i40iw_hw *hw, u32 reg);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_p.h b/drivers/infiniband/hw/i40iw/i40iw_p.h
index 2a4bd32..28a92fe 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_p.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_p.h
@@ -47,8 +47,6 @@ void i40iw_debug_buf(struct i40iw_sc_dev *dev, enum i40iw_debug_flag mask,
enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev,
struct i40iw_device_init_info *info);
-void i40iw_device_init_pestat(struct i40iw_dev_pestat *devstat);
-
void i40iw_sc_cqp_post_sq(struct i40iw_sc_cqp *cqp);
u64 *i40iw_sc_cqp_get_next_send_wqe(struct i40iw_sc_cqp *cqp, u64 scratch);
@@ -64,9 +62,24 @@ enum i40iw_status_code i40iw_sc_init_iw_hmc(struct i40iw_sc_dev *dev,
enum i40iw_status_code i40iw_pf_init_vfhmc(struct i40iw_sc_dev *dev, u8 vf_hmc_fn_id,
u32 *vf_cnt_array);
-/* cqp misc functions */
-void i40iw_change_l2params(struct i40iw_sc_dev *dev, struct i40iw_l2params *l2params);
-void i40iw_qp_add_qos(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp);
+/* stats functions */
+void i40iw_hw_stats_refresh_all(struct i40iw_vsi_pestat *stats);
+void i40iw_hw_stats_read_all(struct i40iw_vsi_pestat *stats, struct i40iw_dev_hw_stats *stats_values);
+void i40iw_hw_stats_read_32(struct i40iw_vsi_pestat *stats,
+ enum i40iw_hw_stats_index_32b index,
+ u64 *value);
+void i40iw_hw_stats_read_64(struct i40iw_vsi_pestat *stats,
+ enum i40iw_hw_stats_index_64b index,
+ u64 *value);
+void i40iw_hw_stats_init(struct i40iw_vsi_pestat *stats, u8 index, bool is_pf);
+
+/* vsi misc functions */
+enum i40iw_status_code i40iw_vsi_stats_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_stats_info *info);
+void i40iw_vsi_stats_free(struct i40iw_sc_vsi *vsi);
+void i40iw_sc_vsi_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_init_info *info);
+
+void i40iw_change_l2params(struct i40iw_sc_vsi *vsi, struct i40iw_l2params *l2params);
+void i40iw_qp_add_qos(struct i40iw_sc_qp *qp);
void i40iw_terminate_send_fin(struct i40iw_sc_qp *qp);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.c b/drivers/infiniband/hw/i40iw/i40iw_puda.c
index c3d28ba..449ba8c 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_puda.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_puda.c
@@ -42,12 +42,13 @@
#include "i40iw_p.h"
#include "i40iw_puda.h"
-static void i40iw_ieq_receive(struct i40iw_sc_dev *dev,
+static void i40iw_ieq_receive(struct i40iw_sc_vsi *vsi,
struct i40iw_puda_buf *buf);
-static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid);
+static void i40iw_ieq_tx_compl(struct i40iw_sc_vsi *vsi, void *sqwrid);
static void i40iw_ilq_putback_rcvbuf(struct i40iw_sc_qp *qp, u32 wqe_idx);
static enum i40iw_status_code i40iw_puda_replenish_rq(struct i40iw_puda_rsrc
*rsrc, bool initial);
+static void i40iw_ieq_cleanup_qp(struct i40iw_puda_rsrc *ieq, struct i40iw_sc_qp *qp);
/**
* i40iw_puda_get_listbuf - get buffer from puda list
* @list: list to use for buffers (ILQ or IEQ)
@@ -292,7 +293,7 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev,
unsigned long flags;
if ((cq_type == I40IW_CQ_TYPE_ILQ) || (cq_type == I40IW_CQ_TYPE_IEQ)) {
- rsrc = (cq_type == I40IW_CQ_TYPE_ILQ) ? dev->ilq : dev->ieq;
+ rsrc = (cq_type == I40IW_CQ_TYPE_ILQ) ? cq->vsi->ilq : cq->vsi->ieq;
} else {
i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s qp_type error\n", __func__);
return I40IW_ERR_BAD_PTR;
@@ -335,7 +336,7 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev,
rsrc->stats_pkt_rcvd++;
rsrc->compl_rxwqe_idx = info.wqe_idx;
i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s RQ completion\n", __func__);
- rsrc->receive(rsrc->dev, buf);
+ rsrc->receive(rsrc->vsi, buf);
if (cq_type == I40IW_CQ_TYPE_ILQ)
i40iw_ilq_putback_rcvbuf(&rsrc->qp, info.wqe_idx);
else
@@ -345,12 +346,12 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev,
i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s SQ completion\n", __func__);
sqwrid = (void *)(uintptr_t)qp->sq_wrtrk_array[info.wqe_idx].wrid;
I40IW_RING_SET_TAIL(qp->sq_ring, info.wqe_idx);
- rsrc->xmit_complete(rsrc->dev, sqwrid);
+ rsrc->xmit_complete(rsrc->vsi, sqwrid);
spin_lock_irqsave(&rsrc->bufpool_lock, flags);
rsrc->tx_wqe_avail_cnt++;
spin_unlock_irqrestore(&rsrc->bufpool_lock, flags);
- if (!list_empty(&dev->ilq->txpend))
- i40iw_puda_send_buf(dev->ilq, NULL);
+ if (!list_empty(&rsrc->vsi->ilq->txpend))
+ i40iw_puda_send_buf(rsrc->vsi->ilq, NULL);
}
done:
@@ -513,10 +514,8 @@ static void i40iw_puda_qp_setctx(struct i40iw_puda_rsrc *rsrc)
* i40iw_puda_qp_wqe - setup wqe for qp create
* @rsrc: resource for qp
*/
-static enum i40iw_status_code i40iw_puda_qp_wqe(struct i40iw_puda_rsrc *rsrc)
+static enum i40iw_status_code i40iw_puda_qp_wqe(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
{
- struct i40iw_sc_qp *qp = &rsrc->qp;
- struct i40iw_sc_dev *dev = rsrc->dev;
struct i40iw_sc_cqp *cqp;
u64 *wqe;
u64 header;
@@ -582,6 +581,7 @@ static enum i40iw_status_code i40iw_puda_qp_create(struct i40iw_puda_rsrc *rsrc)
qp->back_qp = (void *)rsrc;
qp->sq_pa = mem->pa;
qp->rq_pa = qp->sq_pa + sq_size;
+ qp->vsi = rsrc->vsi;
ukqp->sq_base = mem->va;
ukqp->rq_base = &ukqp->sq_base[rsrc->sq_size];
ukqp->shadow_area = ukqp->rq_base[rsrc->rq_size].elem;
@@ -609,15 +609,62 @@ static enum i40iw_status_code i40iw_puda_qp_create(struct i40iw_puda_rsrc *rsrc)
I40E_VFPE_WQEALLOC1);
qp->user_pri = 0;
- i40iw_qp_add_qos(rsrc->dev, qp);
+ i40iw_qp_add_qos(qp);
i40iw_puda_qp_setctx(rsrc);
- ret = i40iw_puda_qp_wqe(rsrc);
+ if (rsrc->ceq_valid)
+ ret = i40iw_cqp_qp_create_cmd(rsrc->dev, qp);
+ else
+ ret = i40iw_puda_qp_wqe(rsrc->dev, qp);
if (ret)
i40iw_free_dma_mem(rsrc->dev->hw, &rsrc->qpmem);
return ret;
}
/**
+ * i40iw_puda_cq_wqe - setup wqe for cq create
+ * @rsrc: resource for cq
+ */
+static enum i40iw_status_code i40iw_puda_cq_wqe(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq)
+{
+ u64 *wqe;
+ struct i40iw_sc_cqp *cqp;
+ u64 header;
+ struct i40iw_ccq_cqe_info compl_info;
+ enum i40iw_status_code status = 0;
+
+ cqp = dev->cqp;
+ wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, 0);
+ if (!wqe)
+ return I40IW_ERR_RING_FULL;
+
+ set_64bit_val(wqe, 0, cq->cq_uk.cq_size);
+ set_64bit_val(wqe, 8, RS_64_1(cq, 1));
+ set_64bit_val(wqe, 16,
+ LS_64(cq->shadow_read_threshold,
+ I40IW_CQPSQ_CQ_SHADOW_READ_THRESHOLD));
+ set_64bit_val(wqe, 32, cq->cq_pa);
+
+ set_64bit_val(wqe, 40, cq->shadow_area_pa);
+
+ header = cq->cq_uk.cq_id |
+ LS_64(I40IW_CQP_OP_CREATE_CQ, I40IW_CQPSQ_OPCODE) |
+ LS_64(1, I40IW_CQPSQ_CQ_CHKOVERFLOW) |
+ LS_64(1, I40IW_CQPSQ_CQ_ENCEQEMASK) |
+ LS_64(1, I40IW_CQPSQ_CQ_CEQIDVALID) |
+ LS_64(cqp->polarity, I40IW_CQPSQ_WQEVALID);
+ set_64bit_val(wqe, 24, header);
+
+ i40iw_debug_buf(dev, I40IW_DEBUG_PUDA, "PUDA CQE",
+ wqe, I40IW_CQP_WQE_SIZE * 8);
+
+ i40iw_sc_cqp_post_sq(dev->cqp);
+ status = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
+ I40IW_CQP_OP_CREATE_CQ,
+ &compl_info);
+ return status;
+}
+
+/**
* i40iw_puda_cq_create - create cq for resource
* @rsrc: resource for which cq to create
*/
@@ -625,18 +672,13 @@ static enum i40iw_status_code i40iw_puda_cq_create(struct i40iw_puda_rsrc *rsrc)
{
struct i40iw_sc_dev *dev = rsrc->dev;
struct i40iw_sc_cq *cq = &rsrc->cq;
- u64 *wqe;
- struct i40iw_sc_cqp *cqp;
- u64 header;
enum i40iw_status_code ret = 0;
u32 tsize, cqsize;
- u32 shadow_read_threshold = 128;
struct i40iw_dma_mem *mem;
- struct i40iw_ccq_cqe_info compl_info;
struct i40iw_cq_init_info info;
struct i40iw_cq_uk_init_info *init_info = &info.cq_uk_init_info;
- cq->back_cq = (void *)rsrc;
+ cq->vsi = rsrc->vsi;
cqsize = rsrc->cq_size * (sizeof(struct i40iw_cqe));
tsize = cqsize + sizeof(struct i40iw_cq_shadow_area);
ret = i40iw_allocate_dma_mem(dev->hw, &rsrc->cqmem, tsize,
@@ -657,43 +699,84 @@ static enum i40iw_status_code i40iw_puda_cq_create(struct i40iw_puda_rsrc *rsrc)
init_info->shadow_area = (u64 *)((u8 *)mem->va + cqsize);
init_info->cq_size = rsrc->cq_size;
init_info->cq_id = rsrc->cq_id;
+ info.ceqe_mask = true;
+ info.ceq_id_valid = true;
ret = dev->iw_priv_cq_ops->cq_init(cq, &info);
if (ret)
goto error;
- cqp = dev->cqp;
- wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, 0);
- if (!wqe) {
- ret = I40IW_ERR_RING_FULL;
- goto error;
- }
+ if (rsrc->ceq_valid)
+ ret = i40iw_cqp_cq_create_cmd(dev, cq);
+ else
+ ret = i40iw_puda_cq_wqe(dev, cq);
+error:
+ if (ret)
+ i40iw_free_dma_mem(dev->hw, &rsrc->cqmem);
+ return ret;
+}
- set_64bit_val(wqe, 0, rsrc->cq_size);
- set_64bit_val(wqe, 8, RS_64_1(cq, 1));
- set_64bit_val(wqe, 16, LS_64(shadow_read_threshold, I40IW_CQPSQ_CQ_SHADOW_READ_THRESHOLD));
- set_64bit_val(wqe, 32, cq->cq_pa);
+/**
+ * i40iw_puda_free_qp - free qp for resource
+ * @rsrc: resource for which qp to free
+ */
+static void i40iw_puda_free_qp(struct i40iw_puda_rsrc *rsrc)
+{
+ enum i40iw_status_code ret;
+ struct i40iw_ccq_cqe_info compl_info;
+ struct i40iw_sc_dev *dev = rsrc->dev;
- set_64bit_val(wqe, 40, cq->shadow_area_pa);
+ if (rsrc->ceq_valid) {
+ i40iw_cqp_qp_destroy_cmd(dev, &rsrc->qp);
+ return;
+ }
- header = rsrc->cq_id |
- LS_64(I40IW_CQP_OP_CREATE_CQ, I40IW_CQPSQ_OPCODE) |
- LS_64(1, I40IW_CQPSQ_CQ_CHKOVERFLOW) |
- LS_64(1, I40IW_CQPSQ_CQ_ENCEQEMASK) |
- LS_64(1, I40IW_CQPSQ_CQ_CEQIDVALID) |
- LS_64(cqp->polarity, I40IW_CQPSQ_WQEVALID);
- set_64bit_val(wqe, 24, header);
+ ret = dev->iw_priv_qp_ops->qp_destroy(&rsrc->qp,
+ 0, false, true, true);
+ if (ret)
+ i40iw_debug(dev, I40IW_DEBUG_PUDA,
+ "%s error puda qp destroy wqe\n",
+ __func__);
- i40iw_debug_buf(dev, I40IW_DEBUG_PUDA, "PUDA CQE",
- wqe, I40IW_CQP_WQE_SIZE * 8);
+ if (!ret) {
+ ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
+ I40IW_CQP_OP_DESTROY_QP,
+ &compl_info);
+ if (ret)
+ i40iw_debug(dev, I40IW_DEBUG_PUDA,
+ "%s error puda qp destroy failed\n",
+ __func__);
+ }
+}
- i40iw_sc_cqp_post_sq(dev->cqp);
- ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
- I40IW_CQP_OP_CREATE_CQ,
- &compl_info);
+/**
+ * i40iw_puda_free_cq - free cq for resource
+ * @rsrc: resource for which cq to free
+ */
+static void i40iw_puda_free_cq(struct i40iw_puda_rsrc *rsrc)
+{
+ enum i40iw_status_code ret;
+ struct i40iw_ccq_cqe_info compl_info;
+ struct i40iw_sc_dev *dev = rsrc->dev;
+
+ if (rsrc->ceq_valid) {
+ i40iw_cqp_cq_destroy_cmd(dev, &rsrc->cq);
+ return;
+ }
+ ret = dev->iw_priv_cq_ops->cq_destroy(&rsrc->cq, 0, true);
-error:
if (ret)
- i40iw_free_dma_mem(dev->hw, &rsrc->cqmem);
- return ret;
+ i40iw_debug(dev, I40IW_DEBUG_PUDA,
+ "%s error ieq cq destroy\n",
+ __func__);
+
+ if (!ret) {
+ ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
+ I40IW_CQP_OP_DESTROY_CQ,
+ &compl_info);
+ if (ret)
+ i40iw_debug(dev, I40IW_DEBUG_PUDA,
+ "%s error ieq qp destroy done\n",
+ __func__);
+ }
}
/**
@@ -702,25 +785,24 @@ static enum i40iw_status_code i40iw_puda_cq_create(struct i40iw_puda_rsrc *rsrc)
* @type: type of resource to dele
* @reset: true if reset chip
*/
-void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev,
+void i40iw_puda_dele_resources(struct i40iw_sc_vsi *vsi,
enum puda_resource_type type,
bool reset)
{
- struct i40iw_ccq_cqe_info compl_info;
+ struct i40iw_sc_dev *dev = vsi->dev;
struct i40iw_puda_rsrc *rsrc;
struct i40iw_puda_buf *buf = NULL;
struct i40iw_puda_buf *nextbuf = NULL;
struct i40iw_virt_mem *vmem;
- enum i40iw_status_code ret;
switch (type) {
case I40IW_PUDA_RSRC_TYPE_ILQ:
- rsrc = dev->ilq;
- vmem = &dev->ilq_mem;
+ rsrc = vsi->ilq;
+ vmem = &vsi->ilq_mem;
break;
case I40IW_PUDA_RSRC_TYPE_IEQ:
- rsrc = dev->ieq;
- vmem = &dev->ieq_mem;
+ rsrc = vsi->ieq;
+ vmem = &vsi->ieq_mem;
break;
default:
i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s: error resource type = 0x%x\n",
@@ -732,45 +814,14 @@ void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev,
case PUDA_HASH_CRC_COMPLETE:
i40iw_free_hash_desc(rsrc->hash_desc);
case PUDA_QP_CREATED:
- do {
- if (reset)
- break;
- ret = dev->iw_priv_qp_ops->qp_destroy(&rsrc->qp,
- 0, false, true, true);
- if (ret)
- i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA,
- "%s error ieq qp destroy\n",
- __func__);
-
- ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
- I40IW_CQP_OP_DESTROY_QP,
- &compl_info);
- if (ret)
- i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA,
- "%s error ieq qp destroy done\n",
- __func__);
- } while (0);
+ if (!reset)
+ i40iw_puda_free_qp(rsrc);
i40iw_free_dma_mem(dev->hw, &rsrc->qpmem);
/* fallthrough */
case PUDA_CQ_CREATED:
- do {
- if (reset)
- break;
- ret = dev->iw_priv_cq_ops->cq_destroy(&rsrc->cq, 0, true);
- if (ret)
- i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA,
- "%s error ieq cq destroy\n",
- __func__);
-
- ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp,
- I40IW_CQP_OP_DESTROY_CQ,
- &compl_info);
- if (ret)
- i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA,
- "%s error ieq qp destroy done\n",
- __func__);
- } while (0);
+ if (!reset)
+ i40iw_puda_free_cq(rsrc);
i40iw_free_dma_mem(dev->hw, &rsrc->cqmem);
break;
@@ -826,9 +877,10 @@ static enum i40iw_status_code i40iw_puda_allocbufs(struct i40iw_puda_rsrc *rsrc,
* @dev: iwarp device
* @info: resource information
*/
-enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
+enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_vsi *vsi,
struct i40iw_puda_rsrc_info *info)
{
+ struct i40iw_sc_dev *dev = vsi->dev;
enum i40iw_status_code ret = 0;
struct i40iw_puda_rsrc *rsrc;
u32 pudasize;
@@ -841,10 +893,10 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
rqwridsize = info->rq_size * 8;
switch (info->type) {
case I40IW_PUDA_RSRC_TYPE_ILQ:
- vmem = &dev->ilq_mem;
+ vmem = &vsi->ilq_mem;
break;
case I40IW_PUDA_RSRC_TYPE_IEQ:
- vmem = &dev->ieq_mem;
+ vmem = &vsi->ieq_mem;
break;
default:
return I40IW_NOT_SUPPORTED;
@@ -857,22 +909,22 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
rsrc = (struct i40iw_puda_rsrc *)vmem->va;
spin_lock_init(&rsrc->bufpool_lock);
if (info->type == I40IW_PUDA_RSRC_TYPE_ILQ) {
- dev->ilq = (struct i40iw_puda_rsrc *)vmem->va;
- dev->ilq_count = info->count;
+ vsi->ilq = (struct i40iw_puda_rsrc *)vmem->va;
+ vsi->ilq_count = info->count;
rsrc->receive = info->receive;
rsrc->xmit_complete = info->xmit_complete;
} else {
- vmem = &dev->ieq_mem;
- dev->ieq_count = info->count;
- dev->ieq = (struct i40iw_puda_rsrc *)vmem->va;
+ vmem = &vsi->ieq_mem;
+ vsi->ieq_count = info->count;
+ vsi->ieq = (struct i40iw_puda_rsrc *)vmem->va;
rsrc->receive = i40iw_ieq_receive;
rsrc->xmit_complete = i40iw_ieq_tx_compl;
}
+ rsrc->ceq_valid = info->ceq_valid;
rsrc->type = info->type;
rsrc->sq_wrtrk_array = (struct i40iw_sq_uk_wr_trk_info *)((u8 *)vmem->va + pudasize);
rsrc->rq_wrid_array = (u64 *)((u8 *)vmem->va + pudasize + sqwridsize);
- rsrc->mss = info->mss;
/* Initialize all ieq lists */
INIT_LIST_HEAD(&rsrc->bufpool);
INIT_LIST_HEAD(&rsrc->txpend);
@@ -886,6 +938,7 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
rsrc->cq_size = info->rq_size + info->sq_size;
rsrc->buf_size = info->buf_size;
rsrc->dev = dev;
+ rsrc->vsi = vsi;
ret = i40iw_puda_cq_create(rsrc);
if (!ret) {
@@ -920,7 +973,7 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
dev->ccq_ops->ccq_arm(&rsrc->cq);
return ret;
error:
- i40iw_puda_dele_resources(dev, info->type, false);
+ i40iw_puda_dele_resources(vsi, info->type, false);
return ret;
}
@@ -1333,7 +1386,7 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq,
}
if (pfpdu->mode && (fps != pfpdu->fps)) {
/* clean up qp as it is new partial sequence */
- i40iw_ieq_cleanup_qp(ieq->dev, qp);
+ i40iw_ieq_cleanup_qp(ieq, qp);
i40iw_debug(ieq->dev, I40IW_DEBUG_IEQ,
"%s: restarting new partial\n", __func__);
pfpdu->mode = false;
@@ -1345,7 +1398,7 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq,
pfpdu->rcv_nxt = fps;
pfpdu->fps = fps;
pfpdu->mode = true;
- pfpdu->max_fpdu_data = ieq->mss;
+ pfpdu->max_fpdu_data = ieq->vsi->mss;
pfpdu->pmode_count++;
INIT_LIST_HEAD(rxlist);
i40iw_ieq_check_first_buf(buf, fps);
@@ -1380,14 +1433,14 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq,
* @dev: iwarp device
* @buf: exception buffer received
*/
-static void i40iw_ieq_receive(struct i40iw_sc_dev *dev,
+static void i40iw_ieq_receive(struct i40iw_sc_vsi *vsi,
struct i40iw_puda_buf *buf)
{
- struct i40iw_puda_rsrc *ieq = dev->ieq;
+ struct i40iw_puda_rsrc *ieq = vsi->ieq;
struct i40iw_sc_qp *qp = NULL;
u32 wqe_idx = ieq->compl_rxwqe_idx;
- qp = i40iw_ieq_get_qp(dev, buf);
+ qp = i40iw_ieq_get_qp(vsi->dev, buf);
if (!qp) {
ieq->stats_bad_qp_id++;
i40iw_puda_ret_bufpool(ieq, buf);
@@ -1405,12 +1458,12 @@ static void i40iw_ieq_receive(struct i40iw_sc_dev *dev,
/**
* i40iw_ieq_tx_compl - put back after sending completed exception buffer
- * @dev: iwarp device
+ * @vsi: pointer to the vsi structure
* @sqwrid: pointer to puda buffer
*/
-static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid)
+static void i40iw_ieq_tx_compl(struct i40iw_sc_vsi *vsi, void *sqwrid)
{
- struct i40iw_puda_rsrc *ieq = dev->ieq;
+ struct i40iw_puda_rsrc *ieq = vsi->ieq;
struct i40iw_puda_buf *buf = (struct i40iw_puda_buf *)sqwrid;
i40iw_puda_ret_bufpool(ieq, buf);
@@ -1422,15 +1475,14 @@ static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid)
/**
* i40iw_ieq_cleanup_qp - qp is being destroyed
- * @dev: iwarp device
+ * @ieq: ieq resource
* @qp: all pending fpdu buffers
*/
-void i40iw_ieq_cleanup_qp(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
+static void i40iw_ieq_cleanup_qp(struct i40iw_puda_rsrc *ieq, struct i40iw_sc_qp *qp)
{
struct i40iw_puda_buf *buf;
struct i40iw_pfpdu *pfpdu = &qp->pfpdu;
struct list_head *rxlist = &pfpdu->rxlist;
- struct i40iw_puda_rsrc *ieq = dev->ieq;
if (!pfpdu->mode)
return;
diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.h b/drivers/infiniband/hw/i40iw/i40iw_puda.h
index 52bf782..dba05ce 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_puda.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_puda.h
@@ -100,6 +100,7 @@ struct i40iw_puda_rsrc_info {
enum puda_resource_type type; /* ILQ or IEQ */
u32 count;
u16 pd_id;
+ bool ceq_valid;
u32 cq_id;
u32 qp_id;
u32 sq_size;
@@ -107,8 +108,8 @@ struct i40iw_puda_rsrc_info {
u16 buf_size;
u16 mss;
u32 tx_buf_cnt; /* total bufs allocated will be rq_size + tx_buf_cnt */
- void (*receive)(struct i40iw_sc_dev *, struct i40iw_puda_buf *);
- void (*xmit_complete)(struct i40iw_sc_dev *, void *);
+ void (*receive)(struct i40iw_sc_vsi *, struct i40iw_puda_buf *);
+ void (*xmit_complete)(struct i40iw_sc_vsi *, void *);
};
struct i40iw_puda_rsrc {
@@ -116,6 +117,7 @@ struct i40iw_puda_rsrc {
struct i40iw_sc_qp qp;
struct i40iw_sc_pd sc_pd;
struct i40iw_sc_dev *dev;
+ struct i40iw_sc_vsi *vsi;
struct i40iw_dma_mem cqmem;
struct i40iw_dma_mem qpmem;
struct i40iw_virt_mem ilq_mem;
@@ -123,6 +125,7 @@ struct i40iw_puda_rsrc {
enum puda_resource_type type;
u16 buf_size; /*buffer must be max datalen + tcpip hdr + mac */
u16 mss;
+ bool ceq_valid;
u32 cq_id;
u32 qp_id;
u32 sq_size;
@@ -142,8 +145,8 @@ struct i40iw_puda_rsrc {
u32 avail_buf_count; /* snapshot of currently available buffers */
spinlock_t bufpool_lock;
struct i40iw_puda_buf *alloclist;
- void (*receive)(struct i40iw_sc_dev *, struct i40iw_puda_buf *);
- void (*xmit_complete)(struct i40iw_sc_dev *, void *);
+ void (*receive)(struct i40iw_sc_vsi *, struct i40iw_puda_buf *);
+ void (*xmit_complete)(struct i40iw_sc_vsi *, void *);
/* puda stats */
u64 stats_buf_alloc_fail;
u64 stats_pkt_rcvd;
@@ -160,14 +163,13 @@ void i40iw_puda_send_buf(struct i40iw_puda_rsrc *rsrc,
struct i40iw_puda_buf *buf);
enum i40iw_status_code i40iw_puda_send(struct i40iw_sc_qp *qp,
struct i40iw_puda_send_info *info);
-enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev,
+enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_vsi *vsi,
struct i40iw_puda_rsrc_info *info);
-void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev,
+void i40iw_puda_dele_resources(struct i40iw_sc_vsi *vsi,
enum puda_resource_type type,
bool reset);
enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev,
struct i40iw_sc_cq *cq, u32 *compl_err);
-void i40iw_ieq_cleanup_qp(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp);
struct i40iw_sc_qp *i40iw_ieq_get_qp(struct i40iw_sc_dev *dev,
struct i40iw_puda_buf *buf);
@@ -180,4 +182,8 @@ void i40iw_ieq_mpa_crc_ae(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp);
void i40iw_free_hash_desc(struct shash_desc *desc);
void i40iw_ieq_update_tcpip_info(struct i40iw_puda_buf *buf, u16 length,
u32 seqnum);
+enum i40iw_status_code i40iw_cqp_qp_create_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp);
+enum i40iw_status_code i40iw_cqp_cq_create_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq);
+void i40iw_cqp_qp_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp);
+void i40iw_cqp_cq_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq);
#endif
diff --git a/drivers/infiniband/hw/i40iw/i40iw_type.h b/drivers/infiniband/hw/i40iw/i40iw_type.h
index ac4869f..68502ba 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_type.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_type.h
@@ -61,7 +61,7 @@ struct i40iw_cq_shadow_area {
struct i40iw_sc_dev;
struct i40iw_hmc_info;
-struct i40iw_dev_pestat;
+struct i40iw_vsi_pestat;
struct i40iw_cqp_ops;
struct i40iw_ccq_ops;
@@ -191,7 +191,7 @@ enum i40iw_debug_flag {
I40IW_DEBUG_ALL = 0xFFFFFFFF
};
-enum i40iw_hw_stat_index_32b {
+enum i40iw_hw_stats_index_32b {
I40IW_HW_STAT_INDEX_IP4RXDISCARD = 0,
I40IW_HW_STAT_INDEX_IP4RXTRUNC,
I40IW_HW_STAT_INDEX_IP4TXNOROUTE,
@@ -204,7 +204,7 @@ enum i40iw_hw_stat_index_32b {
I40IW_HW_STAT_INDEX_MAX_32
};
-enum i40iw_hw_stat_index_64b {
+enum i40iw_hw_stats_index_64b {
I40IW_HW_STAT_INDEX_IP4RXOCTS = 0,
I40IW_HW_STAT_INDEX_IP4RXPKTS,
I40IW_HW_STAT_INDEX_IP4RXFRAGS,
@@ -234,32 +234,23 @@ enum i40iw_hw_stat_index_64b {
I40IW_HW_STAT_INDEX_MAX_64
};
-struct i40iw_dev_hw_stat_offsets {
- u32 stat_offset_32[I40IW_HW_STAT_INDEX_MAX_32];
- u32 stat_offset_64[I40IW_HW_STAT_INDEX_MAX_64];
+struct i40iw_dev_hw_stats_offsets {
+ u32 stats_offset_32[I40IW_HW_STAT_INDEX_MAX_32];
+ u32 stats_offset_64[I40IW_HW_STAT_INDEX_MAX_64];
};
struct i40iw_dev_hw_stats {
- u64 stat_value_32[I40IW_HW_STAT_INDEX_MAX_32];
- u64 stat_value_64[I40IW_HW_STAT_INDEX_MAX_64];
+ u64 stats_value_32[I40IW_HW_STAT_INDEX_MAX_32];
+ u64 stats_value_64[I40IW_HW_STAT_INDEX_MAX_64];
};
-struct i40iw_device_pestat_ops {
- void (*iw_hw_stat_init)(struct i40iw_dev_pestat *, u8, struct i40iw_hw *, bool);
- void (*iw_hw_stat_read_32)(struct i40iw_dev_pestat *, enum i40iw_hw_stat_index_32b, u64 *);
- void (*iw_hw_stat_read_64)(struct i40iw_dev_pestat *, enum i40iw_hw_stat_index_64b, u64 *);
- void (*iw_hw_stat_read_all)(struct i40iw_dev_pestat *, struct i40iw_dev_hw_stats *);
- void (*iw_hw_stat_refresh_all)(struct i40iw_dev_pestat *);
-};
-
-struct i40iw_dev_pestat {
+struct i40iw_vsi_pestat {
struct i40iw_hw *hw;
- struct i40iw_device_pestat_ops ops;
struct i40iw_dev_hw_stats hw_stats;
struct i40iw_dev_hw_stats last_read_hw_stats;
- struct i40iw_dev_hw_stat_offsets hw_stat_offsets;
+ struct i40iw_dev_hw_stats_offsets hw_stats_offsets;
struct timer_list stats_timer;
- spinlock_t stats_lock; /* rdma stats lock */
+ spinlock_t lock; /* rdma stats lock */
};
struct i40iw_hw {
@@ -355,6 +346,7 @@ struct i40iw_sc_cq {
u64 cq_pa;
u64 shadow_area_pa;
struct i40iw_sc_dev *dev;
+ struct i40iw_sc_vsi *vsi;
void *pbl_list;
void *back_cq;
u32 ceq_id;
@@ -378,6 +370,7 @@ struct i40iw_sc_qp {
u64 shadow_area_pa;
u64 q2_pa;
struct i40iw_sc_dev *dev;
+ struct i40iw_sc_vsi *vsi;
struct i40iw_sc_pd *pd;
u64 *hw_host_ctx;
void *llp_stream_handle;
@@ -441,7 +434,7 @@ struct i40iw_qos {
struct i40iw_vfdev {
struct i40iw_sc_dev *pf_dev;
u8 *hmc_info_mem;
- struct i40iw_dev_pestat dev_pestat;
+ struct i40iw_vsi_pestat pestat;
struct i40iw_hmc_pble_info *pble_info;
struct i40iw_hmc_info hmc_info;
struct i40iw_vchnl_vf_msg_buffer vf_msg_buffer;
@@ -455,11 +448,28 @@ struct i40iw_vfdev {
bool stats_initialized;
};
+#define I40IW_INVALID_FCN_ID 0xff
+struct i40iw_sc_vsi {
+ struct i40iw_sc_dev *dev;
+ void *back_vsi; /* Owned by OS */
+ u32 ilq_count;
+ struct i40iw_virt_mem ilq_mem;
+ struct i40iw_puda_rsrc *ilq;
+ u32 ieq_count;
+ struct i40iw_virt_mem ieq_mem;
+ struct i40iw_puda_rsrc *ieq;
+ u16 mss;
+ u8 fcn_id;
+ bool stats_fcn_id_alloc;
+ struct i40iw_qos qos[I40IW_MAX_USER_PRIORITY];
+ struct i40iw_vsi_pestat *pestat;
+};
+
struct i40iw_sc_dev {
struct list_head cqp_cmd_head; /* head of the CQP command list */
spinlock_t cqp_lock; /* cqp list sync */
struct i40iw_dev_uk dev_uk;
- struct i40iw_dev_pestat dev_pestat;
+ bool fcn_id_array[I40IW_MAX_STATS_COUNT];
struct i40iw_dma_mem vf_fpm_query_buf[I40IW_MAX_PE_ENABLED_VF_COUNT];
u64 fpm_query_buf_pa;
u64 fpm_commit_buf_pa;
@@ -486,18 +496,9 @@ struct i40iw_sc_dev {
struct i40iw_cqp_misc_ops *cqp_misc_ops;
struct i40iw_hmc_ops *hmc_ops;
struct i40iw_vchnl_if vchnl_if;
- u32 ilq_count;
- struct i40iw_virt_mem ilq_mem;
- struct i40iw_puda_rsrc *ilq;
- u32 ieq_count;
- struct i40iw_virt_mem ieq_mem;
- struct i40iw_puda_rsrc *ieq;
-
const struct i40iw_vf_cqp_ops *iw_vf_cqp_ops;
struct i40iw_hmc_fpm_misc hmc_fpm_misc;
- struct i40iw_qos qos[I40IW_MAX_USER_PRIORITY];
- u16 mss;
u32 debug_mask;
u16 exception_lan_queue;
u8 hmc_fn_id;
@@ -571,6 +572,19 @@ struct i40iw_l2params {
u16 mss;
};
+struct i40iw_vsi_init_info {
+ struct i40iw_sc_dev *dev;
+ void *back_vsi;
+ struct i40iw_l2params *params;
+};
+
+struct i40iw_vsi_stats_info {
+ struct i40iw_vsi_pestat *pestat;
+ u8 fcn_id;
+ bool alloc_fcn_id;
+ bool stats_initialize;
+};
+
struct i40iw_device_init_info {
u64 fpm_query_buf_pa;
u64 fpm_commit_buf_pa;
@@ -579,7 +593,6 @@ struct i40iw_device_init_info {
struct i40iw_hw *hw;
void __iomem *bar0;
enum i40iw_status_code (*vchnl_send)(struct i40iw_sc_dev *, u32, u8 *, u16);
- struct i40iw_l2params l2params;
u16 exception_lan_queue;
u8 hmc_fn_id;
bool is_pf;
@@ -831,6 +844,7 @@ struct i40iw_register_shared_stag {
struct i40iw_qp_init_info {
struct i40iw_qp_uk_init_info qp_uk_init_info;
struct i40iw_sc_pd *pd;
+ struct i40iw_sc_vsi *vsi;
u64 *host_ctx;
u8 *q2;
u64 sq_pa;
@@ -897,6 +911,7 @@ enum i40iw_quad_hash_manage_type {
};
struct i40iw_qhash_table_info {
+ struct i40iw_sc_vsi *vsi;
enum i40iw_quad_hash_manage_type manage;
enum i40iw_quad_entry_type entry_type;
bool vlan_valid;
diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c
index 7d4af77..0f5d43d 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_utils.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c
@@ -761,7 +761,7 @@ void i40iw_qp_mss_modify(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
memset(&info, 0, sizeof(info));
info.mss_change = true;
- info.new_mss = dev->mss;
+ info.new_mss = qp->vsi->mss;
i40iw_hw_modify_qp(iwdev, iwqp, &info, false);
}
@@ -1068,6 +1068,116 @@ enum i40iw_status_code i40iw_vf_wait_vchnl_resp(struct i40iw_sc_dev *dev)
}
/**
+ * i40iw_cqp_cq_create_cmd - create a cq for the cqp
+ * @dev: device pointer
+ * @cq: pointer to created cq
+ */
+enum i40iw_status_code i40iw_cqp_cq_create_cmd(struct i40iw_sc_dev *dev,
+ struct i40iw_sc_cq *cq)
+{
+ struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev;
+ struct i40iw_cqp *iwcqp = &iwdev->cqp;
+ struct i40iw_cqp_request *cqp_request;
+ struct cqp_commands_info *cqp_info;
+ enum i40iw_status_code status;
+
+ cqp_request = i40iw_get_cqp_request(iwcqp, true);
+ if (!cqp_request)
+ return I40IW_ERR_NO_MEMORY;
+
+ cqp_info = &cqp_request->info;
+ cqp_info->cqp_cmd = OP_CQ_CREATE;
+ cqp_info->post_sq = 1;
+ cqp_info->in.u.cq_create.cq = cq;
+ cqp_info->in.u.cq_create.scratch = (uintptr_t)cqp_request;
+ status = i40iw_handle_cqp_op(iwdev, cqp_request);
+ if (status)
+ i40iw_pr_err("CQP-OP Create QP fail");
+
+ return status;
+}
+
+/**
+ * i40iw_cqp_qp_create_cmd - create a qp for the cqp
+ * @dev: device pointer
+ * @qp: pointer to created qp
+ */
+enum i40iw_status_code i40iw_cqp_qp_create_cmd(struct i40iw_sc_dev *dev,
+ struct i40iw_sc_qp *qp)
+{
+ struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev;
+ struct i40iw_cqp *iwcqp = &iwdev->cqp;
+ struct i40iw_cqp_request *cqp_request;
+ struct cqp_commands_info *cqp_info;
+ struct i40iw_create_qp_info *qp_info;
+ enum i40iw_status_code status;
+
+ cqp_request = i40iw_get_cqp_request(iwcqp, true);
+ if (!cqp_request)
+ return I40IW_ERR_NO_MEMORY;
+
+ cqp_info = &cqp_request->info;
+ qp_info = &cqp_request->info.in.u.qp_create.info;
+
+ memset(qp_info, 0, sizeof(*qp_info));
+
+ qp_info->cq_num_valid = true;
+ qp_info->next_iwarp_state = I40IW_QP_STATE_RTS;
+
+ cqp_info->cqp_cmd = OP_QP_CREATE;
+ cqp_info->post_sq = 1;
+ cqp_info->in.u.qp_create.qp = qp;
+ cqp_info->in.u.qp_create.scratch = (uintptr_t)cqp_request;
+ status = i40iw_handle_cqp_op(iwdev, cqp_request);
+ if (status)
+ i40iw_pr_err("CQP-OP QP create fail");
+ return status;
+}
+
+/**
+ * i40iw_cqp_cq_destroy_cmd - destroy the cqp cq
+ * @dev: device pointer
+ * @cq: pointer to cq
+ */
+void i40iw_cqp_cq_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq)
+{
+ struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev;
+
+ i40iw_cq_wq_destroy(iwdev, cq);
+}
+
+/**
+ * i40iw_cqp_qp_destroy_cmd - destroy the cqp
+ * @dev: device pointer
+ * @qp: pointer to qp
+ */
+void i40iw_cqp_qp_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp)
+{
+ struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev;
+ struct i40iw_cqp *iwcqp = &iwdev->cqp;
+ struct i40iw_cqp_request *cqp_request;
+ struct cqp_commands_info *cqp_info;
+ enum i40iw_status_code status;
+
+ cqp_request = i40iw_get_cqp_request(iwcqp, true);
+ if (!cqp_request)
+ return;
+
+ cqp_info = &cqp_request->info;
+ memset(cqp_info, 0, sizeof(*cqp_info));
+
+ cqp_info->cqp_cmd = OP_QP_DESTROY;
+ cqp_info->post_sq = 1;
+ cqp_info->in.u.qp_destroy.qp = qp;
+ cqp_info->in.u.qp_destroy.scratch = (uintptr_t)cqp_request;
+ cqp_info->in.u.qp_destroy.remove_hash_idx = true;
+ status = i40iw_handle_cqp_op(iwdev, cqp_request);
+ if (status)
+ i40iw_pr_err("CQP QP_DESTROY fail");
+}
+
+
+/**
* i40iw_ieq_mpa_crc_ae - generate AE for crc error
* @dev: hardware control device structure
* @qp: hardware control qp
@@ -1281,27 +1391,29 @@ enum i40iw_status_code i40iw_puda_get_tcpip_info(struct i40iw_puda_completion_in
/**
* i40iw_hw_stats_timeout - Stats timer-handler which updates all HW stats
- * @dev: hardware control device structure
+ * @vsi: pointer to the vsi structure
*/
-static void i40iw_hw_stats_timeout(unsigned long dev)
+static void i40iw_hw_stats_timeout(unsigned long vsi)
{
- struct i40iw_sc_dev *pf_dev = (struct i40iw_sc_dev *)dev;
- struct i40iw_dev_pestat *pf_devstat = &pf_dev->dev_pestat;
- struct i40iw_dev_pestat *vf_devstat = NULL;
+ struct i40iw_sc_vsi *sc_vsi = (struct i40iw_sc_vsi *)vsi;
+ struct i40iw_sc_dev *pf_dev = sc_vsi->dev;
+ struct i40iw_vsi_pestat *pf_devstat = sc_vsi->pestat;
+ struct i40iw_vsi_pestat *vf_devstat = NULL;
u16 iw_vf_idx;
unsigned long flags;
/*PF*/
- pf_devstat->ops.iw_hw_stat_read_all(pf_devstat, &pf_devstat->hw_stats);
+ i40iw_hw_stats_read_all(pf_devstat, &pf_devstat->hw_stats);
+
for (iw_vf_idx = 0; iw_vf_idx < I40IW_MAX_PE_ENABLED_VF_COUNT; iw_vf_idx++) {
- spin_lock_irqsave(&pf_devstat->stats_lock, flags);
+ spin_lock_irqsave(&pf_devstat->lock, flags);
if (pf_dev->vf_dev[iw_vf_idx]) {
if (pf_dev->vf_dev[iw_vf_idx]->stats_initialized) {
- vf_devstat = &pf_dev->vf_dev[iw_vf_idx]->dev_pestat;
- vf_devstat->ops.iw_hw_stat_read_all(vf_devstat, &vf_devstat->hw_stats);
+ vf_devstat = &pf_dev->vf_dev[iw_vf_idx]->pestat;
+ i40iw_hw_stats_read_all(vf_devstat, &vf_devstat->hw_stats);
}
}
- spin_unlock_irqrestore(&pf_devstat->stats_lock, flags);
+ spin_unlock_irqrestore(&pf_devstat->lock, flags);
}
mod_timer(&pf_devstat->stats_timer,
@@ -1310,26 +1422,26 @@ static void i40iw_hw_stats_timeout(unsigned long dev)
/**
* i40iw_hw_stats_start_timer - Start periodic stats timer
- * @dev: hardware control device structure
+ * @vsi: pointer to the vsi structure
*/
-void i40iw_hw_stats_start_timer(struct i40iw_sc_dev *dev)
+void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi)
{
- struct i40iw_dev_pestat *devstat = &dev->dev_pestat;
+ struct i40iw_vsi_pestat *devstat = vsi->pestat;
init_timer(&devstat->stats_timer);
devstat->stats_timer.function = i40iw_hw_stats_timeout;
- devstat->stats_timer.data = (unsigned long)dev;
+ devstat->stats_timer.data = (unsigned long)vsi;
mod_timer(&devstat->stats_timer,
jiffies + msecs_to_jiffies(STATS_TIMER_DELAY));
}
/**
- * i40iw_hw_stats_del_timer - Delete periodic stats timer
- * @dev: hardware control device structure
+ * i40iw_hw_stats_stop_timer - Delete periodic stats timer
+ * @vsi: pointer to the vsi structure
*/
-void i40iw_hw_stats_del_timer(struct i40iw_sc_dev *dev)
+void i40iw_hw_stats_stop_timer(struct i40iw_sc_vsi *vsi)
{
- struct i40iw_dev_pestat *devstat = &dev->dev_pestat;
+ struct i40iw_vsi_pestat *devstat = vsi->pestat;
del_timer_sync(&devstat->stats_timer);
}
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index 18526e6..855e499 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -623,6 +623,7 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd,
sq_size = init_attr->cap.max_send_wr;
rq_size = init_attr->cap.max_recv_wr;
+ init_info.vsi = &iwdev->vsi;
init_info.qp_uk_init_info.sq_size = sq_size;
init_info.qp_uk_init_info.rq_size = rq_size;
init_info.qp_uk_init_info.max_sq_frag_cnt = init_attr->cap.max_send_sge;
@@ -1052,11 +1053,11 @@ static void cq_free_resources(struct i40iw_device *iwdev, struct i40iw_cq *iwcq)
}
/**
- * cq_wq_destroy - send cq destroy cqp
+ * i40iw_cq_wq_destroy - send cq destroy cqp
* @iwdev: iwarp device
* @cq: hardware control cq
*/
-static void cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq)
+void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq)
{
enum i40iw_status_code status;
struct i40iw_cqp_request *cqp_request;
@@ -1095,7 +1096,7 @@ static int i40iw_destroy_cq(struct ib_cq *ib_cq)
iwcq = to_iwcq(ib_cq);
iwdev = to_iwdev(ib_cq->device);
cq = &iwcq->sc_cq;
- cq_wq_destroy(iwdev, cq);
+ i40iw_cq_wq_destroy(iwdev, cq);
cq_free_resources(iwdev, iwcq);
kfree(iwcq);
i40iw_rem_devusecount(iwdev);
@@ -1253,7 +1254,7 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev,
return (struct ib_cq *)iwcq;
cq_destroy:
- cq_wq_destroy(iwdev, cq);
+ i40iw_cq_wq_destroy(iwdev, cq);
cq_free_resources:
cq_free_resources(iwdev, iwcq);
error:
@@ -2632,15 +2633,11 @@ static int i40iw_get_hw_stats(struct ib_device *ibdev,
{
struct i40iw_device *iwdev = to_iwdev(ibdev);
struct i40iw_sc_dev *dev = &iwdev->sc_dev;
- struct i40iw_dev_pestat *devstat = &dev->dev_pestat;
+ struct i40iw_vsi_pestat *devstat = iwdev->vsi.pestat;
struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats;
- unsigned long flags;
if (dev->is_pf) {
- spin_lock_irqsave(&devstat->stats_lock, flags);
- devstat->ops.iw_hw_stat_read_all(devstat,
- &devstat->hw_stats);
- spin_unlock_irqrestore(&devstat->stats_lock, flags);
+ i40iw_hw_stats_read_all(devstat, &devstat->hw_stats);
} else {
if (i40iw_vchnl_vf_get_pe_stats(dev, &devstat->hw_stats))
return -ENOSYS;
diff --git a/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c b/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c
index dbd39c4..f4d1368 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c
@@ -403,6 +403,19 @@ static void pf_del_hmc_obj_callback(void *work_vf_dev)
}
/**
+ * i40iw_vf_init_pestat - Initialize stats for VF
+ * @devL pointer to the VF Device
+ * @stats: Statistics structure pointer
+ * @index: Stats index
+ */
+static void i40iw_vf_init_pestat(struct i40iw_sc_dev *dev, struct i40iw_vsi_pestat *stats, u16 index)
+{
+ stats->hw = dev->hw;
+ i40iw_hw_stats_init(stats, (u8)index, false);
+ spin_lock_init(&stats->lock);
+}
+
+/**
* i40iw_vchnl_recv_pf - Receive PF virtual channel messages
* @dev: IWARP device pointer
* @vf_id: Virtual function ID associated with the message
@@ -421,9 +434,8 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev,
u16 first_avail_iw_vf = I40IW_MAX_PE_ENABLED_VF_COUNT;
struct i40iw_virt_mem vf_dev_mem;
struct i40iw_virtchnl_work_info work_info;
- struct i40iw_dev_pestat *devstat;
+ struct i40iw_vsi_pestat *stats;
enum i40iw_status_code ret_code;
- unsigned long flags;
if (!dev || !msg || !len)
return I40IW_ERR_PARAM;
@@ -496,10 +508,7 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev,
i40iw_debug(dev, I40IW_DEBUG_VIRT,
"VF%u error CQP HMC Function operation.\n",
vf_id);
- i40iw_device_init_pestat(&vf_dev->dev_pestat);
- vf_dev->dev_pestat.ops.iw_hw_stat_init(&vf_dev->dev_pestat,
- (u8)vf_dev->pmf_index,
- dev->hw, false);
+ i40iw_vf_init_pestat(dev, &vf_dev->pestat, vf_dev->pmf_index);
vf_dev->stats_initialized = true;
} else {
if (vf_dev) {
@@ -530,12 +539,10 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev,
case I40IW_VCHNL_OP_GET_STATS:
if (!vf_dev)
return I40IW_ERR_BAD_PTR;
- devstat = &vf_dev->dev_pestat;
- spin_lock_irqsave(&dev->dev_pestat.stats_lock, flags);
- devstat->ops.iw_hw_stat_read_all(devstat, &devstat->hw_stats);
- spin_unlock_irqrestore(&dev->dev_pestat.stats_lock, flags);
+ stats = &vf_dev->pestat;
+ i40iw_hw_stats_read_all(stats, &stats->hw_stats);
vf_dev->msg_count--;
- vchnl_pf_send_get_pe_stats_resp(dev, vf_id, vchnl_msg, &devstat->hw_stats);
+ vchnl_pf_send_get_pe_stats_resp(dev, vf_id, vchnl_msg, &stats->hw_stats);
break;
default:
i40iw_debug(dev, I40IW_DEBUG_VIRT,
--
2.8.3
--
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: [PATCH net-next 7/7] bnxt_en: Add interface to support RDMA driver.
From: kbuild test robot @ 2016-12-06 22:33 UTC (permalink / raw)
To: Michael Chan
Cc: kbuild-all, davem, netdev, selvin.xavier, somnath.kotur, dledford,
linux-rdma
In-Reply-To: <1481044178-25193-8-git-send-email-michael.chan@broadcom.com>
[-- Attachment #1: Type: text/plain, Size: 5469 bytes --]
Hi Michael,
[auto build test WARNING on net-next/master]
url: https://github.com/0day-ci/linux/commits/Michael-Chan/bnxt_en-Add-interface-to-support-RDMA-driver/20161207-053721
config: i386-allmodconfig (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
Note: it may well be a FALSE warning. FWIW you are at least aware of it now.
http://gcc.gnu.org/wiki/Better_Uninitialized_Warnings
All warnings (new ones prefixed by >>):
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c: In function 'bnxt_unregister_dev':
>> drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c:97:4: warning: 'i' may be used uninitialized in this function [-Wmaybe-uninitialized]
i++;
~^~
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c: In function 'bnxt_ulp_stop':
>> drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c:219:6: warning: 'ops' may be used uninitialized in this function [-Wmaybe-uninitialized]
if (!ops || !ops->ulp_stop)
^
vim +/i +97 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
91 RCU_INIT_POINTER(ulp->ulp_ops, NULL);
92 synchronize_rcu();
93 ulp->max_async_event_id = 0;
94 ulp->async_events_bmap = NULL;
95 while (atomic_read(&ulp->ref_count) != 0 && i < 10) {
96 msleep(100);
> 97 i++;
98 }
99 return 0;
100 }
101
102 static int bnxt_req_msix_vecs(struct bnxt_en_dev *edev, int ulp_id,
103 struct bnxt_msix_entry *ent, int num_msix)
104 {
105 struct net_device *dev = edev->net;
106 struct bnxt *bp = netdev_priv(dev);
107 int max_idx, max_cp_rings;
108 int avail_msix, i, idx;
109
110 ASSERT_RTNL();
111 if (ulp_id != BNXT_ROCE_ULP)
112 return -EINVAL;
113
114 if (!(bp->flags & BNXT_FLAG_USING_MSIX))
115 return -ENODEV;
116
117 max_cp_rings = bnxt_get_max_func_cp_rings(bp);
118 max_idx = min_t(int, bp->total_irqs, max_cp_rings);
119 avail_msix = max_idx - bp->cp_nr_rings;
120 if (!avail_msix)
121 return -ENOMEM;
122 if (avail_msix > num_msix)
123 avail_msix = num_msix;
124
125 idx = max_idx - avail_msix;
126 for (i = 0; i < avail_msix; i++) {
127 ent[i].vector = bp->irq_tbl[idx + i].vector;
128 ent[i].ring_idx = idx + i;
129 ent[i].db_offset = (idx + i) * 0x80;
130 }
131 bnxt_set_max_func_irqs(bp, max_idx - avail_msix);
132 bnxt_set_max_func_cp_rings(bp, max_cp_rings - avail_msix);
133 edev->ulp_tbl[ulp_id].msix_requested = avail_msix;
134 return avail_msix;
135 }
136
137 static int bnxt_free_msix_vecs(struct bnxt_en_dev *edev, int ulp_id)
138 {
139 struct net_device *dev = edev->net;
140 struct bnxt *bp = netdev_priv(dev);
141 int max_cp_rings, msix_requested;
142
143 ASSERT_RTNL();
144 if (ulp_id != BNXT_ROCE_ULP)
145 return -EINVAL;
146
147 max_cp_rings = bnxt_get_max_func_cp_rings(bp);
148 msix_requested = edev->ulp_tbl[ulp_id].msix_requested;
149 bnxt_set_max_func_cp_rings(bp, max_cp_rings + msix_requested);
150 edev->ulp_tbl[ulp_id].msix_requested = 0;
151 bnxt_set_max_func_irqs(bp, bp->total_irqs);
152 return 0;
153 }
154
155 void bnxt_subtract_ulp_resources(struct bnxt *bp, int ulp_id)
156 {
157 ASSERT_RTNL();
158 if (bnxt_ulp_registered(bp->edev, ulp_id)) {
159 struct bnxt_en_dev *edev = bp->edev;
160 unsigned int msix_req, max;
161
162 msix_req = edev->ulp_tbl[ulp_id].msix_requested;
163 max = bnxt_get_max_func_cp_rings(bp);
164 bnxt_set_max_func_cp_rings(bp, max - msix_req);
165 max = bnxt_get_max_func_stat_ctxs(bp);
166 bnxt_set_max_func_stat_ctxs(bp, max - 1);
167 }
168 }
169
170 static int bnxt_send_msg(struct bnxt_en_dev *edev, int ulp_id,
171 struct bnxt_fw_msg *fw_msg)
172 {
173 struct net_device *dev = edev->net;
174 struct bnxt *bp = netdev_priv(dev);
175 struct input *req;
176 int rc;
177
178 mutex_lock(&bp->hwrm_cmd_lock);
179 req = fw_msg->msg;
180 req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
181 rc = _hwrm_send_message(bp, fw_msg->msg, fw_msg->msg_len,
182 fw_msg->timeout);
183 if (!rc) {
184 struct output *resp = bp->hwrm_cmd_resp_addr;
185 u32 len = le16_to_cpu(resp->resp_len);
186
187 if (fw_msg->resp_max_len < len)
188 len = fw_msg->resp_max_len;
189
190 memcpy(fw_msg->resp, resp, len);
191 }
192 mutex_unlock(&bp->hwrm_cmd_lock);
193 return rc;
194 }
195
196 static void bnxt_ulp_get(struct bnxt_ulp *ulp)
197 {
198 atomic_inc(&ulp->ref_count);
199 }
200
201 static void bnxt_ulp_put(struct bnxt_ulp *ulp)
202 {
203 atomic_dec(&ulp->ref_count);
204 }
205
206 void bnxt_ulp_stop(struct bnxt *bp)
207 {
208 struct bnxt_en_dev *edev = bp->edev;
209 struct bnxt_ulp_ops *ops;
210 int i;
211
212 if (!edev)
213 return;
214
215 for (i = 0; i < BNXT_MAX_ULP; i++) {
216 struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
217
218 rtnl_dereference(ulp->ulp_ops);
> 219 if (!ops || !ops->ulp_stop)
220 continue;
221 ops->ulp_stop(ulp->handle);
222 }
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 57018 bytes --]
^ permalink raw reply
* Re: [PATCH net-next 2/7] bnxt_en: Enable MSIX early in bnxt_init_one().
From: kbuild test robot @ 2016-12-06 23:55 UTC (permalink / raw)
To: Michael Chan
Cc: kbuild-all-JC7UmRfGjtg, davem-fT/PcQaiUtIeIZ0/mPfg9Q,
netdev-u79uwXL29TY76Z2rM5mHXA,
selvin.xavier-dY08KVG/lbpWk0Htik3J/w,
somnath.kotur-dY08KVG/lbpWk0Htik3J/w,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1481044178-25193-3-git-send-email-michael.chan-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 8301 bytes --]
Hi Michael,
[auto build test WARNING on net-next/master]
url: https://github.com/0day-ci/linux/commits/Michael-Chan/bnxt_en-Add-interface-to-support-RDMA-driver/20161207-053721
config: i386-randconfig-h1-12070631 (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All warnings (new ones prefixed by >>):
drivers/net/ethernet/broadcom/bnxt/bnxt.c: In function 'bnxt_get_max_func_irqs':
>> drivers/net/ethernet/broadcom/bnxt/bnxt.c:4818:1: warning: control reaches end of non-void function [-Wreturn-type]
}
^
Cyclomatic Complexity 5 include/linux/compiler.h:__read_once_size
Cyclomatic Complexity 5 include/linux/compiler.h:__write_once_size
Cyclomatic Complexity 2 arch/x86/include/asm/bitops.h:set_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:__set_bit
Cyclomatic Complexity 2 arch/x86/include/asm/bitops.h:clear_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:__clear_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:test_and_set_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:test_and_clear_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:constant_test_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:variable_test_bit
Cyclomatic Complexity 1 arch/x86/include/asm/bitops.h:fls
Cyclomatic Complexity 1 include/linux/log2.h:__ilog2_u32
Cyclomatic Complexity 1 include/linux/list.h:INIT_LIST_HEAD
Cyclomatic Complexity 1 include/asm-generic/getorder.h:__get_order
Cyclomatic Complexity 1 include/linux/err.h:ERR_PTR
Cyclomatic Complexity 1 arch/x86/include/asm/irqflags.h:arch_irqs_disabled_flags
Cyclomatic Complexity 1 arch/x86/include/asm/processor.h:prefetch
Cyclomatic Complexity 1 arch/x86/include/asm/atomic.h:atomic_read
Cyclomatic Complexity 1 arch/x86/include/asm/atomic.h:atomic_set
Cyclomatic Complexity 1 arch/x86/include/asm/atomic.h:atomic_inc
Cyclomatic Complexity 1 arch/x86/include/asm/atomic.h:atomic_cmpxchg
Cyclomatic Complexity 5 arch/x86/include/asm/preempt.h:__preempt_count_add
Cyclomatic Complexity 1 include/linux/bottom_half.h:__local_bh_disable_ip
Cyclomatic Complexity 1 include/linux/bottom_half.h:local_bh_disable
Cyclomatic Complexity 1 include/linux/spinlock.h:spinlock_check
Cyclomatic Complexity 1 include/linux/spinlock.h:spin_lock
Cyclomatic Complexity 1 include/linux/spinlock.h:spin_lock_bh
Cyclomatic Complexity 1 include/linux/spinlock.h:spin_unlock
Cyclomatic Complexity 1 include/linux/spinlock.h:spin_unlock_bh
Cyclomatic Complexity 1 include/linux/workqueue.h:__init_work
Cyclomatic Complexity 1 arch/x86/include/asm/topology.h:numa_node_id
Cyclomatic Complexity 1 include/linux/topology.h:numa_mem_id
Cyclomatic Complexity 1 include/linux/gfp.h:gfp_zonelist
Cyclomatic Complexity 1 include/linux/gfp.h:node_zonelist
Cyclomatic Complexity 1 include/linux/kasan.h:kasan_kmalloc
Cyclomatic Complexity 28 include/linux/slab.h:kmalloc_index
Cyclomatic Complexity 1 include/linux/slab.h:kmem_cache_alloc_trace
Cyclomatic Complexity 1 include/linux/slab.h:kmalloc_order_trace
Cyclomatic Complexity 67 include/linux/slab.h:kmalloc_large
Cyclomatic Complexity 5 include/linux/slab.h:kmalloc
Cyclomatic Complexity 1 include/linux/slab.h:kzalloc
Cyclomatic Complexity 1 arch/x86/include/asm/io.h:readl
Cyclomatic Complexity 1 arch/x86/include/asm/io.h:writel
Cyclomatic Complexity 1 include/linux/device.h:dev_get_drvdata
Cyclomatic Complexity 1 include/linux/device.h:dev_set_drvdata
Cyclomatic Complexity 1 include/linux/pci.h:pci_is_bridge
Cyclomatic Complexity 1 include/linux/mm.h:lowmem_page_address
Cyclomatic Complexity 1 include/linux/mm.h:page_is_pfmemalloc
Cyclomatic Complexity 1 include/linux/pci.h:pci_disable_msix
Cyclomatic Complexity 1 include/linux/pci.h:pci_enable_msix_range
Cyclomatic Complexity 1 include/linux/pci.h:pci_get_drvdata
Cyclomatic Complexity 1 include/linux/pci.h:pci_set_drvdata
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_map_page
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_mapping_error
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_unmap_page
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_alloc_coherent
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_free_coherent
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_sync_single_for_cpu
Cyclomatic Complexity 1 include/linux/dma-debug.h:debug_dma_sync_single_for_device
Cyclomatic Complexity 1 include/linux/kmemcheck.h:kmemcheck_mark_initialized
Cyclomatic Complexity 1 include/linux/dma-mapping.h:valid_dma_direction
Cyclomatic Complexity 1 arch/x86/include/asm/dma-mapping.h:get_dma_ops
Cyclomatic Complexity 2 include/linux/dma-mapping.h:dma_mapping_error
Cyclomatic Complexity 1 include/linux/dynamic_queue_limits.h:dql_avail
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_frag_size
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_frag_size_set
Cyclomatic Complexity 1 include/linux/skbuff.h:__skb_set_hash
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_set_hash
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_end_pointer
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_headlen
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_reserve
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_inner_transport_header
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_inner_network_header
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_transport_header
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_transport_offset
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_inner_network_header_len
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_inner_network_offset
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_frag_page
Cyclomatic Complexity 1 include/linux/skbuff.h:__skb_frag_set_page
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_get_queue_mapping
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_record_rx_queue
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_is_gso
Cyclomatic Complexity 1 include/linux/skbuff.h:skb_checksum_none_assert
Cyclomatic Complexity 1 include/linux/netdevice.h:napi_disable_pending
Cyclomatic Complexity 3 include/linux/netdevice.h:napi_schedule_prep
Cyclomatic Complexity 1 include/linux/netdevice.h:netdev_get_num_tc
Cyclomatic Complexity 1 include/linux/netdevice.h:netdev_get_tx_queue
Cyclomatic Complexity 1 include/linux/netdevice.h:netdev_priv
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_tx_stop_queue
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_tx_queue_stopped
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_running
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_carrier_ok
Cyclomatic Complexity 1 include/linux/netdevice.h:__netif_tx_lock
Cyclomatic Complexity 1 include/linux/netdevice.h:__netif_tx_unlock
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_addr_lock_bh
Cyclomatic Complexity 1 include/linux/netdevice.h:netif_addr_unlock_bh
Cyclomatic Complexity 1 include/linux/etherdevice.h:is_zero_ether_addr
Cyclomatic Complexity 1 include/linux/etherdevice.h:is_multicast_ether_addr
vim +4818 drivers/net/ethernet/broadcom/bnxt/bnxt.c
4802 if (bp->flags & BNXT_FLAG_USING_MSIX)
4803 bnxt_setup_msix(bp);
4804 else
4805 bnxt_setup_inta(bp);
4806
4807 rc = bnxt_set_real_num_queues(bp);
4808 return rc;
4809 }
4810
4811 static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
4812 {
4813 if (BNXT_PF(bp))
4814 return bp->pf.max_irqs;
4815 #if defined(CONFIG_BNXT_SRIOV)
4816 return bp->vf.max_irqs;
4817 #endif
> 4818 }
4819
4820 void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
4821 {
4822 if (BNXT_PF(bp))
4823 bp->pf.max_irqs = max_irqs;
4824 #if defined(CONFIG_BNXT_SRIOV)
4825 else
4826 bp->vf.max_irqs = max_irqs;
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 28906 bytes --]
^ permalink raw reply
* Re: Use ib_drain_qp instead of ib_drain_rq in ib_srp
From: Bart Van Assche @ 2016-12-07 1:04 UTC (permalink / raw)
To: Max Gurtovoy, sagig, Christoph Hellwig,
swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <7ec8b32c-f813-5a86-b7e9-b1272bc28b2c-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 814 bytes --]
On 12/06/2016 02:01 AM, Max Gurtovoy wrote:
> I understand why the beacon/drain is needed but I'm wondering if it's
> enough draining only the recv_q/send_q of the qp.
> These are 2 separate queues in the qp so receiving the beacon on the
> recv_q, In theory doesn't ensure the all the send_q elements flushed.
> Moreover if the cq is different for recv/send queues.
> I can't prove it currently but I think I saw these happens when
> developing iSER over FreeBSD (It was long time ago so it might have
> happend because we posted stuff after posting the beacon).
> In the current implementation we can't drain with IB_POLL_DIRECT ctx
> (like srp send_cq) but we might think of implementation that can enable
> it if needed.
Hello Max,
Do you agree with the approach of the attached two patches?
Thanks,
Bart.
[-- Attachment #2: 0001-IB-Introduce-ib_set_cq_poll_context.patch --]
[-- Type: text/x-patch, Size: 3583 bytes --]
>From c2d548d4c0bddafe38a3d75da16e3ac5789356b2 Mon Sep 17 00:00:00 2001
From: Bart Van Assche <bart.vanassche-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
Date: Tue, 6 Dec 2016 08:49:01 -0800
Subject: [PATCH 1/2] IB: Introduce ib_set_cq_poll_context()
Signed-off-by: Bart Van Assche <bart.vanassche-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
---
drivers/infiniband/core/cq.c | 66 +++++++++++++++++++++++++++++++-------------
include/rdma/ib_verbs.h | 1 +
2 files changed, 48 insertions(+), 19 deletions(-)
diff --git a/drivers/infiniband/core/cq.c b/drivers/infiniband/core/cq.c
index a754fc727de5..89ec5af63c5f 100644
--- a/drivers/infiniband/core/cq.c
+++ b/drivers/infiniband/core/cq.c
@@ -111,6 +111,50 @@ static void ib_cq_completion_workqueue(struct ib_cq *cq, void *private)
}
/**
+ * ib_set_cq_poll_context() - set CQ polling context
+ * @cq: completion queue pointer.
+ * @poll_ctx: context to poll the CQ from.
+ *
+ * Notes:
+ * - Changing the polling context is only allowed if the current context
+ * is %IB_POLL_DIRECT.
+ * - The caller must ensure that no other context is concurrently calling
+ * ib_poll_cq().
+ */
+int ib_set_cq_poll_context(struct ib_cq *cq, enum ib_poll_context poll_ctx)
+{
+ int ret = -EINVAL;
+
+ if (cq->poll_ctx != IB_POLL_DIRECT)
+ goto out;
+
+ cq->poll_ctx = poll_ctx;
+ switch (cq->poll_ctx) {
+ case IB_POLL_DIRECT:
+ cq->comp_handler = ib_cq_completion_direct;
+ break;
+ case IB_POLL_SOFTIRQ:
+ cq->comp_handler = ib_cq_completion_softirq;
+ irq_poll_init(&cq->iop, IB_POLL_BUDGET_IRQ, ib_poll_handler);
+ ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
+ break;
+ case IB_POLL_WORKQUEUE:
+ cq->comp_handler = ib_cq_completion_workqueue;
+ INIT_WORK(&cq->work, ib_cq_poll_work);
+ ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
+ break;
+ default:
+ goto out;
+ }
+
+ ret = 0;
+
+out:
+ return ret;
+}
+EXPORT_SYMBOL(ib_set_cq_poll_context);
+
+/**
* ib_alloc_cq - allocate a completion queue
* @dev: device to allocate the CQ for
* @private: driver private data, accessible from cq->cq_context
@@ -141,32 +185,16 @@ struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private,
cq->uobject = NULL;
cq->event_handler = NULL;
cq->cq_context = private;
- cq->poll_ctx = poll_ctx;
+ cq->poll_ctx = IB_POLL_DIRECT;
atomic_set(&cq->usecnt, 0);
cq->wc = kmalloc_array(IB_POLL_BATCH, sizeof(*cq->wc), GFP_KERNEL);
if (!cq->wc)
goto out_destroy_cq;
- switch (cq->poll_ctx) {
- case IB_POLL_DIRECT:
- cq->comp_handler = ib_cq_completion_direct;
- break;
- case IB_POLL_SOFTIRQ:
- cq->comp_handler = ib_cq_completion_softirq;
-
- irq_poll_init(&cq->iop, IB_POLL_BUDGET_IRQ, ib_poll_handler);
- ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
- break;
- case IB_POLL_WORKQUEUE:
- cq->comp_handler = ib_cq_completion_workqueue;
- INIT_WORK(&cq->work, ib_cq_poll_work);
- ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
- break;
- default:
- ret = -EINVAL;
+ ret = ib_set_cq_poll_context(cq, poll_ctx);
+ if (ret)
goto out_free_wc;
- }
return cq;
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index de7e13e31b57..e45263ca7f92 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -2751,6 +2751,7 @@ struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private,
int nr_cqe, int comp_vector, enum ib_poll_context poll_ctx);
void ib_free_cq(struct ib_cq *cq);
int ib_process_cq_direct(struct ib_cq *cq, int budget);
+int ib_set_cq_poll_context(struct ib_cq *cq, enum ib_poll_context poll_ctx);
/**
* ib_create_cq - Creates a CQ on the specified device.
--
2.11.0
^ permalink raw reply related
* Donation of $ 1.5 million Dollars
From: Colin-u79uwXL29TY76Z2rM5mHXA, &, Chris-u79uwXL29TY76Z2rM5mHXA, Weir-u79uwXL29TY76Z2rM5mHXA @ 2016-12-07 1:39 UTC (permalink / raw)
To: Recipients-u79uwXL29TY76Z2rM5mHXA
My wife and I have awarded you with a donation of $ 1.5 million Dollars from part of our Jackpot Lottery of 161,653,000 Million Pounds, send your name, address, phone for claims.
View http://www.bbc.co.uk/news/uk-scotland-glasgow-west-18801698
We await your earliest response and God Bless you.
Best of luck.
Colin & Chris Weir
Reply to: chrisweir2011-x64nAZbjQS4AvxtiuMwx3w@public.gmane.org
--
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
* Re: [PATCH] net: return value of skb_linearize should be handled in Linux kernel
From: Cong Wang @ 2016-12-07 5:02 UTC (permalink / raw)
To: Zhouyi Zhou
Cc: faisal.latif-ral2JQCrhuEAvxtiuMwx3w,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w, Jeff Kirsher,
QLogic-Storage-Upstream-h88ZbnxC6KDQT0dZR+AlfA,
jejb-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8, Martin K. Petersen,
jth-DgEjT+Ai2ygdnm+yROfE0A, jon.maloy-IzeFyvvaP7pWk0Htik3J/w,
ying.xue-CWA4WttNNZF54TAoqtyWWQ, David Miller,
linux-rdma-u79uwXL29TY76Z2rM5mHXA, LKML, intel-wired-lan,
Linux Kernel Network Developers,
linux-scsi-u79uwXL29TY76Z2rM5mHXA,
fcoe-devel-s9riP+hp16TNLxjTenLetw,
tipc-discussion-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f
In-Reply-To: <1481008233-16777-1-git-send-email-zhouzhouyi-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
On Mon, Dec 5, 2016 at 11:10 PM, Zhouyi Zhou <zhouzhouyi-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
> index 2a653ec..ab787cb 100644
> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
> @@ -490,7 +490,11 @@ int ixgbe_fcoe_ddp(struct ixgbe_adapter *adapter,
> */
> if ((fh->fh_r_ctl == FC_RCTL_DD_SOL_DATA) &&
> (fctl & FC_FC_END_SEQ)) {
> - skb_linearize(skb);
> + int err = 0;
> +
> + err = skb_linearize(skb);
> + if (err)
> + return err;
You can reuse 'rc' instead of adding 'err'.
> crc = (struct fcoe_crc_eof *)skb_put(skb, sizeof(*crc));
> crc->fcoe_eof = FC_EOF_T;
> }
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> index fee1f29..4926d48 100644
> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> @@ -2173,8 +2173,7 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
> total_rx_bytes += ddp_bytes;
> total_rx_packets += DIV_ROUND_UP(ddp_bytes,
> mss);
> - }
> - if (!ddp_bytes) {
> + } else {
> dev_kfree_skb_any(skb);
> continue;
> }
This piece doesn't seem to be related.
--
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
* [PATCH net-next v2 0/7] bnxt_en: Add interface to support RDMA driver.
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
This series adds an interface to support a brand new RDMA driver bnxt_re.
The first step is to re-arrange some code so that pci_enable_msix() can
be called during pci probe. The purpose is to allow the RDMA driver to
initialize and stay initialized whether the netdev is up or down.
Then we make some changes to VF resource allocation so that there is
enough resources to support RDMA.
Finally the last patch adds a simple interface to allow the RDMA driver to
probe and register itself with any bnxt_en devices that support RDMA.
Once registered, the RDMA driver can request MSIX, send fw messages, and
receive some notifications.
v2: Fixed kbuild test robot warnings.
David, please consider this series for net-next. Thanks.
Michael Chan (7):
bnxt_en: Add bnxt_set_max_func_irqs().
bnxt_en: Enable MSIX early in bnxt_init_one().
bnxt_en: Move function reset to bnxt_init_one().
bnxt_en: Improve completion ring allocation for VFs.
bnxt_en: Reserve RDMA resources by default.
bnxt_en: Refactor the driver registration function with firmware.
bnxt_en: Add interface to support RDMA driver.
drivers/net/ethernet/broadcom/bnxt/Makefile | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 360 +++++++++++++++++-------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 22 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 14 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 346 +++++++++++++++++++++++
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h | 93 ++++++
6 files changed, 722 insertions(+), 115 deletions(-)
create mode 100644 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
create mode 100644 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
--
1.8.3.1
^ permalink raw reply
* [PATCH net-next v2 1/7] bnxt_en: Add bnxt_set_max_func_irqs().
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
By refactoring existing code into this new function. The new function
will be used in subsequent patches.
v2: Fixed compile warning when CONFIG_BNXT_SRIOV is not set.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 17 +++++++++++------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 +
2 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index e84613a..269e757 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4743,6 +4743,16 @@ static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
return 0;
}
+void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ bp->vf.max_irqs = max_irqs;
+ else
+#endif
+ bp->pf.max_irqs = max_irqs;
+}
+
static int bnxt_setup_msix(struct bnxt *bp)
{
struct msix_entry *msix_ent;
@@ -6949,12 +6959,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
bnxt_set_tpa_flags(bp);
bnxt_set_ring_params(bp);
- if (BNXT_PF(bp))
- bp->pf.max_irqs = max_irqs;
-#if defined(CONFIG_BNXT_SRIOV)
- else
- bp->vf.max_irqs = max_irqs;
-#endif
+ bnxt_set_max_func_irqs(bp, max_irqs);
bnxt_set_dflt_rings(bp);
/* Default RSS hash cfg. */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index b4abc1b..8327d0d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1235,6 +1235,7 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
int bnxt_hwrm_set_coal(struct bnxt *);
int bnxt_hwrm_func_qcaps(struct bnxt *);
+void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max);
void bnxt_tx_disable(struct bnxt *bp);
void bnxt_tx_enable(struct bnxt *bp);
int bnxt_hwrm_set_pause(struct bnxt *);
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 2/7] bnxt_en: Enable MSIX early in bnxt_init_one().
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
To better support the new RDMA driver, we need to move pci_enable_msix()
from bnxt_open() to bnxt_init_one(). This way, MSIX vectors are available
to the RDMA driver whether the network device is up or down.
Part of the existing bnxt_setup_int_mode() function is now refactored into
a new bnxt_init_int_mode(). bnxt_init_int_mode() is called during
bnxt_init_one() to enable MSIX. The remaining logic in
bnxt_setup_int_mode() to map the IRQs to the completion rings is called
during bnxt_open().
v2: Fixed compile warning when CONFIG_BNXT_SRIOV is not set.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 183 +++++++++++++++++++-----------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 +
2 files changed, 115 insertions(+), 69 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 269e757..da302eb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4743,6 +4743,80 @@ static int bnxt_trim_rings(struct bnxt *bp, int *rx, int *tx, int max,
return 0;
}
+static void bnxt_setup_msix(struct bnxt *bp)
+{
+ const int len = sizeof(bp->irq_tbl[0].name);
+ struct net_device *dev = bp->dev;
+ int tcs, i;
+
+ tcs = netdev_get_num_tc(dev);
+ if (tcs > 1) {
+ bp->tx_nr_rings_per_tc = bp->tx_nr_rings / tcs;
+ if (bp->tx_nr_rings_per_tc == 0) {
+ netdev_reset_tc(dev);
+ bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
+ } else {
+ int i, off, count;
+
+ bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tcs;
+ for (i = 0; i < tcs; i++) {
+ count = bp->tx_nr_rings_per_tc;
+ off = i * count;
+ netdev_set_tc_queue(dev, i, count, off);
+ }
+ }
+ }
+
+ for (i = 0; i < bp->cp_nr_rings; i++) {
+ char *attr;
+
+ if (bp->flags & BNXT_FLAG_SHARED_RINGS)
+ attr = "TxRx";
+ else if (i < bp->rx_nr_rings)
+ attr = "rx";
+ else
+ attr = "tx";
+
+ snprintf(bp->irq_tbl[i].name, len, "%s-%s-%d", dev->name, attr,
+ i);
+ bp->irq_tbl[i].handler = bnxt_msix;
+ }
+}
+
+static void bnxt_setup_inta(struct bnxt *bp)
+{
+ const int len = sizeof(bp->irq_tbl[0].name);
+
+ if (netdev_get_num_tc(bp->dev))
+ netdev_reset_tc(bp->dev);
+
+ snprintf(bp->irq_tbl[0].name, len, "%s-%s-%d", bp->dev->name, "TxRx",
+ 0);
+ bp->irq_tbl[0].handler = bnxt_inta;
+}
+
+static int bnxt_setup_int_mode(struct bnxt *bp)
+{
+ int rc;
+
+ if (bp->flags & BNXT_FLAG_USING_MSIX)
+ bnxt_setup_msix(bp);
+ else
+ bnxt_setup_inta(bp);
+
+ rc = bnxt_set_real_num_queues(bp);
+ return rc;
+}
+
+static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ return bp->vf.max_irqs;
+#endif
+ return bp->pf.max_irqs;
+}
+
void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
{
#if defined(CONFIG_BNXT_SRIOV)
@@ -4753,16 +4827,12 @@ void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max_irqs)
bp->pf.max_irqs = max_irqs;
}
-static int bnxt_setup_msix(struct bnxt *bp)
+static int bnxt_init_msix(struct bnxt *bp)
{
- struct msix_entry *msix_ent;
- struct net_device *dev = bp->dev;
int i, total_vecs, rc = 0, min = 1;
- const int len = sizeof(bp->irq_tbl[0].name);
-
- bp->flags &= ~BNXT_FLAG_USING_MSIX;
- total_vecs = bp->cp_nr_rings;
+ struct msix_entry *msix_ent;
+ total_vecs = bnxt_get_max_func_irqs(bp);
msix_ent = kcalloc(total_vecs, sizeof(struct msix_entry), GFP_KERNEL);
if (!msix_ent)
return -ENOMEM;
@@ -4783,8 +4853,10 @@ static int bnxt_setup_msix(struct bnxt *bp)
bp->irq_tbl = kcalloc(total_vecs, sizeof(struct bnxt_irq), GFP_KERNEL);
if (bp->irq_tbl) {
- int tcs;
+ for (i = 0; i < total_vecs; i++)
+ bp->irq_tbl[i].vector = msix_ent[i].vector;
+ bp->total_irqs = total_vecs;
/* Trim rings based upon num of vectors allocated */
rc = bnxt_trim_rings(bp, &bp->rx_nr_rings, &bp->tx_nr_rings,
total_vecs, min == 1);
@@ -4792,43 +4864,10 @@ static int bnxt_setup_msix(struct bnxt *bp)
goto msix_setup_exit;
bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
- tcs = netdev_get_num_tc(dev);
- if (tcs > 1) {
- bp->tx_nr_rings_per_tc = bp->tx_nr_rings / tcs;
- if (bp->tx_nr_rings_per_tc == 0) {
- netdev_reset_tc(dev);
- bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
- } else {
- int i, off, count;
+ bp->cp_nr_rings = (min == 1) ?
+ max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) :
+ bp->tx_nr_rings + bp->rx_nr_rings;
- bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tcs;
- for (i = 0; i < tcs; i++) {
- count = bp->tx_nr_rings_per_tc;
- off = i * count;
- netdev_set_tc_queue(dev, i, count, off);
- }
- }
- }
- bp->cp_nr_rings = total_vecs;
-
- for (i = 0; i < bp->cp_nr_rings; i++) {
- char *attr;
-
- bp->irq_tbl[i].vector = msix_ent[i].vector;
- if (bp->flags & BNXT_FLAG_SHARED_RINGS)
- attr = "TxRx";
- else if (i < bp->rx_nr_rings)
- attr = "rx";
- else
- attr = "tx";
-
- snprintf(bp->irq_tbl[i].name, len,
- "%s-%s-%d", dev->name, attr, i);
- bp->irq_tbl[i].handler = bnxt_msix;
- }
- rc = bnxt_set_real_num_queues(bp);
- if (rc)
- goto msix_setup_exit;
} else {
rc = -ENOMEM;
goto msix_setup_exit;
@@ -4838,52 +4877,54 @@ static int bnxt_setup_msix(struct bnxt *bp)
return 0;
msix_setup_exit:
- netdev_err(bp->dev, "bnxt_setup_msix err: %x\n", rc);
+ netdev_err(bp->dev, "bnxt_init_msix err: %x\n", rc);
+ kfree(bp->irq_tbl);
+ bp->irq_tbl = NULL;
pci_disable_msix(bp->pdev);
kfree(msix_ent);
return rc;
}
-static int bnxt_setup_inta(struct bnxt *bp)
+static int bnxt_init_inta(struct bnxt *bp)
{
- int rc;
- const int len = sizeof(bp->irq_tbl[0].name);
-
- if (netdev_get_num_tc(bp->dev))
- netdev_reset_tc(bp->dev);
-
bp->irq_tbl = kcalloc(1, sizeof(struct bnxt_irq), GFP_KERNEL);
- if (!bp->irq_tbl) {
- rc = -ENOMEM;
- return rc;
- }
+ if (!bp->irq_tbl)
+ return -ENOMEM;
+
+ bp->total_irqs = 1;
bp->rx_nr_rings = 1;
bp->tx_nr_rings = 1;
bp->cp_nr_rings = 1;
bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
bp->flags |= BNXT_FLAG_SHARED_RINGS;
bp->irq_tbl[0].vector = bp->pdev->irq;
- snprintf(bp->irq_tbl[0].name, len,
- "%s-%s-%d", bp->dev->name, "TxRx", 0);
- bp->irq_tbl[0].handler = bnxt_inta;
- rc = bnxt_set_real_num_queues(bp);
- return rc;
+ return 0;
}
-static int bnxt_setup_int_mode(struct bnxt *bp)
+static int bnxt_init_int_mode(struct bnxt *bp)
{
int rc = 0;
if (bp->flags & BNXT_FLAG_MSIX_CAP)
- rc = bnxt_setup_msix(bp);
+ rc = bnxt_init_msix(bp);
if (!(bp->flags & BNXT_FLAG_USING_MSIX) && BNXT_PF(bp)) {
/* fallback to INTA */
- rc = bnxt_setup_inta(bp);
+ rc = bnxt_init_inta(bp);
}
return rc;
}
+static void bnxt_clear_int_mode(struct bnxt *bp)
+{
+ if (bp->flags & BNXT_FLAG_USING_MSIX)
+ pci_disable_msix(bp->pdev);
+
+ kfree(bp->irq_tbl);
+ bp->irq_tbl = NULL;
+ bp->flags &= ~BNXT_FLAG_USING_MSIX;
+}
+
static void bnxt_free_irq(struct bnxt *bp)
{
struct bnxt_irq *irq;
@@ -4902,10 +4943,6 @@ static void bnxt_free_irq(struct bnxt *bp)
free_irq(irq->vector, bp->bnapi[i]);
irq->requested = 0;
}
- if (bp->flags & BNXT_FLAG_USING_MSIX)
- pci_disable_msix(bp->pdev);
- kfree(bp->irq_tbl);
- bp->irq_tbl = NULL;
}
static int bnxt_request_irq(struct bnxt *bp)
@@ -6695,6 +6732,7 @@ static void bnxt_remove_one(struct pci_dev *pdev)
cancel_work_sync(&bp->sp_task);
bp->sp_event = 0;
+ bnxt_clear_int_mode(bp);
bnxt_hwrm_func_drv_unrgtr(bp);
bnxt_free_hwrm_resources(bp);
bnxt_dcb_free(bp);
@@ -6990,10 +7028,14 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if (rc)
goto init_err;
- rc = register_netdev(dev);
+ rc = bnxt_init_int_mode(bp);
if (rc)
goto init_err;
+ rc = register_netdev(dev);
+ if (rc)
+ goto init_err_clr_int;
+
netdev_info(dev, "%s found at mem %lx, node addr %pM\n",
board_info[ent->driver_data].name,
(long)pci_resource_start(pdev, 0), dev->dev_addr);
@@ -7002,6 +7044,9 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
return 0;
+init_err_clr_int:
+ bnxt_clear_int_mode(bp);
+
init_err:
pci_iounmap(pdev, bp->bar0);
pci_release_regions(pdev);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 8327d0d..1461355 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1024,6 +1024,7 @@ struct bnxt {
#define BNXT_STATE_FN_RST_DONE 2
struct bnxt_irq *irq_tbl;
+ int total_irqs;
u8 mac_addr[ETH_ALEN];
#ifdef CONFIG_BNXT_DCB
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 3/7] bnxt_en: Move function reset to bnxt_init_one().
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
Now that MSIX is enabled in bnxt_init_one(), resources may be allocated by
the RDMA driver before the network device is opened. So we cannot do
function reset in bnxt_open() which will clear all the resources.
The proper place to do function reset now is in bnxt_init_one().
If we get AER, we'll do function reset as well.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 25 ++++++-------------------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 -
2 files changed, 6 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index da302eb..30b482b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5613,22 +5613,7 @@ int bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
static int bnxt_open(struct net_device *dev)
{
struct bnxt *bp = netdev_priv(dev);
- int rc = 0;
- if (!test_bit(BNXT_STATE_FN_RST_DONE, &bp->state)) {
- rc = bnxt_hwrm_func_reset(bp);
- if (rc) {
- netdev_err(bp->dev, "hwrm chip reset failure rc: %x\n",
- rc);
- rc = -EBUSY;
- return rc;
- }
- /* Do func_reset during the 1st PF open only to prevent killing
- * the VFs when the PF is brought down and up.
- */
- if (BNXT_PF(bp))
- set_bit(BNXT_STATE_FN_RST_DONE, &bp->state);
- }
return __bnxt_open_nic(bp, true, true);
}
@@ -7028,6 +7013,10 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if (rc)
goto init_err;
+ rc = bnxt_hwrm_func_reset(bp);
+ if (rc)
+ goto init_err;
+
rc = bnxt_init_int_mode(bp);
if (rc)
goto init_err;
@@ -7069,7 +7058,6 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
pci_channel_state_t state)
{
struct net_device *netdev = pci_get_drvdata(pdev);
- struct bnxt *bp = netdev_priv(netdev);
netdev_info(netdev, "PCI I/O error detected\n");
@@ -7084,8 +7072,6 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
if (netif_running(netdev))
bnxt_close(netdev);
- /* So that func_reset will be done during slot_reset */
- clear_bit(BNXT_STATE_FN_RST_DONE, &bp->state);
pci_disable_device(pdev);
rtnl_unlock();
@@ -7119,7 +7105,8 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
} else {
pci_set_master(pdev);
- if (netif_running(netdev))
+ err = bnxt_hwrm_func_reset(bp);
+ if (!err && netif_running(netdev))
err = bnxt_open(netdev);
if (!err)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 1461355..0ee2cc4 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1021,7 +1021,6 @@ struct bnxt {
unsigned long state;
#define BNXT_STATE_OPEN 0
#define BNXT_STATE_IN_SP_TASK 1
-#define BNXT_STATE_FN_RST_DONE 2
struct bnxt_irq *irq_tbl;
int total_irqs;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 4/7] bnxt_en: Improve completion ring allocation for VFs.
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
All available remaining completion rings not used by the PF should be
made available for the VFs so that there are enough rings in the VF to
support RDMA. The earlier workaround code of capping the rings by the
statistics context is removed.
When SRIOV is disabled, call a new function bnxt_restore_pf_fw_resources()
to restore FW resources. Later on we need to add some logic to account
for RDMA resources.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 8 +++++++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 14 ++++----------
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 30b482b..52b8ad4 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4152,7 +4152,7 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp)
return rc;
}
-int bnxt_hwrm_func_qcaps(struct bnxt *bp)
+static int bnxt_hwrm_func_qcaps(struct bnxt *bp)
{
int rc = 0;
struct hwrm_func_qcaps_input req = {0};
@@ -6856,6 +6856,12 @@ static int bnxt_set_dflt_rings(struct bnxt *bp)
return rc;
}
+void bnxt_restore_pf_fw_resources(struct bnxt *bp)
+{
+ ASSERT_RTNL();
+ bnxt_hwrm_func_qcaps(bp);
+}
+
static void bnxt_parse_log_pcie_link(struct bnxt *bp)
{
enum pcie_link_width width = PCIE_LNK_WIDTH_UNKNOWN;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 0ee2cc4..43a4b17 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1234,7 +1234,6 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int hwrm_send_message(struct bnxt *, void *, u32, int);
int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
int bnxt_hwrm_set_coal(struct bnxt *);
-int bnxt_hwrm_func_qcaps(struct bnxt *);
void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max);
void bnxt_tx_disable(struct bnxt *bp);
void bnxt_tx_enable(struct bnxt *bp);
@@ -1245,4 +1244,5 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int bnxt_close_nic(struct bnxt *, bool, bool);
int bnxt_setup_mq_tc(struct net_device *dev, u8 tc);
int bnxt_get_max_rings(struct bnxt *, int *, int *, bool);
+void bnxt_restore_pf_fw_resources(struct bnxt *bp);
#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index bff626a..c696025 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -420,15 +420,7 @@ static int bnxt_hwrm_func_cfg(struct bnxt *bp, int num_vfs)
bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1);
/* Remaining rings are distributed equally amongs VF's for now */
- /* TODO: the following workaroud is needed to restrict total number
- * of vf_cp_rings not exceed number of HW ring groups. This WA should
- * be removed once new HWRM provides HW ring groups capability in
- * hwrm_func_qcap.
- */
- vf_cp_rings = min_t(u16, pf->max_cp_rings, pf->max_stat_ctxs);
- vf_cp_rings = (vf_cp_rings - bp->cp_nr_rings) / num_vfs;
- /* TODO: restore this logic below once the WA above is removed */
- /* vf_cp_rings = (pf->max_cp_rings - bp->cp_nr_rings) / num_vfs; */
+ vf_cp_rings = (pf->max_cp_rings - bp->cp_nr_rings) / num_vfs;
vf_stat_ctx = (pf->max_stat_ctxs - bp->num_stat_ctxs) / num_vfs;
if (bp->flags & BNXT_FLAG_AGG_RINGS)
vf_rx_rings = (pf->max_rx_rings - bp->rx_nr_rings * 2) /
@@ -590,7 +582,9 @@ void bnxt_sriov_disable(struct bnxt *bp)
bp->pf.active_vfs = 0;
/* Reclaim all resources for the PF. */
- bnxt_hwrm_func_qcaps(bp);
+ rtnl_lock();
+ bnxt_restore_pf_fw_resources(bp);
+ rtnl_unlock();
}
int bnxt_sriov_configure(struct pci_dev *pdev, int num_vfs)
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 5/7] bnxt_en: Reserve RDMA resources by default.
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
If the device supports RDMA, we'll setup network default rings so that
there are enough minimum resources for RDMA, if possible. However, the
user can still increase network rings to the max if he wants. The actual
RDMA resources won't be reserved until the RDMA driver registers.
v2: Fix compile warning when BNXT_CONFIG_SRIOV is not set.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 58 ++++++++++++++++++++++++++++++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 9 +++++
2 files changed, 66 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 52b8ad4..57285bd 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4166,6 +4166,11 @@ static int bnxt_hwrm_func_qcaps(struct bnxt *bp)
if (rc)
goto hwrm_func_qcaps_exit;
+ if (resp->flags & cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED))
+ bp->flags |= BNXT_FLAG_ROCEV1_CAP;
+ if (resp->flags & cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED))
+ bp->flags |= BNXT_FLAG_ROCEV2_CAP;
+
bp->tx_push_thresh = 0;
if (resp->flags &
cpu_to_le32(FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED))
@@ -4808,6 +4813,24 @@ static int bnxt_setup_int_mode(struct bnxt *bp)
return rc;
}
+unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ return bp->vf.max_stat_ctxs;
+#endif
+ return bp->pf.max_stat_ctxs;
+}
+
+unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ return bp->vf.max_cp_rings;
+#endif
+ return bp->pf.max_cp_rings;
+}
+
static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
{
#if defined(CONFIG_BNXT_SRIOV)
@@ -6832,6 +6855,39 @@ int bnxt_get_max_rings(struct bnxt *bp, int *max_rx, int *max_tx, bool shared)
return bnxt_trim_rings(bp, max_rx, max_tx, cp, shared);
}
+static int bnxt_get_dflt_rings(struct bnxt *bp, int *max_rx, int *max_tx,
+ bool shared)
+{
+ int rc;
+
+ rc = bnxt_get_max_rings(bp, max_rx, max_tx, shared);
+ if (rc)
+ return rc;
+
+ if (bp->flags & BNXT_FLAG_ROCE_CAP) {
+ int max_cp, max_stat, max_irq;
+
+ /* Reserve minimum resources for RoCE */
+ max_cp = bnxt_get_max_func_cp_rings(bp);
+ max_stat = bnxt_get_max_func_stat_ctxs(bp);
+ max_irq = bnxt_get_max_func_irqs(bp);
+ if (max_cp <= BNXT_MIN_ROCE_CP_RINGS ||
+ max_irq <= BNXT_MIN_ROCE_CP_RINGS ||
+ max_stat <= BNXT_MIN_ROCE_STAT_CTXS)
+ return 0;
+
+ max_cp -= BNXT_MIN_ROCE_CP_RINGS;
+ max_irq -= BNXT_MIN_ROCE_CP_RINGS;
+ max_stat -= BNXT_MIN_ROCE_STAT_CTXS;
+ max_cp = min_t(int, max_cp, max_irq);
+ max_cp = min_t(int, max_cp, max_stat);
+ rc = bnxt_trim_rings(bp, max_rx, max_tx, max_cp, shared);
+ if (rc)
+ rc = 0;
+ }
+ return rc;
+}
+
static int bnxt_set_dflt_rings(struct bnxt *bp)
{
int dflt_rings, max_rx_rings, max_tx_rings, rc;
@@ -6840,7 +6896,7 @@ static int bnxt_set_dflt_rings(struct bnxt *bp)
if (sh)
bp->flags |= BNXT_FLAG_SHARED_RINGS;
dflt_rings = netif_get_num_default_rss_queues();
- rc = bnxt_get_max_rings(bp, &max_rx_rings, &max_tx_rings, sh);
+ rc = bnxt_get_dflt_rings(bp, &max_rx_rings, &max_tx_rings, sh);
if (rc)
return rc;
bp->rx_nr_rings = min_t(int, dflt_rings, max_rx_rings);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 43a4b17..d796836 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -387,6 +387,9 @@ struct rx_tpa_end_cmp_ext {
#define DB_KEY_TX_PUSH (0x4 << 28)
#define DB_LONG_TX_PUSH (0x2 << 24)
+#define BNXT_MIN_ROCE_CP_RINGS 2
+#define BNXT_MIN_ROCE_STAT_CTXS 1
+
#define INVALID_HW_RING_ID ((u16)-1)
/* The hardware supports certain page sizes. Use the supported page sizes
@@ -953,6 +956,10 @@ struct bnxt {
#define BNXT_FLAG_PORT_STATS 0x400
#define BNXT_FLAG_UDP_RSS_CAP 0x800
#define BNXT_FLAG_EEE_CAP 0x1000
+ #define BNXT_FLAG_ROCEV1_CAP 0x8000
+ #define BNXT_FLAG_ROCEV2_CAP 0x10000
+ #define BNXT_FLAG_ROCE_CAP (BNXT_FLAG_ROCEV1_CAP | \
+ BNXT_FLAG_ROCEV2_CAP)
#define BNXT_FLAG_CHIP_NITRO_A0 0x1000000
#define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA | \
@@ -1234,6 +1241,8 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int hwrm_send_message(struct bnxt *, void *, u32, int);
int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
int bnxt_hwrm_set_coal(struct bnxt *);
+unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp);
+unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp);
void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max);
void bnxt_tx_disable(struct bnxt *bp);
void bnxt_tx_enable(struct bnxt *bp);
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 6/7] bnxt_en: Refactor the driver registration function with firmware.
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
The driver register function with firmware consists of passing version
information and registering for async events. To support the RDMA driver,
the async events that we need to register may change. Separate the
driver register function into 2 parts so that we can just update the
async events for the RDMA driver.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 34 ++++++++++++++++++++++++++-----
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 ++
2 files changed, 31 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 57285bd..c782942 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3117,27 +3117,46 @@ int hwrm_send_message_silent(struct bnxt *bp, void *msg, u32 msg_len,
return rc;
}
-static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
+int bnxt_hwrm_func_rgtr_async_events(struct bnxt *bp, unsigned long *bmap,
+ int bmap_size)
{
struct hwrm_func_drv_rgtr_input req = {0};
- int i;
DECLARE_BITMAP(async_events_bmap, 256);
u32 *events = (u32 *)async_events_bmap;
+ int i;
bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);
req.enables =
- cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
- FUNC_DRV_RGTR_REQ_ENABLES_VER |
- FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
+ cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
memset(async_events_bmap, 0, sizeof(async_events_bmap));
for (i = 0; i < ARRAY_SIZE(bnxt_async_events_arr); i++)
__set_bit(bnxt_async_events_arr[i], async_events_bmap);
+ if (bmap && bmap_size) {
+ for (i = 0; i < bmap_size; i++) {
+ if (test_bit(i, bmap))
+ __set_bit(i, async_events_bmap);
+ }
+ }
+
for (i = 0; i < 8; i++)
req.async_event_fwd[i] |= cpu_to_le32(events[i]);
+ return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+}
+
+static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
+{
+ struct hwrm_func_drv_rgtr_input req = {0};
+
+ bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);
+
+ req.enables =
+ cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
+ FUNC_DRV_RGTR_REQ_ENABLES_VER);
+
req.os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
req.ver_maj = DRV_VER_MAJ;
req.ver_min = DRV_VER_MIN;
@@ -3146,6 +3165,7 @@ static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
if (BNXT_PF(bp)) {
DECLARE_BITMAP(vf_req_snif_bmap, 256);
u32 *data = (u32 *)vf_req_snif_bmap;
+ int i;
memset(vf_req_snif_bmap, 0, sizeof(vf_req_snif_bmap));
for (i = 0; i < ARRAY_SIZE(bnxt_vf_req_snif); i++)
@@ -7023,6 +7043,10 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if (rc)
goto init_err;
+ rc = bnxt_hwrm_func_rgtr_async_events(bp, NULL, 0);
+ if (rc)
+ goto init_err;
+
/* Get the MAX capabilities for this function */
rc = bnxt_hwrm_func_qcaps(bp);
if (rc) {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index d796836..eec2415 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1240,6 +1240,8 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int _hwrm_send_message(struct bnxt *, void *, u32, int);
int hwrm_send_message(struct bnxt *, void *, u32, int);
int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
+int bnxt_hwrm_func_rgtr_async_events(struct bnxt *bp, unsigned long *bmap,
+ int bmap_size);
int bnxt_hwrm_set_coal(struct bnxt *);
unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp);
unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp);
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 7/7] bnxt_en: Add interface to support RDMA driver.
From: Michael Chan @ 2016-12-07 5:26 UTC (permalink / raw)
To: davem; +Cc: netdev, selvin.xavier, somnath.kotur, dledford, linux-rdma
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan@broadcom.com>
Since the network driver and RDMA driver operate on the same PCI function,
we need to create an interface to allow the RDMA driver to share resources
with the network driver.
1. Create a new bnxt_en_dev struct which will be returned by
bnxt_ulp_probe() upon success. After that, all calls from the RDMA driver
to bnxt_en will pass a pointer to this struct.
2. This struct contains additional function pointers to register, request
msix, send fw messages, register for async events.
3. If the RDMA driver wants to enable RDMA on the function, it needs to
call the function pointer bnxt_register_device(). A ulp_ops structure
is passed for RCU protected upcalls from bnxt_en to the RDMA driver.
4. The RDMA driver can call firmware APIs using the bnxt_send_fw_msg()
function pointer.
5. 1 stats context is reserved when the RDMA driver registers. MSIX
and completion rings are reserved when the RDMA driver calls
bnxt_request_msix() function pointer.
6. When the RDMA driver calls bnxt_unregister_device(), all RDMA resources
will be cleaned up.
v2: Fixed 2 uninitialized variable warnings.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/Makefile | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 41 ++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 6 +
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 346 ++++++++++++++++++++++++++
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h | 93 +++++++
5 files changed, 483 insertions(+), 5 deletions(-)
create mode 100644 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
create mode 100644 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
diff --git a/drivers/net/ethernet/broadcom/bnxt/Makefile b/drivers/net/ethernet/broadcom/bnxt/Makefile
index b233a86..6082ed1 100644
--- a/drivers/net/ethernet/broadcom/bnxt/Makefile
+++ b/drivers/net/ethernet/broadcom/bnxt/Makefile
@@ -1,3 +1,3 @@
obj-$(CONFIG_BNXT) += bnxt_en.o
-bnxt_en-y := bnxt.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o
+bnxt_en-y := bnxt.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c782942..9608cb4 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -52,6 +52,7 @@
#include "bnxt_hsi.h"
#include "bnxt.h"
+#include "bnxt_ulp.h"
#include "bnxt_sriov.h"
#include "bnxt_ethtool.h"
#include "bnxt_dcb.h"
@@ -1528,12 +1529,11 @@ static int bnxt_async_event_process(struct bnxt *bp,
set_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event);
break;
default:
- netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n",
- event_id);
goto async_event_process_exit;
}
schedule_work(&bp->sp_task);
async_event_process_exit:
+ bnxt_ulp_async_events(bp, cmpl);
return 0;
}
@@ -3547,7 +3547,7 @@ static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, u16 vnic_id, u16 ctx_idx)
return rc;
}
-static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id)
+int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id)
{
unsigned int ring = 0, grp_idx;
struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];
@@ -3595,6 +3595,9 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id)
#endif
if ((bp->flags & BNXT_FLAG_STRIP_VLAN) || def_vlan)
req.flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE);
+ if (!vnic_id && bnxt_ulp_registered(bp->edev, BNXT_ROCE_ULP))
+ req.flags |=
+ cpu_to_le32(VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE);
return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}
@@ -4842,6 +4845,16 @@ unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp)
return bp->pf.max_stat_ctxs;
}
+void bnxt_set_max_func_stat_ctxs(struct bnxt *bp, unsigned int max)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ bp->vf.max_stat_ctxs = max;
+ else
+#endif
+ bp->pf.max_stat_ctxs = max;
+}
+
unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp)
{
#if defined(CONFIG_BNXT_SRIOV)
@@ -4851,6 +4864,16 @@ unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp)
return bp->pf.max_cp_rings;
}
+void bnxt_set_max_func_cp_rings(struct bnxt *bp, unsigned int max)
+{
+#if defined(CONFIG_BNXT_SRIOV)
+ if (BNXT_VF(bp))
+ bp->vf.max_cp_rings = max;
+ else
+#endif
+ bp->pf.max_cp_rings = max;
+}
+
static unsigned int bnxt_get_max_func_irqs(struct bnxt *bp)
{
#if defined(CONFIG_BNXT_SRIOV)
@@ -6767,6 +6790,8 @@ static void bnxt_remove_one(struct pci_dev *pdev)
pci_iounmap(pdev, bp->bar2);
pci_iounmap(pdev, bp->bar1);
pci_iounmap(pdev, bp->bar0);
+ kfree(bp->edev);
+ bp->edev = NULL;
free_netdev(dev);
pci_release_regions(pdev);
@@ -6936,6 +6961,7 @@ void bnxt_restore_pf_fw_resources(struct bnxt *bp)
{
ASSERT_RTNL();
bnxt_hwrm_func_qcaps(bp);
+ bnxt_subtract_ulp_resources(bp, BNXT_ROCE_ULP);
}
static void bnxt_parse_log_pcie_link(struct bnxt *bp)
@@ -7047,6 +7073,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if (rc)
goto init_err;
+ bp->ulp_probe = bnxt_ulp_probe;
+
/* Get the MAX capabilities for this function */
rc = bnxt_hwrm_func_qcaps(bp);
if (rc) {
@@ -7144,12 +7172,15 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
pci_channel_state_t state)
{
struct net_device *netdev = pci_get_drvdata(pdev);
+ struct bnxt *bp = netdev_priv(netdev);
netdev_info(netdev, "PCI I/O error detected\n");
rtnl_lock();
netif_device_detach(netdev);
+ bnxt_ulp_stop(bp);
+
if (state == pci_channel_io_perm_failure) {
rtnl_unlock();
return PCI_ERS_RESULT_DISCONNECT;
@@ -7195,8 +7226,10 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
if (!err && netif_running(netdev))
err = bnxt_open(netdev);
- if (!err)
+ if (!err) {
result = PCI_ERS_RESULT_RECOVERED;
+ bnxt_ulp_start(bp);
+ }
}
if (result != PCI_ERS_RESULT_RECOVERED && netif_running(netdev))
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index eec2415..16defe9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -972,6 +972,9 @@ struct bnxt {
#define BNXT_SINGLE_PF(bp) (BNXT_PF(bp) && !BNXT_NPAR(bp))
#define BNXT_CHIP_TYPE_NITRO_A0(bp) ((bp)->flags & BNXT_FLAG_CHIP_NITRO_A0)
+ struct bnxt_en_dev *edev;
+ struct bnxt_en_dev * (*ulp_probe)(struct net_device *);
+
struct bnxt_napi **bnapi;
struct bnxt_rx_ring_info *rx_ring;
@@ -1242,9 +1245,12 @@ static inline void bnxt_disable_poll(struct bnxt_napi *bnapi)
int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
int bnxt_hwrm_func_rgtr_async_events(struct bnxt *bp, unsigned long *bmap,
int bmap_size);
+int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id);
int bnxt_hwrm_set_coal(struct bnxt *);
unsigned int bnxt_get_max_func_stat_ctxs(struct bnxt *bp);
+void bnxt_set_max_func_stat_ctxs(struct bnxt *bp, unsigned int max);
unsigned int bnxt_get_max_func_cp_rings(struct bnxt *bp);
+void bnxt_set_max_func_cp_rings(struct bnxt *bp, unsigned int max);
void bnxt_set_max_func_irqs(struct bnxt *bp, unsigned int max);
void bnxt_tx_disable(struct bnxt *bp);
void bnxt_tx_enable(struct bnxt *bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
new file mode 100644
index 0000000..8b7464b
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c
@@ -0,0 +1,346 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2016 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/bitops.h>
+#include <linux/irq.h>
+#include <asm/byteorder.h>
+#include <linux/bitmap.h>
+
+#include "bnxt_hsi.h"
+#include "bnxt.h"
+#include "bnxt_ulp.h"
+
+static int bnxt_register_dev(struct bnxt_en_dev *edev, int ulp_id,
+ struct bnxt_ulp_ops *ulp_ops, void *handle)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ struct bnxt_ulp *ulp;
+
+ ASSERT_RTNL();
+ if (ulp_id >= BNXT_MAX_ULP)
+ return -EINVAL;
+
+ ulp = &edev->ulp_tbl[ulp_id];
+ if (rcu_access_pointer(ulp->ulp_ops)) {
+ netdev_err(bp->dev, "ulp id %d already registered\n", ulp_id);
+ return -EBUSY;
+ }
+ if (ulp_id == BNXT_ROCE_ULP) {
+ unsigned int max_stat_ctxs;
+
+ max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp);
+ if (max_stat_ctxs <= BNXT_MIN_ROCE_STAT_CTXS ||
+ bp->num_stat_ctxs == max_stat_ctxs)
+ return -ENOMEM;
+ bnxt_set_max_func_stat_ctxs(bp, max_stat_ctxs -
+ BNXT_MIN_ROCE_STAT_CTXS);
+ }
+
+ atomic_set(&ulp->ref_count, 0);
+ ulp->handle = handle;
+ rcu_assign_pointer(ulp->ulp_ops, ulp_ops);
+
+ if (ulp_id == BNXT_ROCE_ULP) {
+ if (test_bit(BNXT_STATE_OPEN, &bp->state))
+ bnxt_hwrm_vnic_cfg(bp, 0);
+ }
+
+ return 0;
+}
+
+static int bnxt_unregister_dev(struct bnxt_en_dev *edev, int ulp_id)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ struct bnxt_ulp *ulp;
+ int i = 0;
+
+ ASSERT_RTNL();
+ if (ulp_id >= BNXT_MAX_ULP)
+ return -EINVAL;
+
+ ulp = &edev->ulp_tbl[ulp_id];
+ if (!rcu_access_pointer(ulp->ulp_ops)) {
+ netdev_err(bp->dev, "ulp id %d not registered\n", ulp_id);
+ return -EINVAL;
+ }
+ if (ulp_id == BNXT_ROCE_ULP) {
+ unsigned int max_stat_ctxs;
+
+ max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp);
+ bnxt_set_max_func_stat_ctxs(bp, max_stat_ctxs + 1);
+ }
+ if (ulp->max_async_event_id)
+ bnxt_hwrm_func_rgtr_async_events(bp, NULL, 0);
+
+ RCU_INIT_POINTER(ulp->ulp_ops, NULL);
+ synchronize_rcu();
+ ulp->max_async_event_id = 0;
+ ulp->async_events_bmap = NULL;
+ while (atomic_read(&ulp->ref_count) != 0 && i < 10) {
+ msleep(100);
+ i++;
+ }
+ return 0;
+}
+
+static int bnxt_req_msix_vecs(struct bnxt_en_dev *edev, int ulp_id,
+ struct bnxt_msix_entry *ent, int num_msix)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ int max_idx, max_cp_rings;
+ int avail_msix, i, idx;
+
+ ASSERT_RTNL();
+ if (ulp_id != BNXT_ROCE_ULP)
+ return -EINVAL;
+
+ if (!(bp->flags & BNXT_FLAG_USING_MSIX))
+ return -ENODEV;
+
+ max_cp_rings = bnxt_get_max_func_cp_rings(bp);
+ max_idx = min_t(int, bp->total_irqs, max_cp_rings);
+ avail_msix = max_idx - bp->cp_nr_rings;
+ if (!avail_msix)
+ return -ENOMEM;
+ if (avail_msix > num_msix)
+ avail_msix = num_msix;
+
+ idx = max_idx - avail_msix;
+ for (i = 0; i < avail_msix; i++) {
+ ent[i].vector = bp->irq_tbl[idx + i].vector;
+ ent[i].ring_idx = idx + i;
+ ent[i].db_offset = (idx + i) * 0x80;
+ }
+ bnxt_set_max_func_irqs(bp, max_idx - avail_msix);
+ bnxt_set_max_func_cp_rings(bp, max_cp_rings - avail_msix);
+ edev->ulp_tbl[ulp_id].msix_requested = avail_msix;
+ return avail_msix;
+}
+
+static int bnxt_free_msix_vecs(struct bnxt_en_dev *edev, int ulp_id)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ int max_cp_rings, msix_requested;
+
+ ASSERT_RTNL();
+ if (ulp_id != BNXT_ROCE_ULP)
+ return -EINVAL;
+
+ max_cp_rings = bnxt_get_max_func_cp_rings(bp);
+ msix_requested = edev->ulp_tbl[ulp_id].msix_requested;
+ bnxt_set_max_func_cp_rings(bp, max_cp_rings + msix_requested);
+ edev->ulp_tbl[ulp_id].msix_requested = 0;
+ bnxt_set_max_func_irqs(bp, bp->total_irqs);
+ return 0;
+}
+
+void bnxt_subtract_ulp_resources(struct bnxt *bp, int ulp_id)
+{
+ ASSERT_RTNL();
+ if (bnxt_ulp_registered(bp->edev, ulp_id)) {
+ struct bnxt_en_dev *edev = bp->edev;
+ unsigned int msix_req, max;
+
+ msix_req = edev->ulp_tbl[ulp_id].msix_requested;
+ max = bnxt_get_max_func_cp_rings(bp);
+ bnxt_set_max_func_cp_rings(bp, max - msix_req);
+ max = bnxt_get_max_func_stat_ctxs(bp);
+ bnxt_set_max_func_stat_ctxs(bp, max - 1);
+ }
+}
+
+static int bnxt_send_msg(struct bnxt_en_dev *edev, int ulp_id,
+ struct bnxt_fw_msg *fw_msg)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ struct input *req;
+ int rc;
+
+ mutex_lock(&bp->hwrm_cmd_lock);
+ req = fw_msg->msg;
+ req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
+ rc = _hwrm_send_message(bp, fw_msg->msg, fw_msg->msg_len,
+ fw_msg->timeout);
+ if (!rc) {
+ struct output *resp = bp->hwrm_cmd_resp_addr;
+ u32 len = le16_to_cpu(resp->resp_len);
+
+ if (fw_msg->resp_max_len < len)
+ len = fw_msg->resp_max_len;
+
+ memcpy(fw_msg->resp, resp, len);
+ }
+ mutex_unlock(&bp->hwrm_cmd_lock);
+ return rc;
+}
+
+static void bnxt_ulp_get(struct bnxt_ulp *ulp)
+{
+ atomic_inc(&ulp->ref_count);
+}
+
+static void bnxt_ulp_put(struct bnxt_ulp *ulp)
+{
+ atomic_dec(&ulp->ref_count);
+}
+
+void bnxt_ulp_stop(struct bnxt *bp)
+{
+ struct bnxt_en_dev *edev = bp->edev;
+ struct bnxt_ulp_ops *ops;
+ int i;
+
+ if (!edev)
+ return;
+
+ for (i = 0; i < BNXT_MAX_ULP; i++) {
+ struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+
+ ops = rtnl_dereference(ulp->ulp_ops);
+ if (!ops || !ops->ulp_stop)
+ continue;
+ ops->ulp_stop(ulp->handle);
+ }
+}
+
+void bnxt_ulp_start(struct bnxt *bp)
+{
+ struct bnxt_en_dev *edev = bp->edev;
+ struct bnxt_ulp_ops *ops;
+ int i;
+
+ if (!edev)
+ return;
+
+ for (i = 0; i < BNXT_MAX_ULP; i++) {
+ struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+
+ ops = rtnl_dereference(ulp->ulp_ops);
+ if (!ops || !ops->ulp_start)
+ continue;
+ ops->ulp_start(ulp->handle);
+ }
+}
+
+void bnxt_ulp_sriov_cfg(struct bnxt *bp, int num_vfs)
+{
+ struct bnxt_en_dev *edev = bp->edev;
+ struct bnxt_ulp_ops *ops;
+ int i;
+
+ if (!edev)
+ return;
+
+ for (i = 0; i < BNXT_MAX_ULP; i++) {
+ struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+
+ rcu_read_lock();
+ ops = rcu_dereference(ulp->ulp_ops);
+ if (!ops || !ops->ulp_sriov_config) {
+ rcu_read_unlock();
+ continue;
+ }
+ bnxt_ulp_get(ulp);
+ rcu_read_unlock();
+ ops->ulp_sriov_config(ulp->handle, num_vfs);
+ bnxt_ulp_put(ulp);
+ }
+}
+
+void bnxt_ulp_async_events(struct bnxt *bp, struct hwrm_async_event_cmpl *cmpl)
+{
+ u16 event_id = le16_to_cpu(cmpl->event_id);
+ struct bnxt_en_dev *edev = bp->edev;
+ struct bnxt_ulp_ops *ops;
+ int i;
+
+ if (!edev)
+ return;
+
+ rcu_read_lock();
+ for (i = 0; i < BNXT_MAX_ULP; i++) {
+ struct bnxt_ulp *ulp = &edev->ulp_tbl[i];
+
+ ops = rcu_dereference(ulp->ulp_ops);
+ if (!ops || !ops->ulp_async_notifier)
+ continue;
+ if (!ulp->async_events_bmap ||
+ event_id > ulp->max_async_event_id)
+ continue;
+
+ /* Read max_async_event_id first before testing the bitmap. */
+ smp_rmb();
+ if (test_bit(event_id, ulp->async_events_bmap))
+ ops->ulp_async_notifier(ulp->handle, cmpl);
+ }
+ rcu_read_unlock();
+}
+
+static int bnxt_register_async_events(struct bnxt_en_dev *edev, int ulp_id,
+ unsigned long *events_bmap, u16 max_id)
+{
+ struct net_device *dev = edev->net;
+ struct bnxt *bp = netdev_priv(dev);
+ struct bnxt_ulp *ulp;
+
+ if (ulp_id >= BNXT_MAX_ULP)
+ return -EINVAL;
+
+ ulp = &edev->ulp_tbl[ulp_id];
+ ulp->async_events_bmap = events_bmap;
+ /* Make sure bnxt_ulp_async_events() sees this order */
+ smp_wmb();
+ ulp->max_async_event_id = max_id;
+ bnxt_hwrm_func_rgtr_async_events(bp, events_bmap, max_id + 1);
+ return 0;
+}
+
+static const struct bnxt_en_ops bnxt_en_ops_tbl = {
+ .bnxt_register_device = bnxt_register_dev,
+ .bnxt_unregister_device = bnxt_unregister_dev,
+ .bnxt_request_msix = bnxt_req_msix_vecs,
+ .bnxt_free_msix = bnxt_free_msix_vecs,
+ .bnxt_send_fw_msg = bnxt_send_msg,
+ .bnxt_register_fw_async_events = bnxt_register_async_events,
+};
+
+struct bnxt_en_dev *bnxt_ulp_probe(struct net_device *dev)
+{
+ struct bnxt *bp = netdev_priv(dev);
+ struct bnxt_en_dev *edev;
+
+ edev = bp->edev;
+ if (!edev) {
+ edev = kzalloc(sizeof(*edev), GFP_KERNEL);
+ if (!edev)
+ return ERR_PTR(-ENOMEM);
+ edev->en_ops = &bnxt_en_ops_tbl;
+ if (bp->flags & BNXT_FLAG_ROCEV1_CAP)
+ edev->flags |= BNXT_EN_FLAG_ROCEV1_CAP;
+ if (bp->flags & BNXT_FLAG_ROCEV2_CAP)
+ edev->flags |= BNXT_EN_FLAG_ROCEV2_CAP;
+ edev->net = dev;
+ edev->pdev = bp->pdev;
+ bp->edev = edev;
+ }
+ return bp->edev;
+}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
new file mode 100644
index 0000000..74f816e
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h
@@ -0,0 +1,93 @@
+/* Broadcom NetXtreme-C/E network driver.
+ *
+ * Copyright (c) 2016 Broadcom Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef BNXT_ULP_H
+#define BNXT_ULP_H
+
+#define BNXT_ROCE_ULP 0
+#define BNXT_OTHER_ULP 1
+#define BNXT_MAX_ULP 2
+
+#define BNXT_MIN_ROCE_CP_RINGS 2
+#define BNXT_MIN_ROCE_STAT_CTXS 1
+
+struct hwrm_async_event_cmpl;
+struct bnxt;
+
+struct bnxt_ulp_ops {
+ /* async_notifier() cannot sleep (in BH context) */
+ void (*ulp_async_notifier)(void *, struct hwrm_async_event_cmpl *);
+ void (*ulp_stop)(void *);
+ void (*ulp_start)(void *);
+ void (*ulp_sriov_config)(void *, int);
+};
+
+struct bnxt_msix_entry {
+ u32 vector;
+ u32 ring_idx;
+ u32 db_offset;
+};
+
+struct bnxt_fw_msg {
+ void *msg;
+ int msg_len;
+ void *resp;
+ int resp_max_len;
+ int timeout;
+};
+
+struct bnxt_ulp {
+ void *handle;
+ struct bnxt_ulp_ops __rcu *ulp_ops;
+ unsigned long *async_events_bmap;
+ u16 max_async_event_id;
+ u16 msix_requested;
+ atomic_t ref_count;
+};
+
+struct bnxt_en_dev {
+ struct net_device *net;
+ struct pci_dev *pdev;
+ u32 flags;
+ #define BNXT_EN_FLAG_ROCEV1_CAP 0x1
+ #define BNXT_EN_FLAG_ROCEV2_CAP 0x2
+ #define BNXT_EN_FLAG_ROCE_CAP (BNXT_EN_FLAG_ROCEV1_CAP | \
+ BNXT_EN_FLAG_ROCEV2_CAP)
+ const struct bnxt_en_ops *en_ops;
+ struct bnxt_ulp ulp_tbl[BNXT_MAX_ULP];
+};
+
+struct bnxt_en_ops {
+ int (*bnxt_register_device)(struct bnxt_en_dev *, int,
+ struct bnxt_ulp_ops *, void *);
+ int (*bnxt_unregister_device)(struct bnxt_en_dev *, int);
+ int (*bnxt_request_msix)(struct bnxt_en_dev *, int,
+ struct bnxt_msix_entry *, int);
+ int (*bnxt_free_msix)(struct bnxt_en_dev *, int);
+ int (*bnxt_send_fw_msg)(struct bnxt_en_dev *, int,
+ struct bnxt_fw_msg *);
+ int (*bnxt_register_fw_async_events)(struct bnxt_en_dev *, int,
+ unsigned long *, u16);
+};
+
+static inline bool bnxt_ulp_registered(struct bnxt_en_dev *edev, int ulp_id)
+{
+ if (edev && rcu_access_pointer(edev->ulp_tbl[ulp_id].ulp_ops))
+ return true;
+ return false;
+}
+
+void bnxt_subtract_ulp_resources(struct bnxt *bp, int ulp_id);
+void bnxt_ulp_stop(struct bnxt *bp);
+void bnxt_ulp_start(struct bnxt *bp);
+void bnxt_ulp_sriov_cfg(struct bnxt *bp, int num_vfs);
+void bnxt_ulp_async_events(struct bnxt *bp, struct hwrm_async_event_cmpl *cmpl);
+struct bnxt_en_dev *bnxt_ulp_probe(struct net_device *dev);
+
+#endif
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH] net: return value of skb_linearize should be handled in Linux kernel
From: Zhouyi Zhou @ 2016-12-07 6:27 UTC (permalink / raw)
To: Cong Wang
Cc: faisal.latif, dledford, sean.hefty, Hal Rosenstock, Jeff Kirsher,
QLogic-Storage-Upstream, jejb, Martin K. Petersen,
Johannes Thumshirn, jon.maloy, ying.xue, David Miller, linux-rdma,
LKML, intel-wired-lan, Linux Kernel Network Developers,
linux-scsi, fcoe-devel, tipc-discussion
In-Reply-To: <CAM_iQpX0nA10tCzfZr_q5i6V2hWqfMUkvXdfo+Ro1TWHOEVi_A@mail.gmail.com>
On Wed, Dec 7, 2016 at 1:02 PM, Cong Wang <xiyou.wangcong@gmail.com> wrote:
> On Mon, Dec 5, 2016 at 11:10 PM, Zhouyi Zhou <zhouzhouyi@gmail.com> wrote:
>> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
>> index 2a653ec..ab787cb 100644
>> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
>> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
>> @@ -490,7 +490,11 @@ int ixgbe_fcoe_ddp(struct ixgbe_adapter *adapter,
>> */
>> if ((fh->fh_r_ctl == FC_RCTL_DD_SOL_DATA) &&
>> (fctl & FC_FC_END_SEQ)) {
>> - skb_linearize(skb);
>> + int err = 0;
>> +
>> + err = skb_linearize(skb);
>> + if (err)
>> + return err;
>
>
> You can reuse 'rc' instead of adding 'err'.
rc here is meaningful for the length of data being ddped. If using rc
here, a successful
skb_linearize will assign rc to 0.
>
>
>
>> crc = (struct fcoe_crc_eof *)skb_put(skb, sizeof(*crc));
>> crc->fcoe_eof = FC_EOF_T;
>> }
>> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
>> index fee1f29..4926d48 100644
>> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
>> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
>> @@ -2173,8 +2173,7 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
>> total_rx_bytes += ddp_bytes;
>> total_rx_packets += DIV_ROUND_UP(ddp_bytes,
>> mss);
>> - }
>> - if (!ddp_bytes) {
>> + } else {
>> dev_kfree_skb_any(skb);
>> continue;
>> }
>
>
> This piece doesn't seem to be related.
if ddp_bytes is negative there will be some error, I think the skb
should not pass to upper layer.
^ permalink raw reply
* Re: [PATCH] net: return value of skb_linearize should be handled in Linux kernel
From: Eric Dumazet @ 2016-12-07 6:51 UTC (permalink / raw)
To: Zhouyi Zhou
Cc: faisal.latif, dledford, sean.hefty, hal.rosenstock,
jeffrey.t.kirsher, QLogic-Storage-Upstream, jejb, martin.petersen,
jth, jon.maloy, ying.xue, davem, linux-rdma, linux-kernel,
intel-wired-lan, netdev, linux-scsi, fcoe-devel, tipc-discussion
In-Reply-To: <1481008233-16777-1-git-send-email-zhouzhouyi@gmail.com>
On Tue, 2016-12-06 at 15:10 +0800, Zhouyi Zhou wrote:
> kmalloc_reserve may fail to allocate memory inside skb_linearize,
> which means skb_linearize's return value should not be ignored.
> Following patch correct the uses of skb_linearize.
>
> Compiled in x86_64
>
> Signed-off-by: Zhouyi Zhou <zhouzhouyi@gmail.com>
> ---
> drivers/infiniband/hw/nes/nes_nic.c | 5 +++--
> drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c | 6 +++++-
> drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 3 +--
> drivers/scsi/bnx2fc/bnx2fc_fcoe.c | 7 +++++--
> drivers/scsi/fcoe/fcoe.c | 5 ++++-
> net/tipc/link.c | 3 ++-
> net/tipc/name_distr.c | 5 ++++-
> 7 files changed, 24 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c
> index 2b27d13..69372ea 100644
> --- a/drivers/infiniband/hw/nes/nes_nic.c
> +++ b/drivers/infiniband/hw/nes/nes_nic.c
> @@ -662,10 +662,11 @@ static int nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev)
> nesnic->sq_head &= nesnic->sq_size-1;
> }
> } else {
> - nesvnic->linearized_skbs++;
> hoffset = skb_transport_header(skb) - skb->data;
> nhoffset = skb_network_header(skb) - skb->data;
> - skb_linearize(skb);
> + if (skb_linearize(skb))
> + return NETDEV_TX_BUSY;
This would live lock.
Please drop the packet.
You probably should send one patch per driver, to ease code review and
acceptance.
^ permalink raw reply
* Re: [PATCH net-next v2 0/7] bnxt_en: Add interface to support RDMA driver.
From: Christoph Hellwig @ 2016-12-07 7:29 UTC (permalink / raw)
To: Michael Chan
Cc: davem-fT/PcQaiUtIeIZ0/mPfg9Q, netdev-u79uwXL29TY76Z2rM5mHXA,
selvin.xavier-dY08KVG/lbpWk0Htik3J/w,
somnath.kotur-dY08KVG/lbpWk0Htik3J/w,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1481088381-30411-1-git-send-email-michael.chan-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
On Wed, Dec 07, 2016 at 12:26:14AM -0500, Michael Chan wrote:
> This series adds an interface to support a brand new RDMA driver bnxt_re.
> The first step is to re-arrange some code so that pci_enable_msix() can
> be called during pci probe. The purpose is to allow the RDMA driver to
> initialize and stay initialized whether the netdev is up or down.
Please switch form pci_enable_msix to pci_alloc_irq_vectors for any
changes to MSI-X code, thanks!
--
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
* [PATCH 1/1] infiniband: nes: return value of skb_linearize should be handled
From: Zhouyi Zhou @ 2016-12-07 7:30 UTC (permalink / raw)
To: faisal.latif, dledford, sean.hefty, hal.rosenstock, eric.dumazet,
linux-rdma, linux-kernel
Cc: Zhouyi Zhou, Zhouyi Zhou
Return value of skb_linearize should be handled in function
nes_netdev_start_xmit.
Compiled in x86_64
Signed-off-by: Zhouyi Zhou <yizhouzhou@ict.ac.cn>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
---
drivers/infiniband/hw/nes/nes_nic.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c
index 2b27d13..dfd1f57 100644
--- a/drivers/infiniband/hw/nes/nes_nic.c
+++ b/drivers/infiniband/hw/nes/nes_nic.c
@@ -662,10 +662,14 @@ static int nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev)
nesnic->sq_head &= nesnic->sq_size-1;
}
} else {
- nesvnic->linearized_skbs++;
hoffset = skb_transport_header(skb) - skb->data;
nhoffset = skb_network_header(skb) - skb->data;
- skb_linearize(skb);
+ if (skb_linearize(skb)) {
+ nesvnic->tx_sw_dropped++;
+ kfree_skb(skb);
+ return NETDEV_TX_OK;
+ }
+ nesvnic->linearized_skbs++;
skb_set_transport_header(skb, hoffset);
skb_set_network_header(skb, nhoffset);
if (!nes_nic_send(skb, netdev))
--
1.9.1
^ permalink raw reply related
* Re: [PATCH 1/5] mlx5: Use NULL instead of 0 to represent a pointer
From: Leon Romanovsky @ 2016-12-07 8:21 UTC (permalink / raw)
To: Bart Van Assche
Cc: Doug Ledford, Eli Cohen,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <a8d3dbb7-aafa-c5af-5d06-78d1f4914c5d-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 1865 bytes --]
On Tue, Dec 06, 2016 at 08:29:56AM -0800, Bart Van Assche wrote:
> On 12/06/2016 05:59 AM, Leon Romanovsky wrote:
> > On Mon, Dec 05, 2016 at 05:18:08PM -0800, Bart Van Assche wrote:
> > > Detected by sparse.
> > >
> > > Signed-off-by: Bart Van Assche <bart.vanassche-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
> > > Cc: Eli Cohen <eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > > ---
> > > drivers/infiniband/hw/mlx5/main.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
> > > index 32b09f059c84..abd200e3e299 100644
> > > --- a/drivers/infiniband/hw/mlx5/main.c
> > > +++ b/drivers/infiniband/hw/mlx5/main.c
> > > @@ -127,7 +127,7 @@ static int mlx5_netdev_event(struct notifier_block *this,
> > >
> > > if ((upper == ndev || (!upper && ndev == ibdev->roce.netdev))
> > > && ibdev->ib_active) {
> > > - struct ib_event ibev = {0};
> > > + struct ib_event ibev = { NULL };
> >
> > I afraid that it is sparse anomality and because NULL==0, the ibev.event
> > will be initialized to zero, but it is a matter of time when the sparse
> > will complain about wrong initialization again.
>
> Hello Leon,
>
> The first member of struct ib_event is a pointer so I think the sparse
> complaint is correct.
Yes, sparse didn't iterate all fields of this struct. This is why I
called it "sparse anomality".
> Anyway, how about one of the following two
> alternatives:
> * Change {0} into { }.
Yes, please.
Thanks
> * Use memset() instead of an initializer.
>
> Bart.
> --
> 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
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply
* Re: I/O error on dd commands
From: Max Gurtovoy @ 2016-12-07 10:47 UTC (permalink / raw)
To: Tomita.Haruo-IGagC74glE2s6Rmoc/2Z03gSJqDPrsil,
monis-VPRAkNaXOzVWk0Htik3J/w
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <0980d1e6e7d34e098900ee293aa0e487-DqNMWkYM789gYsFYm0uEO7jjLBE8jN/0@public.gmane.org>
Any interesting logs in dmesg ?
On 12/6/2016 7:55 AM, Tomita.Haruo-IGagC74glE2s6Rmoc/2Z03gSJqDPrsil@public.gmane.org wrote:
> Hi Max,
>
> Thank you for your reply.
> This issue reproduced vanilla 4.9-rc8.
>
>> can you try to repro it with iSER ?
>
> I'm sorry. I can't try iSER.
>
>> what is your backing store device ?
>
> I do now using TOSHIBA Enterprise SSD PX04PMB160.
> https://toshiba.semicon-storage.com/ap-en/product/storage-products/enterprise-ssd.html
>
>> is this happens in 1k bs only or in different bs as well ?
>
> This issue is confirming the reproduction in 512k/1024k/2048k/4096k/8192k bs.
> It occurs by more than one machine, so it isn't a failed of a NIC.
> LAN is a connection directly (It isn't a failed of a hub.)
>
>> rxe_req.c | 9 +++++----
>> 1 file changed, 5 insertions(+), 4 deletions(-)
>> --- linux-4.9-rc7/drivers/infiniband/sw/rxe/rxe_req.c.orig 2016-12-05 10:11:38.000000000 +0900
>> +++ linux-4.9-rc7/drivers/infiniband/sw/rxe/rxe_req.c 2016-12-05 10:15:43.000000000 +0900
>> @@ -705,12 +705,12 @@ next_wqe:
>> skb = init_req_packet(qp, wqe, opcode, payload, &pkt);
>> if (unlikely(!skb)) {
>> pr_err("qp#%d Failed allocating skb\n", qp_num(qp));
>> - goto err;
>> + goto err1;
>> }
>>
>> if (fill_packet(qp, wqe, &pkt, skb, payload)) {
>> pr_debug("qp#%d Error during fill packet\n", qp_num(qp));
>> - goto err;
>> + goto err2;
>> }
>>
>> /*
>> @@ -734,15 +734,16 @@ next_wqe:
>> goto exit;
>> }
>>
>> - goto err;
>> + goto err1;
>> }
>>
>> update_state(qp, wqe, &pkt, payload);
>>
>> goto next_wqe;
>>
>> -err:
>> +err2:
>> kfree_skb(skb);
>> +err1:
>> wqe->status = IB_WC_LOC_PROT_ERR;
>> wqe->state = wqe_state_error;
>>
>
> It's unrelated to this issue, please apply this patch.
>
> thanks,
> Haruo.
>
--
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
* Re: [PATCH 00/28] Broadcom RoCE Driver (bnxt_re)
From: Selvin Xavier @ 2016-12-07 11:33 UTC (permalink / raw)
To: Jason Gunthorpe; +Cc: Doug Ledford, linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20161206163456.GB28066-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
On Tue, Dec 6, 2016 at 10:04 PM, Jason Gunthorpe
<jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> wrote:
> On Tue, Dec 06, 2016 at 11:31:14AM +0530, Selvin Xavier wrote:
>> > Do you have a git tree someplace for this?
>
>> I assume you are referring to our uverbs library.
>
> No, I mean this series for linux. It is very hard to read these
> patches when they touch the same file again and again and have
> dependencies on net trees and things. I encourage people to include a
> git tree for such large works..
I will send out our repository info along with the v2 patch set.
>
>> > Do you have the user space component ready for rdma-core as well?
>> >
>> This is not ready yet. Currently working on it.
>
> I'm also encouraging people to have their user space ready so the
> kernel driver can be properly reviewed.
The user space library is also ready. We would be posting it out soon.
>
> Jason
--
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
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