[PATCH AUTOSEL 6.12 116/486] net/smc: use the correct ndev to find pnetid by pnetid table

[PATCH AUTOSEL 6.12 116/486] net/smc: use the correct ndev to find pnetid by pnetid table

[Sasha Levin]

From: Guangguan Wang <guangguan.wang@linux.alibaba.com>

[ Upstream commit bfc6c67ec2d64d0ca4e5cc3e1ac84298a10b8d62 ]

When using smc_pnet in SMC, it will only search the pnetid in the
base_ndev of the netdev hierarchy(both HW PNETID and User-defined
sw pnetid). This may not work for some scenarios when using SMC in
container on cloud environment.
In container, there have choices of different container network,
such as directly using host network, virtual network IPVLAN, veth,
etc. Different choices of container network have different netdev
hierarchy. Examples of netdev hierarchy show below. (eth0 and eth1
in host below is the netdev directly related to the physical device).
            _______________________________
           |   _________________           |
           |  |POD              |          |
           |  |                 |          |
           |  | eth0_________   |          |
           |  |____|         |__|          |
           |       |         |             |
           |       |         |             |
           |   eth1|base_ndev| eth0_______ |
           |       |         |    | RDMA  ||
           | host  |_________|    |_______||
           ---------------------------------
     netdev hierarchy if directly using host network
           ________________________________
           |   _________________           |
           |  |POD  __________  |          |
           |  |    |upper_ndev| |          |
           |  |eth0|__________| |          |
           |  |_______|_________|          |
           |          |lower netdev        |
           |        __|______              |
           |   eth1|         | eth0_______ |
           |       |base_ndev|    | RDMA  ||
           | host  |_________|    |_______||
           ---------------------------------
            netdev hierarchy if using IPVLAN
            _______________________________
           |   _____________________       |
           |  |POD        _________ |      |
           |  |          |base_ndev||      |
           |  |eth0(veth)|_________||      |
           |  |____________|________|      |
           |               |pairs          |
           |        _______|_              |
           |       |         | eth0_______ |
           |   veth|base_ndev|    | RDMA  ||
           |       |_________|    |_______||
           |        _________              |
           |   eth1|base_ndev|             |
           | host  |_________|             |
           ---------------------------------
             netdev hierarchy if using veth
Due to some reasons, the eth1 in host is not RDMA attached netdevice,
pnetid is needed to map the eth1(in host) with RDMA device so that POD
can do SMC-R. Because the eth1(in host) is managed by CNI plugin(such
as Terway, network management plugin in container environment), and in
cloud environment the eth(in host) can dynamically be inserted by CNI
when POD create and dynamically be removed by CNI when POD destroy and
no POD related to the eth(in host) anymore. It is hard to config the
pnetid to the eth1(in host). But it is easy to config the pnetid to the
netdevice which can be seen in POD. When do SMC-R, both the container
directly using host network and the container using veth network can
successfully match the RDMA device, because the configured pnetid netdev
is a base_ndev. But the container using IPVLAN can not successfully
match the RDMA device and 0x03030000 fallback happens, because the
configured pnetid netdev is not a base_ndev. Additionally, if config
pnetid to the eth1(in host) also can not work for matching RDMA device
when using veth network and doing SMC-R in POD.

To resolve the problems list above, this patch extends to search user
-defined sw pnetid in the clc handshake ndev when no pnetid can be found
in the base_ndev, and the base_ndev take precedence over ndev for backward
compatibility. This patch also can unify the pnetid setup of different
network choices list above in container(Config user-defined sw pnetid in
the netdevice can be seen in POD).

Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/smc/smc_pnet.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/smc/smc_pnet.c b/net/smc/smc_pnet.c
index 716808f374a8d..b391c2ef463f2 100644
--- a/net/smc/smc_pnet.c
+++ b/net/smc/smc_pnet.c
@@ -1079,14 +1079,16 @@ static void smc_pnet_find_roce_by_pnetid(struct net_device *ndev,
 					 struct smc_init_info *ini)
 {
 	u8 ndev_pnetid[SMC_MAX_PNETID_LEN];
+	struct net_device *base_ndev;
 	struct net *net;
 
-	ndev = pnet_find_base_ndev(ndev);
+	base_ndev = pnet_find_base_ndev(ndev);
 	net = dev_net(ndev);
-	if (smc_pnetid_by_dev_port(ndev->dev.parent, ndev->dev_port,
+	if (smc_pnetid_by_dev_port(base_ndev->dev.parent, base_ndev->dev_port,
 				   ndev_pnetid) &&
+	    smc_pnet_find_ndev_pnetid_by_table(base_ndev, ndev_pnetid) &&
 	    smc_pnet_find_ndev_pnetid_by_table(ndev, ndev_pnetid)) {
-		smc_pnet_find_rdma_dev(ndev, ini);
+		smc_pnet_find_rdma_dev(base_ndev, ini);
 		return; /* pnetid could not be determined */
 	}
 	_smc_pnet_find_roce_by_pnetid(ndev_pnetid, ini, NULL, net);
-- 
2.39.5

[PATCH AUTOSEL 6.12 128/486] RDMA/uverbs: Propagate errors from rdma_lookup_get_uobject()

[Sasha Levin]

From: Maher Sanalla <msanalla@nvidia.com>

[ Upstream commit 81f8f7454ad9e0bf95efdec6542afdc9a6ab1e24 ]

Currently, the IB uverbs API calls uobj_get_uobj_read(), which in turn
uses the rdma_lookup_get_uobject() helper to retrieve user objects.
In case of failure, uobj_get_uobj_read() returns NULL, overriding the
error code from rdma_lookup_get_uobject(). The IB uverbs API then
translates this NULL to -EINVAL, masking the actual error and
complicating debugging. For example, applications calling ibv_modify_qp
that fails with EBUSY when retrieving the QP uobject will see the
overridden error code EINVAL instead, masking the actual error.

Furthermore, based on rdma-core commit:
"2a22f1ced5f3 ("Merge pull request #1568 from jakemoroni/master")"
Kernel's IB uverbs return values are either ignored and passed on as is
to application or overridden with other errnos in a few cases.

Thus, to improve error reporting and debuggability, propagate the
original error from rdma_lookup_get_uobject() instead of replacing it
with EINVAL.

Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://patch.msgid.link/64f9d3711b183984e939962c2f83383904f97dfb.1740577869.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/core/uverbs_cmd.c | 144 ++++++++++++++-------------
 include/rdma/uverbs_std_types.h      |   2 +-
 2 files changed, 77 insertions(+), 69 deletions(-)

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index edef79daed3fa..535bb99ed9f5f 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -718,8 +718,8 @@ static int ib_uverbs_reg_mr(struct uverbs_attr_bundle *attrs)
 		goto err_free;
 
 	pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd.pd_handle, attrs);
-	if (!pd) {
-		ret = -EINVAL;
+	if (IS_ERR(pd)) {
+		ret = PTR_ERR(pd);
 		goto err_free;
 	}
 
@@ -809,8 +809,8 @@ static int ib_uverbs_rereg_mr(struct uverbs_attr_bundle *attrs)
 	if (cmd.flags & IB_MR_REREG_PD) {
 		new_pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd.pd_handle,
 					   attrs);
-		if (!new_pd) {
-			ret = -EINVAL;
+		if (IS_ERR(new_pd)) {
+			ret = PTR_ERR(new_pd);
 			goto put_uobjs;
 		}
 	} else {
@@ -919,8 +919,8 @@ static int ib_uverbs_alloc_mw(struct uverbs_attr_bundle *attrs)
 		return PTR_ERR(uobj);
 
 	pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd.pd_handle, attrs);
-	if (!pd) {
-		ret = -EINVAL;
+	if (IS_ERR(pd)) {
+		ret = PTR_ERR(pd);
 		goto err_free;
 	}
 
@@ -1127,8 +1127,8 @@ static int ib_uverbs_resize_cq(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ, cmd.cq_handle, attrs);
-	if (!cq)
-		return -EINVAL;
+	if (IS_ERR(cq))
+		return PTR_ERR(cq);
 
 	ret = cq->device->ops.resize_cq(cq, cmd.cqe, &attrs->driver_udata);
 	if (ret)
@@ -1189,8 +1189,8 @@ static int ib_uverbs_poll_cq(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ, cmd.cq_handle, attrs);
-	if (!cq)
-		return -EINVAL;
+	if (IS_ERR(cq))
+		return PTR_ERR(cq);
 
 	/* we copy a struct ib_uverbs_poll_cq_resp to user space */
 	header_ptr = attrs->ucore.outbuf;
@@ -1238,8 +1238,8 @@ static int ib_uverbs_req_notify_cq(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ, cmd.cq_handle, attrs);
-	if (!cq)
-		return -EINVAL;
+	if (IS_ERR(cq))
+		return PTR_ERR(cq);
 
 	ib_req_notify_cq(cq, cmd.solicited_only ?
 			 IB_CQ_SOLICITED : IB_CQ_NEXT_COMP);
@@ -1321,8 +1321,8 @@ static int create_qp(struct uverbs_attr_bundle *attrs,
 		ind_tbl = uobj_get_obj_read(rwq_ind_table,
 					    UVERBS_OBJECT_RWQ_IND_TBL,
 					    cmd->rwq_ind_tbl_handle, attrs);
-		if (!ind_tbl) {
-			ret = -EINVAL;
+		if (IS_ERR(ind_tbl)) {
+			ret = PTR_ERR(ind_tbl);
 			goto err_put;
 		}
 
@@ -1360,8 +1360,10 @@ static int create_qp(struct uverbs_attr_bundle *attrs,
 			if (cmd->is_srq) {
 				srq = uobj_get_obj_read(srq, UVERBS_OBJECT_SRQ,
 							cmd->srq_handle, attrs);
-				if (!srq || srq->srq_type == IB_SRQT_XRC) {
-					ret = -EINVAL;
+				if (IS_ERR(srq) ||
+				    srq->srq_type == IB_SRQT_XRC) {
+					ret = IS_ERR(srq) ? PTR_ERR(srq) :
+								  -EINVAL;
 					goto err_put;
 				}
 			}
@@ -1371,23 +1373,29 @@ static int create_qp(struct uverbs_attr_bundle *attrs,
 					rcq = uobj_get_obj_read(
 						cq, UVERBS_OBJECT_CQ,
 						cmd->recv_cq_handle, attrs);
-					if (!rcq) {
-						ret = -EINVAL;
+					if (IS_ERR(rcq)) {
+						ret = PTR_ERR(rcq);
 						goto err_put;
 					}
 				}
 			}
 		}
 
-		if (has_sq)
+		if (has_sq) {
 			scq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ,
 						cmd->send_cq_handle, attrs);
+			if (IS_ERR(scq)) {
+				ret = PTR_ERR(scq);
+				goto err_put;
+			}
+		}
+
 		if (!ind_tbl && cmd->qp_type != IB_QPT_XRC_INI)
 			rcq = rcq ?: scq;
 		pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd->pd_handle,
 				       attrs);
-		if (!pd || (!scq && has_sq)) {
-			ret = -EINVAL;
+		if (IS_ERR(pd)) {
+			ret = PTR_ERR(pd);
 			goto err_put;
 		}
 
@@ -1482,18 +1490,18 @@ static int create_qp(struct uverbs_attr_bundle *attrs,
 err_put:
 	if (!IS_ERR(xrcd_uobj))
 		uobj_put_read(xrcd_uobj);
-	if (pd)
+	if (!IS_ERR_OR_NULL(pd))
 		uobj_put_obj_read(pd);
-	if (scq)
+	if (!IS_ERR_OR_NULL(scq))
 		rdma_lookup_put_uobject(&scq->uobject->uevent.uobject,
 					UVERBS_LOOKUP_READ);
-	if (rcq && rcq != scq)
+	if (!IS_ERR_OR_NULL(rcq) && rcq != scq)
 		rdma_lookup_put_uobject(&rcq->uobject->uevent.uobject,
 					UVERBS_LOOKUP_READ);
-	if (srq)
+	if (!IS_ERR_OR_NULL(srq))
 		rdma_lookup_put_uobject(&srq->uobject->uevent.uobject,
 					UVERBS_LOOKUP_READ);
-	if (ind_tbl)
+	if (!IS_ERR_OR_NULL(ind_tbl))
 		uobj_put_obj_read(ind_tbl);
 
 	uobj_alloc_abort(&obj->uevent.uobject, attrs);
@@ -1655,8 +1663,8 @@ static int ib_uverbs_query_qp(struct uverbs_attr_bundle *attrs)
 	}
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp) {
-		ret = -EINVAL;
+	if (IS_ERR(qp)) {
+		ret = PTR_ERR(qp);
 		goto out;
 	}
 
@@ -1761,8 +1769,8 @@ static int modify_qp(struct uverbs_attr_bundle *attrs,
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd->base.qp_handle,
 			       attrs);
-	if (!qp) {
-		ret = -EINVAL;
+	if (IS_ERR(qp)) {
+		ret = PTR_ERR(qp);
 		goto out;
 	}
 
@@ -2028,8 +2036,8 @@ static int ib_uverbs_post_send(struct uverbs_attr_bundle *attrs)
 		return -ENOMEM;
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp) {
-		ret = -EINVAL;
+	if (IS_ERR(qp)) {
+		ret = PTR_ERR(qp);
 		goto out;
 	}
 
@@ -2066,9 +2074,9 @@ static int ib_uverbs_post_send(struct uverbs_attr_bundle *attrs)
 
 			ud->ah = uobj_get_obj_read(ah, UVERBS_OBJECT_AH,
 						   user_wr->wr.ud.ah, attrs);
-			if (!ud->ah) {
+			if (IS_ERR(ud->ah)) {
+				ret = PTR_ERR(ud->ah);
 				kfree(ud);
-				ret = -EINVAL;
 				goto out_put;
 			}
 			ud->remote_qpn = user_wr->wr.ud.remote_qpn;
@@ -2305,8 +2313,8 @@ static int ib_uverbs_post_recv(struct uverbs_attr_bundle *attrs)
 		return PTR_ERR(wr);
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp) {
-		ret = -EINVAL;
+	if (IS_ERR(qp)) {
+		ret = PTR_ERR(qp);
 		goto out;
 	}
 
@@ -2356,8 +2364,8 @@ static int ib_uverbs_post_srq_recv(struct uverbs_attr_bundle *attrs)
 		return PTR_ERR(wr);
 
 	srq = uobj_get_obj_read(srq, UVERBS_OBJECT_SRQ, cmd.srq_handle, attrs);
-	if (!srq) {
-		ret = -EINVAL;
+	if (IS_ERR(srq)) {
+		ret = PTR_ERR(srq);
 		goto out;
 	}
 
@@ -2413,8 +2421,8 @@ static int ib_uverbs_create_ah(struct uverbs_attr_bundle *attrs)
 	}
 
 	pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd.pd_handle, attrs);
-	if (!pd) {
-		ret = -EINVAL;
+	if (IS_ERR(pd)) {
+		ret = PTR_ERR(pd);
 		goto err;
 	}
 
@@ -2483,8 +2491,8 @@ static int ib_uverbs_attach_mcast(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp)
-		return -EINVAL;
+	if (IS_ERR(qp))
+		return PTR_ERR(qp);
 
 	obj = qp->uobject;
 
@@ -2533,8 +2541,8 @@ static int ib_uverbs_detach_mcast(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp)
-		return -EINVAL;
+	if (IS_ERR(qp))
+		return PTR_ERR(qp);
 
 	obj = qp->uobject;
 	mutex_lock(&obj->mcast_lock);
@@ -2668,8 +2676,8 @@ static int kern_spec_to_ib_spec_action(struct uverbs_attr_bundle *attrs,
 							UVERBS_OBJECT_FLOW_ACTION,
 							kern_spec->action.handle,
 							attrs);
-		if (!ib_spec->action.act)
-			return -EINVAL;
+		if (IS_ERR(ib_spec->action.act))
+			return PTR_ERR(ib_spec->action.act);
 		ib_spec->action.size =
 			sizeof(struct ib_flow_spec_action_handle);
 		flow_resources_add(uflow_res,
@@ -2686,8 +2694,8 @@ static int kern_spec_to_ib_spec_action(struct uverbs_attr_bundle *attrs,
 					  UVERBS_OBJECT_COUNTERS,
 					  kern_spec->flow_count.handle,
 					  attrs);
-		if (!ib_spec->flow_count.counters)
-			return -EINVAL;
+		if (IS_ERR(ib_spec->flow_count.counters))
+			return PTR_ERR(ib_spec->flow_count.counters);
 		ib_spec->flow_count.size =
 				sizeof(struct ib_flow_spec_action_count);
 		flow_resources_add(uflow_res,
@@ -2905,14 +2913,14 @@ static int ib_uverbs_ex_create_wq(struct uverbs_attr_bundle *attrs)
 		return PTR_ERR(obj);
 
 	pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd.pd_handle, attrs);
-	if (!pd) {
-		err = -EINVAL;
+	if (IS_ERR(pd)) {
+		err = PTR_ERR(pd);
 		goto err_uobj;
 	}
 
 	cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ, cmd.cq_handle, attrs);
-	if (!cq) {
-		err = -EINVAL;
+	if (IS_ERR(cq)) {
+		err = PTR_ERR(cq);
 		goto err_put_pd;
 	}
 
@@ -3013,8 +3021,8 @@ static int ib_uverbs_ex_modify_wq(struct uverbs_attr_bundle *attrs)
 		return -EINVAL;
 
 	wq = uobj_get_obj_read(wq, UVERBS_OBJECT_WQ, cmd.wq_handle, attrs);
-	if (!wq)
-		return -EINVAL;
+	if (IS_ERR(wq))
+		return PTR_ERR(wq);
 
 	if (cmd.attr_mask & IB_WQ_FLAGS) {
 		wq_attr.flags = cmd.flags;
@@ -3097,8 +3105,8 @@ static int ib_uverbs_ex_create_rwq_ind_table(struct uverbs_attr_bundle *attrs)
 			num_read_wqs++) {
 		wq = uobj_get_obj_read(wq, UVERBS_OBJECT_WQ,
 				       wqs_handles[num_read_wqs], attrs);
-		if (!wq) {
-			err = -EINVAL;
+		if (IS_ERR(wq)) {
+			err = PTR_ERR(wq);
 			goto put_wqs;
 		}
 
@@ -3253,8 +3261,8 @@ static int ib_uverbs_ex_create_flow(struct uverbs_attr_bundle *attrs)
 	}
 
 	qp = uobj_get_obj_read(qp, UVERBS_OBJECT_QP, cmd.qp_handle, attrs);
-	if (!qp) {
-		err = -EINVAL;
+	if (IS_ERR(qp)) {
+		err = PTR_ERR(qp);
 		goto err_uobj;
 	}
 
@@ -3400,15 +3408,15 @@ static int __uverbs_create_xsrq(struct uverbs_attr_bundle *attrs,
 	if (ib_srq_has_cq(cmd->srq_type)) {
 		attr.ext.cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ,
 						cmd->cq_handle, attrs);
-		if (!attr.ext.cq) {
-			ret = -EINVAL;
+		if (IS_ERR(attr.ext.cq)) {
+			ret = PTR_ERR(attr.ext.cq);
 			goto err_put_xrcd;
 		}
 	}
 
 	pd = uobj_get_obj_read(pd, UVERBS_OBJECT_PD, cmd->pd_handle, attrs);
-	if (!pd) {
-		ret = -EINVAL;
+	if (IS_ERR(pd)) {
+		ret = PTR_ERR(pd);
 		goto err_put_cq;
 	}
 
@@ -3515,8 +3523,8 @@ static int ib_uverbs_modify_srq(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	srq = uobj_get_obj_read(srq, UVERBS_OBJECT_SRQ, cmd.srq_handle, attrs);
-	if (!srq)
-		return -EINVAL;
+	if (IS_ERR(srq))
+		return PTR_ERR(srq);
 
 	attr.max_wr    = cmd.max_wr;
 	attr.srq_limit = cmd.srq_limit;
@@ -3543,8 +3551,8 @@ static int ib_uverbs_query_srq(struct uverbs_attr_bundle *attrs)
 		return ret;
 
 	srq = uobj_get_obj_read(srq, UVERBS_OBJECT_SRQ, cmd.srq_handle, attrs);
-	if (!srq)
-		return -EINVAL;
+	if (IS_ERR(srq))
+		return PTR_ERR(srq);
 
 	ret = ib_query_srq(srq, &attr);
 
@@ -3669,8 +3677,8 @@ static int ib_uverbs_ex_modify_cq(struct uverbs_attr_bundle *attrs)
 		return -EOPNOTSUPP;
 
 	cq = uobj_get_obj_read(cq, UVERBS_OBJECT_CQ, cmd.cq_handle, attrs);
-	if (!cq)
-		return -EINVAL;
+	if (IS_ERR(cq))
+		return PTR_ERR(cq);
 
 	ret = rdma_set_cq_moderation(cq, cmd.attr.cq_count, cmd.attr.cq_period);
 
diff --git a/include/rdma/uverbs_std_types.h b/include/rdma/uverbs_std_types.h
index fe05121169589..555ea3d142a46 100644
--- a/include/rdma/uverbs_std_types.h
+++ b/include/rdma/uverbs_std_types.h
@@ -34,7 +34,7 @@
 static inline void *_uobj_get_obj_read(struct ib_uobject *uobj)
 {
 	if (IS_ERR(uobj))
-		return NULL;
+		return ERR_CAST(uobj);
 	return uobj->object;
 }
 #define uobj_get_obj_read(_object, _type, _id, _attrs)                         \
-- 
2.39.5

[PATCH AUTOSEL 6.12 238/486] net/mlx5: Avoid report two health errors on same syndrome

[Sasha Levin]

From: Moshe Shemesh <moshe@nvidia.com>

[ Upstream commit b5d7b2f04ebcff740f44ef4d295b3401aeb029f4 ]

In case health counter has not increased for few polling intervals, miss
counter will reach max misses threshold and health report will be
triggered for FW health reporter. In case syndrome found on same health
poll another health report will be triggered.

Avoid two health reports on same syndrome by marking this syndrome as
already known.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shahar Shitrit <shshitrit@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/health.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index a6329ca2d9bff..52c8035547be5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -799,6 +799,7 @@ static void poll_health(struct timer_list *t)
 	health->prev = count;
 	if (health->miss_counter == MAX_MISSES) {
 		mlx5_core_err(dev, "device's health compromised - reached miss count\n");
+		health->synd = ioread8(&h->synd);
 		print_health_info(dev);
 		queue_work(health->wq, &health->report_work);
 	}
-- 
2.39.5

[PATCH AUTOSEL 6.12 263/486] net/mlx5e: Add correct match to check IPSec syndromes for switchdev mode

[Sasha Levin]

From: Jianbo Liu <jianbol@nvidia.com>

[ Upstream commit 85e4a808af2545fefaf18c8fe50071b06fcbdabc ]

In commit dddb49b63d86 ("net/mlx5e: Add IPsec and ASO syndromes check
in HW"), IPSec and ASO syndromes checks after decryption for the
specified ASO object were added. But they are correct only for eswith
in legacy mode. For switchdev mode, metadata register c1 is used to
save the mapped id (not ASO object id). So, need to change the match
accordingly for the check rules in status table.

Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Patrisious Haddad <phaddad@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250220213959.504304-4-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../mellanox/mlx5/core/en_accel/ipsec_fs.c    | 28 ++++++++++++++-----
 .../mellanox/mlx5/core/esw/ipsec_fs.c         | 13 +++++++++
 .../mellanox/mlx5/core/esw/ipsec_fs.h         |  5 ++++
 include/linux/mlx5/eswitch.h                  |  2 ++
 4 files changed, 41 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
index 57861d34d46f8..59b9653f573c8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c
@@ -165,6 +165,25 @@ static void ipsec_rx_status_pass_destroy(struct mlx5e_ipsec *ipsec,
 #endif
 }
 
+static void ipsec_rx_rule_add_match_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
+					struct mlx5e_ipsec_rx *rx,
+					struct mlx5_flow_spec *spec)
+{
+	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
+
+	if (rx == ipsec->rx_esw) {
+		mlx5_esw_ipsec_rx_rule_add_match_obj(sa_entry, spec);
+	} else {
+		MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria,
+				 misc_parameters_2.metadata_reg_c_2);
+		MLX5_SET(fte_match_param, spec->match_value,
+			 misc_parameters_2.metadata_reg_c_2,
+			 sa_entry->ipsec_obj_id | BIT(31));
+
+		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2;
+	}
+}
+
 static int rx_add_rule_drop_auth_trailer(struct mlx5e_ipsec_sa_entry *sa_entry,
 					 struct mlx5e_ipsec_rx *rx)
 {
@@ -200,11 +219,8 @@ static int rx_add_rule_drop_auth_trailer(struct mlx5e_ipsec_sa_entry *sa_entry,
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.ipsec_syndrome);
 	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.ipsec_syndrome, 1);
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
-	MLX5_SET(fte_match_param, spec->match_value,
-		 misc_parameters_2.metadata_reg_c_2,
-		 sa_entry->ipsec_obj_id | BIT(31));
 	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	ipsec_rx_rule_add_match_obj(sa_entry, rx, spec);
 	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
@@ -281,10 +297,8 @@ static int rx_add_rule_drop_replay(struct mlx5e_ipsec_sa_entry *sa_entry, struct
 
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_4);
 	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_c_4, 1);
-	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_c_2);
-	MLX5_SET(fte_match_param, spec->match_value,  misc_parameters_2.metadata_reg_c_2,
-		 sa_entry->ipsec_obj_id | BIT(31));
 	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
+	ipsec_rx_rule_add_match_obj(sa_entry, rx, spec);
 	rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
 	if (IS_ERR(rule)) {
 		err = PTR_ERR(rule);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
index ed977ae75fab8..4bba2884c1c05 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c
@@ -85,6 +85,19 @@ int mlx5_esw_ipsec_rx_setup_modify_header(struct mlx5e_ipsec_sa_entry *sa_entry,
 	return err;
 }
 
+void mlx5_esw_ipsec_rx_rule_add_match_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
+					  struct mlx5_flow_spec *spec)
+{
+	MLX5_SET(fte_match_param, spec->match_criteria,
+		 misc_parameters_2.metadata_reg_c_1,
+		 ESW_IPSEC_RX_MAPPED_ID_MATCH_MASK);
+	MLX5_SET(fte_match_param, spec->match_value,
+		 misc_parameters_2.metadata_reg_c_1,
+		 sa_entry->rx_mapped_id << ESW_ZONE_ID_BITS);
+
+	spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2;
+}
+
 void mlx5_esw_ipsec_rx_id_mapping_remove(struct mlx5e_ipsec_sa_entry *sa_entry)
 {
 	struct mlx5e_ipsec *ipsec = sa_entry->ipsec;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.h
index ac9c65b89166e..514c15258b1d1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.h
@@ -20,6 +20,8 @@ int mlx5_esw_ipsec_rx_ipsec_obj_id_search(struct mlx5e_priv *priv, u32 id,
 void mlx5_esw_ipsec_tx_create_attr_set(struct mlx5e_ipsec *ipsec,
 				       struct mlx5e_ipsec_tx_create_attr *attr);
 void mlx5_esw_ipsec_restore_dest_uplink(struct mlx5_core_dev *mdev);
+void mlx5_esw_ipsec_rx_rule_add_match_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
+					  struct mlx5_flow_spec *spec);
 #else
 static inline void mlx5_esw_ipsec_rx_create_attr_set(struct mlx5e_ipsec *ipsec,
 						     struct mlx5e_ipsec_rx_create_attr *attr) {}
@@ -48,5 +50,8 @@ static inline void mlx5_esw_ipsec_tx_create_attr_set(struct mlx5e_ipsec *ipsec,
 						     struct mlx5e_ipsec_tx_create_attr *attr) {}
 
 static inline void mlx5_esw_ipsec_restore_dest_uplink(struct mlx5_core_dev *mdev) {}
+static inline void
+mlx5_esw_ipsec_rx_rule_add_match_obj(struct mlx5e_ipsec_sa_entry *sa_entry,
+				     struct mlx5_flow_spec *spec) {}
 #endif /* CONFIG_MLX5_ESWITCH */
 #endif /* __MLX5_ESW_IPSEC_FS_H__ */
diff --git a/include/linux/mlx5/eswitch.h b/include/linux/mlx5/eswitch.h
index df73a2ccc9af3..67256e776566c 100644
--- a/include/linux/mlx5/eswitch.h
+++ b/include/linux/mlx5/eswitch.h
@@ -147,6 +147,8 @@ u32 mlx5_eswitch_get_vport_metadata_for_set(struct mlx5_eswitch *esw,
 
 /* reuse tun_opts for the mapped ipsec obj id when tun_id is 0 (invalid) */
 #define ESW_IPSEC_RX_MAPPED_ID_MASK GENMASK(ESW_TUN_OPTS_BITS - 1, 0)
+#define ESW_IPSEC_RX_MAPPED_ID_MATCH_MASK \
+	GENMASK(31 - ESW_RESERVED_BITS, ESW_ZONE_ID_BITS)
 
 u8 mlx5_eswitch_mode(const struct mlx5_core_dev *dev);
 u16 mlx5_eswitch_get_total_vports(const struct mlx5_core_dev *dev);
-- 
2.39.5

[PATCH AUTOSEL 6.12 270/486] net/mlx5: Change POOL_NEXT_SIZE define value and make it global

[Sasha Levin]

From: Patrisious Haddad <phaddad@nvidia.com>

[ Upstream commit 80df31f384b4146a62a01b3d4beb376cc7b9a89e ]

Change POOL_NEXT_SIZE define value from 0 to BIT(30), since this define
is used to request the available maximum sized flow table, and zero doesn't
make sense for it, whereas some places in the driver use zero explicitly
expecting the smallest table size possible but instead due to this
define they end up allocating the biggest table size unawarely.

In addition move the definition to "include/linux/mlx5/fs.h" to expose the
define to IB driver as well, while appropriately renaming it.

Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250219085808.349923-3-tariqt@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c    | 2 +-
 drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.c    | 6 ++++--
 drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.h    | 2 --
 drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c | 3 ++-
 include/linux/mlx5/fs.h                                 | 2 ++
 5 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
index 8587cd572da53..bdb825aa87268 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/legacy.c
@@ -96,7 +96,7 @@ static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw)
 	if (!flow_group_in)
 		return -ENOMEM;
 
-	ft_attr.max_fte = POOL_NEXT_SIZE;
+	ft_attr.max_fte = MLX5_FS_MAX_POOL_SIZE;
 	ft_attr.prio = LEGACY_FDB_PRIO;
 	fdb = mlx5_create_flow_table(root_ns, &ft_attr);
 	if (IS_ERR(fdb)) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.c
index c14590acc7726..f6abfd00d7e68 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.c
@@ -50,10 +50,12 @@ mlx5_ft_pool_get_avail_sz(struct mlx5_core_dev *dev, enum fs_flow_table_type tab
 	int i, found_i = -1;
 
 	for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) {
-		if (dev->priv.ft_pool->ft_left[i] && FT_POOLS[i] >= desired_size &&
+		if (dev->priv.ft_pool->ft_left[i] &&
+		    (FT_POOLS[i] >= desired_size ||
+		     desired_size == MLX5_FS_MAX_POOL_SIZE) &&
 		    FT_POOLS[i] <= max_ft_size) {
 			found_i = i;
-			if (desired_size != POOL_NEXT_SIZE)
+			if (desired_size != MLX5_FS_MAX_POOL_SIZE)
 				break;
 		}
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.h
index 25f4274b372b5..173e312db7204 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_ft_pool.h
@@ -7,8 +7,6 @@
 #include <linux/mlx5/driver.h>
 #include "fs_core.h"
 
-#define POOL_NEXT_SIZE 0
-
 int mlx5_ft_pool_init(struct mlx5_core_dev *dev);
 void mlx5_ft_pool_destroy(struct mlx5_core_dev *dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
index 711d14dea2485..d313cb7f0ed88 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c
@@ -161,7 +161,8 @@ mlx5_chains_create_table(struct mlx5_fs_chains *chains,
 		ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
 				  MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
 
-	sz = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? FT_TBL_SZ : POOL_NEXT_SIZE;
+	sz = (chain == mlx5_chains_get_nf_ft_chain(chains)) ?
+		FT_TBL_SZ : MLX5_FS_MAX_POOL_SIZE;
 	ft_attr.max_fte = sz;
 
 	/* We use chains_default_ft(chains) as the table's next_ft till
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index b744e554f014d..db5c9ddef1702 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -40,6 +40,8 @@
 
 #define MLX5_SET_CFG(p, f, v) MLX5_SET(create_flow_group_in, p, f, v)
 
+#define MLX5_FS_MAX_POOL_SIZE BIT(30)
+
 enum mlx5_flow_destination_type {
 	MLX5_FLOW_DESTINATION_TYPE_NONE,
 	MLX5_FLOW_DESTINATION_TYPE_VPORT,
-- 
2.39.5

[PATCH AUTOSEL 6.12 307/486] RDMA/core: Fix best page size finding when it can cross SG entries

[Sasha Levin]

From: Michael Margolin <mrgolin@amazon.com>

[ Upstream commit 486055f5e09df959ad4e3aa4ee75b5c91ddeec2e ]

A single scatter-gather entry is limited by a 32 bits "length" field
that is practically 4GB - PAGE_SIZE. This means that even when the
memory is physically contiguous, we might need more than one entry to
represent it. Additionally when using dmabuf, the sg_table might be
originated outside the subsystem and optimized for other needs.

For instance an SGT of 16GB GPU continuous memory might look like this:
(a real life example)

dma_address 34401400000, length fffff000
dma_address 345013ff000, length fffff000
dma_address 346013fe000, length fffff000
dma_address 347013fd000, length fffff000
dma_address 348013fc000, length 4000

Since ib_umem_find_best_pgsz works within SG entries, in the above case
we will result with the worst possible 4KB page size.

Fix this by taking into consideration only the alignment of addresses of
real discontinuity points rather than treating SG entries as such, and
adjust the page iterator to correctly handle cross SG entry pages.

There is currently an assumption that drivers do not ask for pages
bigger than maximal DMA size supported by their devices.

Reviewed-by: Firas Jahjah <firasj@amazon.com>
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20250217141623.12428-1-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/core/umem.c  | 36 ++++++++++++++++++++++++---------
 drivers/infiniband/core/verbs.c | 11 +++++-----
 2 files changed, 32 insertions(+), 15 deletions(-)

diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index 07c571c7b6999..c5b6863947605 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -80,9 +80,12 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
 				     unsigned long pgsz_bitmap,
 				     unsigned long virt)
 {
-	struct scatterlist *sg;
+	unsigned long curr_len = 0;
+	dma_addr_t curr_base = ~0;
 	unsigned long va, pgoff;
+	struct scatterlist *sg;
 	dma_addr_t mask;
+	dma_addr_t end;
 	int i;
 
 	umem->iova = va = virt;
@@ -107,17 +110,30 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
 	pgoff = umem->address & ~PAGE_MASK;
 
 	for_each_sgtable_dma_sg(&umem->sgt_append.sgt, sg, i) {
-		/* Walk SGL and reduce max page size if VA/PA bits differ
-		 * for any address.
+		/* If the current entry is physically contiguous with the previous
+		 * one, no need to take its start addresses into consideration.
 		 */
-		mask |= (sg_dma_address(sg) + pgoff) ^ va;
+		if (check_add_overflow(curr_base, curr_len, &end) ||
+		    end != sg_dma_address(sg)) {
+
+			curr_base = sg_dma_address(sg);
+			curr_len = 0;
+
+			/* Reduce max page size if VA/PA bits differ */
+			mask |= (curr_base + pgoff) ^ va;
+
+			/* The alignment of any VA matching a discontinuity point
+			* in the physical memory sets the maximum possible page
+			* size as this must be a starting point of a new page that
+			* needs to be aligned.
+			*/
+			if (i != 0)
+				mask |= va;
+		}
+
+		curr_len += sg_dma_len(sg);
 		va += sg_dma_len(sg) - pgoff;
-		/* Except for the last entry, the ending iova alignment sets
-		 * the maximum possible page size as the low bits of the iova
-		 * must be zero when starting the next chunk.
-		 */
-		if (i != (umem->sgt_append.sgt.nents - 1))
-			mask |= va;
+
 		pgoff = 0;
 	}
 
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 473ee0831307c..dc40001072a5e 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -3109,22 +3109,23 @@ EXPORT_SYMBOL(__rdma_block_iter_start);
 bool __rdma_block_iter_next(struct ib_block_iter *biter)
 {
 	unsigned int block_offset;
-	unsigned int sg_delta;
+	unsigned int delta;
 
 	if (!biter->__sg_nents || !biter->__sg)
 		return false;
 
 	biter->__dma_addr = sg_dma_address(biter->__sg) + biter->__sg_advance;
 	block_offset = biter->__dma_addr & (BIT_ULL(biter->__pg_bit) - 1);
-	sg_delta = BIT_ULL(biter->__pg_bit) - block_offset;
+	delta = BIT_ULL(biter->__pg_bit) - block_offset;
 
-	if (sg_dma_len(biter->__sg) - biter->__sg_advance > sg_delta) {
-		biter->__sg_advance += sg_delta;
-	} else {
+	while (biter->__sg_nents && biter->__sg &&
+	       sg_dma_len(biter->__sg) - biter->__sg_advance <= delta) {
+		delta -= sg_dma_len(biter->__sg) - biter->__sg_advance;
 		biter->__sg_advance = 0;
 		biter->__sg = sg_next(biter->__sg);
 		biter->__sg_nents--;
 	}
+	biter->__sg_advance += delta;
 
 	return true;
 }
-- 
2.39.5

[PATCH AUTOSEL 6.12 312/486] eth: mlx4: don't try to complete XDP frames in netpoll

[Sasha Levin]

From: Jakub Kicinski <kuba@kernel.org>

[ Upstream commit 8fdeafd66edaf420ea0063a1f13442fe3470fe70 ]

mlx4 doesn't support ndo_xdp_xmit / XDP_REDIRECT and wasn't
using page pool until now, so it could run XDP completions
in netpoll (NAPI budget == 0) just fine. Page pool has calling
context requirements, make sure we don't try to call it from
what is potentially HW IRQ context.

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250213010635.1354034-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_tx.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 1ddb11cb25f91..6e077d202827a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -450,6 +450,8 @@ int mlx4_en_process_tx_cq(struct net_device *dev,
 
 	if (unlikely(!priv->port_up))
 		return 0;
+	if (unlikely(!napi_budget) && cq->type == TX_XDP)
+		return 0;
 
 	netdev_txq_bql_complete_prefetchw(ring->tx_queue);
 
-- 
2.39.5

[PATCH AUTOSEL 6.12 321/486] net/mlx5: Modify LSB bitmask in temperature event to include only the first bit

[Sasha Levin]

From: Shahar Shitrit <shshitrit@nvidia.com>

[ Upstream commit 633f16d7e07c129a36b882c05379e01ce5bdb542 ]

In the sensor_count field of the MTEWE register, bits 1-62 are
supported only for unmanaged switches, not for NICs, and bit 63
is reserved for internal use.

To prevent confusing output that may include set bits that are
not relevant to NIC sensors, we update the bitmask to retain only
the first bit, which corresponds to the sensor ASIC.

Signed-off-by: Shahar Shitrit <shshitrit@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Link: https://patch.msgid.link/20250213094641.226501-4-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/events.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c
index d91ea53eb394d..cd8d107f7d9e3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/events.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c
@@ -163,6 +163,10 @@ static int temp_warn(struct notifier_block *nb, unsigned long type, void *data)
 	u64 value_msb;
 
 	value_lsb = be64_to_cpu(eqe->data.temp_warning.sensor_warning_lsb);
+	/* bit 1-63 are not supported for NICs,
+	 * hence read only bit 0 (asic) from lsb.
+	 */
+	value_lsb &= 0x1;
 	value_msb = be64_to_cpu(eqe->data.temp_warning.sensor_warning_msb);
 
 	mlx5_core_warn(events->dev,
-- 
2.39.5

[PATCH AUTOSEL 6.12 322/486] net/mlx5: Apply rate-limiting to high temperature warning

[Sasha Levin]

From: Shahar Shitrit <shshitrit@nvidia.com>

[ Upstream commit 9dd3d5d258aceb37bdf09c8b91fa448f58ea81f0 ]

Wrap the high temperature warning in a temperature event with
a call to net_ratelimit() to prevent flooding the kernel log
with repeated warning messages when temperature exceeds the
threshold multiple times within a short duration.

Signed-off-by: Shahar Shitrit <shshitrit@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Link: https://patch.msgid.link/20250213094641.226501-2-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/events.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c
index cd8d107f7d9e3..fc6e56305cbbc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/events.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c
@@ -169,9 +169,10 @@ static int temp_warn(struct notifier_block *nb, unsigned long type, void *data)
 	value_lsb &= 0x1;
 	value_msb = be64_to_cpu(eqe->data.temp_warning.sensor_warning_msb);
 
-	mlx5_core_warn(events->dev,
-		       "High temperature on sensors with bit set %llx %llx",
-		       value_msb, value_lsb);
+	if (net_ratelimit())
+		mlx5_core_warn(events->dev,
+			       "High temperature on sensors with bit set %llx %llx",
+			       value_msb, value_lsb);
 
 	return NOTIFY_OK;
 }
-- 
2.39.5

[PATCH AUTOSEL 6.12 342/486] net/mlx4_core: Avoid impossible mlx4_db_alloc() order value

[Sasha Levin]

From: Kees Cook <kees@kernel.org>

[ Upstream commit 4a6f18f28627e121bd1f74b5fcc9f945d6dbeb1e ]

GCC can see that the value range for "order" is capped, but this leads
it to consider that it might be negative, leading to a false positive
warning (with GCC 15 with -Warray-bounds -fdiagnostics-details):

../drivers/net/ethernet/mellanox/mlx4/alloc.c:691:47: error: array subscript -1 is below array bounds of 'long unsigned int *[2]' [-Werror=array-bounds=]
  691 |                 i = find_first_bit(pgdir->bits[o], MLX4_DB_PER_PAGE >> o);
      |                                    ~~~~~~~~~~~^~~
  'mlx4_alloc_db_from_pgdir': events 1-2
  691 |                 i = find_first_bit(pgdir->bits[o], MLX4_DB_PER_PAGE >> o);                        |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      |                     |                         |                                                   |                     |                         (2) out of array bounds here
      |                     (1) when the condition is evaluated to true                             In file included from ../drivers/net/ethernet/mellanox/mlx4/mlx4.h:53,
                 from ../drivers/net/ethernet/mellanox/mlx4/alloc.c:42:
../include/linux/mlx4/device.h:664:33: note: while referencing 'bits'
  664 |         unsigned long          *bits[2];
      |                                 ^~~~

Switch the argument to unsigned int, which removes the compiler needing
to consider negative values.

Signed-off-by: Kees Cook <kees@kernel.org>
Link: https://patch.msgid.link/20250210174504.work.075-kees@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/alloc.c | 6 +++---
 include/linux/mlx4/device.h                | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/alloc.c b/drivers/net/ethernet/mellanox/mlx4/alloc.c
index b330020dc0d67..f2bded847e61d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/alloc.c
+++ b/drivers/net/ethernet/mellanox/mlx4/alloc.c
@@ -682,9 +682,9 @@ static struct mlx4_db_pgdir *mlx4_alloc_db_pgdir(struct device *dma_device)
 }
 
 static int mlx4_alloc_db_from_pgdir(struct mlx4_db_pgdir *pgdir,
-				    struct mlx4_db *db, int order)
+				    struct mlx4_db *db, unsigned int order)
 {
-	int o;
+	unsigned int o;
 	int i;
 
 	for (o = order; o <= 1; ++o) {
@@ -712,7 +712,7 @@ static int mlx4_alloc_db_from_pgdir(struct mlx4_db_pgdir *pgdir,
 	return 0;
 }
 
-int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order)
+int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, unsigned int order)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_db_pgdir *pgdir;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 27f42f713c891..86f0f2a25a3d6 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -1135,7 +1135,7 @@ int mlx4_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 int mlx4_buf_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 		       struct mlx4_buf *buf);
 
-int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, int order);
+int mlx4_db_alloc(struct mlx4_dev *dev, struct mlx4_db *db, unsigned int order);
 void mlx4_db_free(struct mlx4_dev *dev, struct mlx4_db *db);
 
 int mlx4_alloc_hwq_res(struct mlx4_dev *dev, struct mlx4_hwq_resources *wqres,
-- 
2.39.5

[PATCH AUTOSEL 6.12 379/486] net/mlx5: XDP, Enable TX side XDP multi-buffer support

[Sasha Levin]

From: Alexei Lazar <alazar@nvidia.com>

[ Upstream commit 1a9304859b3a4119579524c293b902a8927180f3 ]

In XDP scenarios, fragmented packets can occur if the MTU is larger
than the page size, even when the packet size fits within the linear
part.
If XDP multi-buffer support is disabled, the fragmented part won't be
handled in the TX flow, leading to packet drops.

Since XDP multi-buffer support is always available, this commit removes
the conditional check for enabling it.
This ensures that XDP multi-buffer support is always enabled,
regardless of the `is_xdp_mb` parameter, and guarantees the handling of
fragmented packets in such scenarios.

Signed-off-by: Alexei Lazar <alazar@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250209101716.112774-16-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  1 -
 .../ethernet/mellanox/mlx5/core/en/params.c   |  1 -
 .../ethernet/mellanox/mlx5/core/en/params.h   |  1 -
 .../mellanox/mlx5/core/en/reporter_tx.c       |  1 -
 .../net/ethernet/mellanox/mlx5/core/en/xdp.c  | 49 ++++++++-----------
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 29 -----------
 6 files changed, 21 insertions(+), 61 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 57b7298a0e793..d6266f6a96d6e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -385,7 +385,6 @@ enum {
 	MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE,
 	MLX5E_SQ_STATE_PENDING_XSK_TX,
 	MLX5E_SQ_STATE_PENDING_TLS_RX_RESYNC,
-	MLX5E_SQ_STATE_XDP_MULTIBUF,
 	MLX5E_NUM_SQ_STATES, /* Must be kept last */
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 31eb99f09c63c..8c4d710e85675 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -1242,7 +1242,6 @@ void mlx5e_build_xdpsq_param(struct mlx5_core_dev *mdev,
 	mlx5e_build_sq_param_common(mdev, param);
 	MLX5_SET(wq, wq, log_wq_sz, params->log_sq_size);
 	param->is_mpw = MLX5E_GET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE);
-	param->is_xdp_mb = !mlx5e_rx_is_linear_skb(mdev, params, xsk);
 	mlx5e_build_tx_cq_param(mdev, params, &param->cqp);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
index 3f8986f9d8629..bd5877acc5b1e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
@@ -33,7 +33,6 @@ struct mlx5e_sq_param {
 	struct mlx5_wq_param       wq;
 	bool                       is_mpw;
 	bool                       is_tls;
-	bool                       is_xdp_mb;
 	u16                        stop_room;
 };
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
index 09433b91be176..532c7fa94d172 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
@@ -16,7 +16,6 @@ static const char * const sq_sw_state_type_name[] = {
 	[MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE] = "vlan_need_l2_inline",
 	[MLX5E_SQ_STATE_PENDING_XSK_TX] = "pending_xsk_tx",
 	[MLX5E_SQ_STATE_PENDING_TLS_RX_RESYNC] = "pending_tls_rx_resync",
-	[MLX5E_SQ_STATE_XDP_MULTIBUF] = "xdp_multibuf",
 };
 
 static int mlx5e_wait_for_sq_flush(struct mlx5e_txqsq *sq)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 4610621a340e5..08ab0999f7b31 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -546,6 +546,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 	bool inline_ok;
 	bool linear;
 	u16 pi;
+	int i;
 
 	struct mlx5e_xdpsq_stats *stats = sq->stats;
 
@@ -612,41 +613,33 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
 
 	cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_SEND);
 
-	if (test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state)) {
-		int i;
-
-		memset(&cseg->trailer, 0, sizeof(cseg->trailer));
-		memset(eseg, 0, sizeof(*eseg) - sizeof(eseg->trailer));
-
-		eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz);
+	memset(&cseg->trailer, 0, sizeof(cseg->trailer));
+	memset(eseg, 0, sizeof(*eseg) - sizeof(eseg->trailer));
 
-		for (i = 0; i < num_frags; i++) {
-			skb_frag_t *frag = &xdptxdf->sinfo->frags[i];
-			dma_addr_t addr;
+	eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz);
 
-			addr = xdptxdf->dma_arr ? xdptxdf->dma_arr[i] :
-				page_pool_get_dma_addr(skb_frag_page(frag)) +
-				skb_frag_off(frag);
+	for (i = 0; i < num_frags; i++) {
+		skb_frag_t *frag = &xdptxdf->sinfo->frags[i];
+		dma_addr_t addr;
 
-			dseg->addr = cpu_to_be64(addr);
-			dseg->byte_count = cpu_to_be32(skb_frag_size(frag));
-			dseg->lkey = sq->mkey_be;
-			dseg++;
-		}
+		addr = xdptxdf->dma_arr ? xdptxdf->dma_arr[i] :
+			page_pool_get_dma_addr(skb_frag_page(frag)) +
+			skb_frag_off(frag);
 
-		cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
+		dseg->addr = cpu_to_be64(addr);
+		dseg->byte_count = cpu_to_be32(skb_frag_size(frag));
+		dseg->lkey = sq->mkey_be;
+		dseg++;
+	}
 
-		sq->db.wqe_info[pi] = (struct mlx5e_xdp_wqe_info) {
-			.num_wqebbs = num_wqebbs,
-			.num_pkts = 1,
-		};
+	cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
 
-		sq->pc += num_wqebbs;
-	} else {
-		cseg->fm_ce_se = 0;
+	sq->db.wqe_info[pi] = (struct mlx5e_xdp_wqe_info) {
+		.num_wqebbs = num_wqebbs,
+		.num_pkts = 1,
+	};
 
-		sq->pc++;
-	}
+	sq->pc += num_wqebbs;
 
 	xsk_tx_metadata_request(meta, &mlx5e_xsk_tx_metadata_ops, eseg);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 1c087fa1ca269..15ec9750d4be0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2030,41 +2030,12 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params,
 	csp.min_inline_mode = sq->min_inline_mode;
 	set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state);
 
-	if (param->is_xdp_mb)
-		set_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state);
-
 	err = mlx5e_create_sq_rdy(c->mdev, param, &csp, 0, &sq->sqn);
 	if (err)
 		goto err_free_xdpsq;
 
 	mlx5e_set_xmit_fp(sq, param->is_mpw);
 
-	if (!param->is_mpw && !test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state)) {
-		unsigned int ds_cnt = MLX5E_TX_WQE_EMPTY_DS_COUNT + 1;
-		unsigned int inline_hdr_sz = 0;
-		int i;
-
-		if (sq->min_inline_mode != MLX5_INLINE_MODE_NONE) {
-			inline_hdr_sz = MLX5E_XDP_MIN_INLINE;
-			ds_cnt++;
-		}
-
-		/* Pre initialize fixed WQE fields */
-		for (i = 0; i < mlx5_wq_cyc_get_size(&sq->wq); i++) {
-			struct mlx5e_tx_wqe      *wqe  = mlx5_wq_cyc_get_wqe(&sq->wq, i);
-			struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl;
-			struct mlx5_wqe_eth_seg  *eseg = &wqe->eth;
-
-			sq->db.wqe_info[i] = (struct mlx5e_xdp_wqe_info) {
-				.num_wqebbs = 1,
-				.num_pkts   = 1,
-			};
-
-			cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
-			eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz);
-		}
-	}
-
 	return 0;
 
 err_free_xdpsq:
-- 
2.39.5

[PATCH AUTOSEL 6.12 380/486] net/mlx5: Extend Ethtool loopback selftest to support non-linear SKB

[Sasha Levin]

From: Alexei Lazar <alazar@nvidia.com>

[ Upstream commit 95b9606b15bb3ce1198d28d2393dd0e1f0a5f3e9 ]

Current loopback test validation ignores non-linear SKB case in
the SKB access, which can lead to failures in scenarios such as
when HW GRO is enabled.
Linearize the SKB so both cases will be handled.

Signed-off-by: Alexei Lazar <alazar@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250209101716.112774-15-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c
index 1d60465cc2ca4..2f7a543feca62 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c
@@ -166,6 +166,9 @@ mlx5e_test_loopback_validate(struct sk_buff *skb,
 	struct udphdr *udph;
 	struct iphdr *iph;
 
+	if (skb_linearize(skb))
+		goto out;
+
 	/* We are only going to peek, no need to clone the SKB */
 	if (MLX5E_TEST_PKT_SIZE - ETH_HLEN > skb_headlen(skb))
 		goto out;
-- 
2.39.5

[PATCH AUTOSEL 6.12 381/486] net/mlx5e: set the tx_queue_len for pfifo_fast

[Sasha Levin]

From: William Tu <witu@nvidia.com>

[ Upstream commit a38cc5706fb9f7dc4ee3a443f61de13ce1e410ed ]

By default, the mq netdev creates a pfifo_fast qdisc. On a
system with 16 core, the pfifo_fast with 3 bands consumes
16 * 3 * 8 (size of pointer) * 1024 (default tx queue len)
= 393KB. The patch sets the tx qlen to representor default
value, 128 (1<<MLX5E_REP_PARAMS_DEF_LOG_SQ_SIZE), which
consumes 16 * 3 * 8 * 128 = 49KB, saving 344KB for each
representor at ECPF.

Signed-off-by: William Tu <witu@nvidia.com>
Reviewed-by: Daniel Jurgens <danielj@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Link: https://patch.msgid.link/20250209101716.112774-9-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 0657d10765357..fd1f460b7be65 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -885,6 +885,8 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev,
 	netdev->ethtool_ops = &mlx5e_rep_ethtool_ops;
 
 	netdev->watchdog_timeo    = 15 * HZ;
+	if (mlx5_core_is_ecpf(mdev))
+		netdev->tx_queue_len = 1 << MLX5E_REP_PARAMS_DEF_LOG_SQ_SIZE;
 
 #if IS_ENABLED(CONFIG_MLX5_CLS_ACT)
 	netdev->hw_features    |= NETIF_F_HW_TC;
-- 
2.39.5

[PATCH AUTOSEL 6.12 382/486] net/mlx5e: reduce rep rxq depth to 256 for ECPF

[Sasha Levin]

From: William Tu <witu@nvidia.com>

[ Upstream commit b9cc8f9d700867aaa77aedddfea85e53d5e5d584 ]

By experiments, a single queue representor netdev consumes kernel
memory around 2.8MB, and 1.8MB out of the 2.8MB is due to page
pool for the RXQ. Scaling to a thousand representors consumes 2.8GB,
which becomes a memory pressure issue for embedded devices such as
BlueField-2 16GB / BlueField-3 32GB memory.

Since representor netdevs mostly handles miss traffic, and ideally,
most of the traffic will be offloaded, reduce the default non-uplink
rep netdev's RXQ default depth from 1024 to 256 if mdev is ecpf eswitch
manager. This saves around 1MB of memory per regular RQ,
(1024 - 256) * 2KB, allocated from page pool.

With rxq depth of 256, the netlink page pool tool reports
$./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
	 --dump page-pool-get
 {'id': 277,
  'ifindex': 9,
  'inflight': 128,
  'inflight-mem': 786432,
  'napi-id': 775}]

This is due to mtu 1500 + headroom consumes half pages, so 256 rxq
entries consumes around 128 pages (thus create a page pool with
size 128), shown above at inflight.

Note that each netdev has multiple types of RQs, including
Regular RQ, XSK, PTP, Drop, Trap RQ. Since non-uplink representor
only supports regular rq, this patch only changes the regular RQ's
default depth.

Signed-off-by: William Tu <witu@nvidia.com>
Reviewed-by: Bodong Wang <bodong@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Link: https://patch.msgid.link/20250209101716.112774-8-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index fd1f460b7be65..18ec392d17404 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -65,6 +65,7 @@
 #define MLX5E_REP_PARAMS_DEF_LOG_SQ_SIZE \
 	max(0x7, MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE)
 #define MLX5E_REP_PARAMS_DEF_NUM_CHANNELS 1
+#define MLX5E_REP_PARAMS_DEF_LOG_RQ_SIZE 0x8
 
 static const char mlx5e_rep_driver_name[] = "mlx5e_rep";
 
@@ -854,6 +855,8 @@ static void mlx5e_build_rep_params(struct net_device *netdev)
 
 	/* RQ */
 	mlx5e_build_rq_params(mdev, params);
+	if (!mlx5e_is_uplink_rep(priv) && mlx5_core_is_ecpf(mdev))
+		params->log_rq_mtu_frames = MLX5E_REP_PARAMS_DEF_LOG_RQ_SIZE;
 
 	/* If netdev is already registered (e.g. move from nic profile to uplink,
 	 * RTNL lock must be held before triggering netdev notifiers.
-- 
2.39.5

[PATCH AUTOSEL 6.12 383/486] net/mlx5e: reduce the max log mpwrq sz for ECPF and reps

[Sasha Levin]

From: William Tu <witu@nvidia.com>

[ Upstream commit e1d68ea58c7e9ebacd9ad7a99b25a3578fa62182 ]

For the ECPF and representors, reduce the max MPWRQ size from 256KB (18)
to 128KB (17). This prepares the later patch for saving representor
memory.

With Striding RQ, there is a minimum of 4 MPWQEs. So with 128KB of max
MPWRQ size, the minimal memory is 4 * 128KB = 512KB. When creating page
pool, consider 1500 mtu, the minimal page pool size will be 512KB/4KB =
128 pages = 256 rx ring entries (2 entries per page).

Before this patch, setting RX ringsize (ethtool -G rx) to 256 causes
driver to allocate page pool size more than it needs due to max MPWRQ
is 256KB (18). Ex: 4 * 256KB = 1MB, 1MB/4KB = 256 pages, but actually
128 pages is good enough. Reducing the max MPWRQ to 128KB fixes the
limitation.

Signed-off-by: William Tu <witu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Link: https://patch.msgid.link/20250209101716.112774-7-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  2 --
 .../net/ethernet/mellanox/mlx5/core/en/params.c   | 15 +++++++++++----
 2 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index d6266f6a96d6e..e048a667e0758 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -94,8 +94,6 @@ struct page_pool;
 #define MLX5_MPWRQ_DEF_LOG_STRIDE_SZ(mdev) \
 	MLX5_MPWRQ_LOG_STRIDE_SZ(mdev, order_base_2(MLX5E_RX_MAX_HEAD))
 
-#define MLX5_MPWRQ_MAX_LOG_WQE_SZ 18
-
 /* Keep in sync with mlx5e_mpwrq_log_wqe_sz.
  * These are theoretical maximums, which can be further restricted by
  * capabilities. These values are used for static resource allocations and
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 8c4d710e85675..58ec5e44aa7ad 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -10,6 +10,9 @@
 #include <net/page_pool/types.h>
 #include <net/xdp_sock_drv.h>
 
+#define MLX5_MPWRQ_MAX_LOG_WQE_SZ 18
+#define MLX5_REP_MPWRQ_MAX_LOG_WQE_SZ 17
+
 static u8 mlx5e_mpwrq_min_page_shift(struct mlx5_core_dev *mdev)
 {
 	u8 min_page_shift = MLX5_CAP_GEN_2(mdev, log_min_mkey_entity_size);
@@ -103,18 +106,22 @@ u8 mlx5e_mpwrq_log_wqe_sz(struct mlx5_core_dev *mdev, u8 page_shift,
 			  enum mlx5e_mpwrq_umr_mode umr_mode)
 {
 	u8 umr_entry_size = mlx5e_mpwrq_umr_entry_size(umr_mode);
-	u8 max_pages_per_wqe, max_log_mpwqe_size;
+	u8 max_pages_per_wqe, max_log_wqe_size_calc;
+	u8 max_log_wqe_size_cap;
 	u16 max_wqe_size;
 
 	/* Keep in sync with MLX5_MPWRQ_MAX_PAGES_PER_WQE. */
 	max_wqe_size = mlx5e_get_max_sq_aligned_wqebbs(mdev) * MLX5_SEND_WQE_BB;
 	max_pages_per_wqe = ALIGN_DOWN(max_wqe_size - sizeof(struct mlx5e_umr_wqe),
 				       MLX5_UMR_FLEX_ALIGNMENT) / umr_entry_size;
-	max_log_mpwqe_size = ilog2(max_pages_per_wqe) + page_shift;
+	max_log_wqe_size_calc = ilog2(max_pages_per_wqe) + page_shift;
+
+	WARN_ON_ONCE(max_log_wqe_size_calc < MLX5E_ORDER2_MAX_PACKET_MTU);
 
-	WARN_ON_ONCE(max_log_mpwqe_size < MLX5E_ORDER2_MAX_PACKET_MTU);
+	max_log_wqe_size_cap = mlx5_core_is_ecpf(mdev) ?
+			   MLX5_REP_MPWRQ_MAX_LOG_WQE_SZ : MLX5_MPWRQ_MAX_LOG_WQE_SZ;
 
-	return min_t(u8, max_log_mpwqe_size, MLX5_MPWRQ_MAX_LOG_WQE_SZ);
+	return min_t(u8, max_log_wqe_size_calc, max_log_wqe_size_cap);
 }
 
 u8 mlx5e_mpwrq_pages_per_wqe(struct mlx5_core_dev *mdev, u8 page_shift,
-- 
2.39.5

[PATCH AUTOSEL 6.12 422/486] net/mlx5e: Avoid WARN_ON when configuring MQPRIO with HTB offload enabled

[Sasha Levin]

From: Carolina Jubran <cjubran@nvidia.com>

[ Upstream commit 689805dcc474c2accb5cffbbcea1c06ee4a54570 ]

When attempting to enable MQPRIO while HTB offload is already
configured, the driver currently returns `-EINVAL` and triggers a
`WARN_ON`, leading to an unnecessary call trace.

Update the code to handle this case more gracefully by returning
`-EOPNOTSUPP` instead, while also providing a helpful user message.

Signed-off-by: Carolina Jubran <cjubran@nvidia.com>
Reviewed-by: Yael Chemla <ychemla@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 15ec9750d4be0..36a2c935267b0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3759,8 +3759,11 @@ static int mlx5e_setup_tc_mqprio(struct mlx5e_priv *priv,
 	/* MQPRIO is another toplevel qdisc that can't be attached
 	 * simultaneously with the offloaded HTB.
 	 */
-	if (WARN_ON(mlx5e_selq_is_htb_enabled(&priv->selq)))
-		return -EINVAL;
+	if (mlx5e_selq_is_htb_enabled(&priv->selq)) {
+		NL_SET_ERR_MSG_MOD(mqprio->extack,
+				   "MQPRIO cannot be configured when HTB offload is enabled.");
+		return -EOPNOTSUPP;
+	}
 
 	switch (mqprio->mode) {
 	case TC_MQPRIO_MODE_DCB:
-- 
2.39.5