* [PATCH] rhashtable: Remove obsolete rhashtable_walk_init function
From: Herbert Xu @ 2019-02-13 5:34 UTC (permalink / raw)
To: David Miller, johannes, linux-wireless, netdev, j, tgraf,
johannes.berg
In-Reply-To: <20190213050551.x3jffq3ipghw6g2m@gondor.apana.org.au>
Here is another patch on top the fixes to mac80211 to finally
remove the obsolete rhashtable_walk_init API.
---8<---
The rhashtable_walk_init function has been obsolete for more than
two years. This patch finally converts its last users over to
rhashtable_walk_enter and removes it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h
index 20f9c6af7473..ae9c0f71f311 100644
--- a/include/linux/rhashtable.h
+++ b/include/linux/rhashtable.h
@@ -1113,14 +1113,6 @@ static inline int rhashtable_replace_fast(
return err;
}
-/* Obsolete function, do not use in new code. */
-static inline int rhashtable_walk_init(struct rhashtable *ht,
- struct rhashtable_iter *iter, gfp_t gfp)
-{
- rhashtable_walk_enter(ht, iter);
- return 0;
-}
-
/**
* rhltable_walk_enter - Initialise an iterator
* @hlt: Table to walk over
diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index 852ffa5160f1..0a105d4af166 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -682,7 +682,7 @@ EXPORT_SYMBOL_GPL(rhashtable_walk_enter);
* rhashtable_walk_exit - Free an iterator
* @iter: Hash table Iterator
*
- * This function frees resources allocated by rhashtable_walk_init.
+ * This function frees resources allocated by rhashtable_walk_enter.
*/
void rhashtable_walk_exit(struct rhashtable_iter *iter)
{
diff --git a/lib/test_rhashtable.c b/lib/test_rhashtable.c
index 6a8ac7626797..d96962eea942 100644
--- a/lib/test_rhashtable.c
+++ b/lib/test_rhashtable.c
@@ -177,16 +177,11 @@ static int __init test_rht_lookup(struct rhashtable *ht, struct test_obj *array,
static void test_bucket_stats(struct rhashtable *ht, unsigned int entries)
{
- unsigned int err, total = 0, chain_len = 0;
+ unsigned int total = 0, chain_len = 0;
struct rhashtable_iter hti;
struct rhash_head *pos;
- err = rhashtable_walk_init(ht, &hti, GFP_KERNEL);
- if (err) {
- pr_warn("Test failed: allocation error");
- return;
- }
-
+ rhashtable_walk_enter(ht, &hti);
rhashtable_walk_start(&hti);
while ((pos = rhashtable_walk_next(&hti))) {
diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c
index 17c455ff69ff..ae6cd4cef8db 100644
--- a/net/ipv6/ila/ila_xlat.c
+++ b/net/ipv6/ila/ila_xlat.c
@@ -385,10 +385,7 @@ int ila_xlat_nl_cmd_flush(struct sk_buff *skb, struct genl_info *info)
spinlock_t *lock;
int ret;
- ret = rhashtable_walk_init(&ilan->xlat.rhash_table, &iter, GFP_KERNEL);
- if (ret)
- goto done;
-
+ rhashtable_walk_enter(&ilan->xlat.rhash_table, &iter);
rhashtable_walk_start(&iter);
for (;;) {
@@ -509,23 +506,17 @@ int ila_xlat_nl_dump_start(struct netlink_callback *cb)
struct net *net = sock_net(cb->skb->sk);
struct ila_net *ilan = net_generic(net, ila_net_id);
struct ila_dump_iter *iter;
- int ret;
iter = kmalloc(sizeof(*iter), GFP_KERNEL);
if (!iter)
return -ENOMEM;
- ret = rhashtable_walk_init(&ilan->xlat.rhash_table, &iter->rhiter,
- GFP_KERNEL);
- if (ret) {
- kfree(iter);
- return ret;
- }
+ rhashtable_walk_enter(&ilan->xlat.rhash_table, &iter->rhiter);
iter->skip = 0;
cb->args[0] = (long)iter;
- return ret;
+ return 0;
}
int ila_xlat_nl_dump_done(struct netlink_callback *cb)
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 3c023d6120f6..f369cc751cea 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2541,15 +2541,7 @@ struct nl_seq_iter {
static int netlink_walk_start(struct nl_seq_iter *iter)
{
- int err;
-
- err = rhashtable_walk_init(&nl_table[iter->link].hash, &iter->hti,
- GFP_KERNEL);
- if (err) {
- iter->link = MAX_LINKS;
- return err;
- }
-
+ rhashtable_walk_enter(&nl_table[iter->link].hash, &iter->hti);
rhashtable_walk_start(&iter->hti);
return 0;
--
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply related
* Re: [PATCH iproute2] lib/libnetlink: ensure a minimum of 32KB for the buffer used in rtnl_recvmsg()
From: Hangbin Liu @ 2019-02-13 6:20 UTC (permalink / raw)
To: David Ahern; +Cc: Eric Dumazet, Stephen Hemminger, netdev, Phil Sutter
In-Reply-To: <5f91bd70-98af-8108-859b-d0482f80674e@gmail.com>
Hi David,
On Tue, Feb 12, 2019 at 06:08:13PM -0800, Eric Dumazet wrote:
> >> diff --git a/lib/libnetlink.c b/lib/libnetlink.c
> >> index 1892a02ab5d0d73776c9882ffc77edcd2c663d01..0d48a3d43cf03065dacbd419578ab10af56431a4 100644
> >> --- a/lib/libnetlink.c
> >> +++ b/lib/libnetlink.c
> >> @@ -718,6 +718,8 @@ static int rtnl_recvmsg(int fd, struct msghdr *msg, char **answer)
> >> if (len < 0)
> >> return len;
> >>
> >> + if (len < 32768)
> >> + len = 32768;
> >> buf = malloc(len);
> >> if (!buf) {
> >> fprintf(stderr, "malloc error: not enough buffer\n");
> >>
> >
> > I believe that negates the whole point of 2d34851cd341 - which I have no
> > problem with. 2 recvmsg calls per message is overkill.
It should not affects ip cmd too much. But for ss, as Eric pointed, it
cause performance issue.
> >
>
> It does not negates the point at all.
>
> The main point was to eventually be able to allocate more than 32KB.
>
> We need to have a minimum size of 32KB so that the kernel can cook reasonably sized skbs
>
> Because trying to allocate 4KB only in 2019 is kind of stupid...
>
> ( Especially considering ss currently buffers the whole thing before calling render() !!! )
This makes sense to me.
> > Do we know of any single message sizes > 32k? 2d34851cd341 cites
> > increasing VF's but at some point there is a limit. If not, the whole
> > PEEK thing should go away and we just malloc 32k (or 64k) buffers for
> > each recvmsg.
Apart from the 200 VFs example, I think there will be more and more virtual
interfaces be used in cloud environment, like openstack/OVS, so useing 32K
or 64K is still not safe.
What do you think.
Thanks
Hangbin
^ permalink raw reply
* [PATCH mlx5-next 00/15] mlx5-next updates (ECPF & 50Gbps per lane)
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky; +Cc: netdev, linux-rdma
Hi All,
This series from Bodong and Aya provides two low level updates to mlx5
core driver that are required for both mlx5e netdev and mlx5_ib drivers.
From Bodong Wang:
1) Introduction of ECPF (Embedded CPU Physical Function), and low level
bits for mlx5 SmartNic capabilities support.
2) Vport enumeration refactoring that affect mlx5_ib and mlx5_core
From Aya Levin,
3) Add support for 50Gbps per lane link modes in the Port Type and Speed
register (PTYS)
4) Refactor low level query functions for PTYS register
5) Add support for 50Gbps per lane link modes to mlx5_ib
Thanks,
Saeed.
---
Aya Levin (4):
net/mlx5: Refactor queries to speed fields in Port Type and Speed
register
net/mlx5: Add new fields to Port Type and Speed register
net/mlx5: Add support to ext_* fields introduced in Port Type and
Speed register
IB/mlx5: Add support for 50Gbps per lane link modes
Bodong Wang (11):
net/mlx5: Use void pointer as the type in address_of macro
net/mlx5: Use consistent vport num argument type
IB/mlx5: Use unified register/load function for uplink and VF vports
net/mlx5: Introduce Mellanox SmartNIC and modify page management logic
net/mlx5: Update enable HCA dependency
net/mlx5: Add query host params command
net/mlx5: Add host params change event
net/mlx5: Provide an alternative VF upper bound for ECPF
net/mlx5: E-Switch, Normalize the name of uplink vport number
net/mlx5: Relocate vport macros to the vport header file
net/mlx5: E-Switch, Avoid magic numbers when initializing offloads
mode
drivers/infiniband/hw/mlx5/ib_rep.c | 73 +++------
drivers/infiniband/hw/mlx5/ib_rep.h | 10 +-
drivers/infiniband/hw/mlx5/main.c | 95 +++++++++---
.../net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 +
.../net/ethernet/mellanox/mlx5/core/ecpf.c | 112 ++++++++++++++
.../net/ethernet/mellanox/mlx5/core/ecpf.h | 33 ++++
.../net/ethernet/mellanox/mlx5/core/en/port.c | 142 +++++++++++++++---
.../net/ethernet/mellanox/mlx5/core/en/port.h | 16 +-
.../ethernet/mellanox/mlx5/core/en_ethtool.c | 28 ++--
.../net/ethernet/mellanox/mlx5/core/en_rep.c | 22 +--
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 6 +-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 4 +
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 42 +++---
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 2 -
.../mellanox/mlx5/core/eswitch_offloads.c | 39 ++---
.../net/ethernet/mellanox/mlx5/core/events.c | 2 +
.../net/ethernet/mellanox/mlx5/core/fs_core.c | 1 +
.../net/ethernet/mellanox/mlx5/core/main.c | 16 ++
.../ethernet/mellanox/mlx5/core/mlx5_core.h | 2 +-
.../ethernet/mellanox/mlx5/core/pagealloc.c | 54 +++++--
.../net/ethernet/mellanox/mlx5/core/port.c | 106 -------------
.../net/ethernet/mellanox/mlx5/core/sriov.c | 2 +-
.../net/ethernet/mellanox/mlx5/core/vport.c | 8 +-
include/linux/mlx5/device.h | 6 +-
include/linux/mlx5/driver.h | 27 +++-
include/linux/mlx5/mlx5_ifc.h | 71 +++++++--
include/linux/mlx5/port.h | 30 ++--
include/linux/mlx5/vport.h | 21 ++-
29 files changed, 641 insertions(+), 333 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
--
2.20.1
^ permalink raw reply
* [PATCH mlx5-next 01/15] net/mlx5: Use void pointer as the type in address_of macro
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen, Or Gerlitz,
Daniel Jurgens
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Better to use void * and avoid unnecessary casts.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/linux/mlx5/device.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 0845a227a7b2..46223efa1877 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -67,7 +67,7 @@
#define MLX5_UN_SZ_BYTES(typ) (sizeof(union mlx5_ifc_##typ##_bits) / 8)
#define MLX5_UN_SZ_DW(typ) (sizeof(union mlx5_ifc_##typ##_bits) / 32)
#define MLX5_BYTE_OFF(typ, fld) (__mlx5_bit_off(typ, fld) / 8)
-#define MLX5_ADDR_OF(typ, p, fld) ((char *)(p) + MLX5_BYTE_OFF(typ, fld))
+#define MLX5_ADDR_OF(typ, p, fld) ((void *)((uint8_t *)(p) + MLX5_BYTE_OFF(typ, fld)))
/* insert a value to a struct */
#define MLX5_SET(typ, p, fld, v) do { \
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 02/15] net/mlx5: Use consistent vport num argument type
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Use u16 for vport number, which matches how hardware refers to this
argument throughout commands.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 32 +++++++++----------
.../net/ethernet/mellanox/mlx5/core/vport.c | 8 ++---
include/linux/mlx5/vport.h | 8 ++---
3 files changed, 24 insertions(+), 24 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index a44ea7b85614..d7382892e81c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -52,7 +52,7 @@ enum {
struct vport_addr {
struct l2addr_node node;
u8 action;
- u32 vport;
+ u16 vport;
struct mlx5_flow_handle *flow_rule;
bool mpfs; /* UC MAC was added to MPFs */
/* A flag indicating that mac was added due to mc promiscuous vport */
@@ -115,7 +115,7 @@ static int modify_esw_vport_context_cmd(struct mlx5_core_dev *dev, u16 vport,
return mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
}
-static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport,
+static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u16 vport,
u16 vlan, u8 qos, u8 set_flags)
{
u32 in[MLX5_ST_SZ_DW(modify_esw_vport_context_in)] = {0};
@@ -152,7 +152,7 @@ static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport,
/* E-Switch FDB */
static struct mlx5_flow_handle *
-__esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule,
+__esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u16 vport, bool rx_rule,
u8 mac_c[ETH_ALEN], u8 mac_v[ETH_ALEN])
{
int match_header = (is_zero_ether_addr(mac_c) ? 0 :
@@ -215,7 +215,7 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule,
}
static struct mlx5_flow_handle *
-esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
+esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u16 vport)
{
u8 mac_c[ETH_ALEN];
@@ -224,7 +224,7 @@ esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
}
static struct mlx5_flow_handle *
-esw_fdb_set_vport_allmulti_rule(struct mlx5_eswitch *esw, u32 vport)
+esw_fdb_set_vport_allmulti_rule(struct mlx5_eswitch *esw, u16 vport)
{
u8 mac_c[ETH_ALEN];
u8 mac_v[ETH_ALEN];
@@ -237,7 +237,7 @@ esw_fdb_set_vport_allmulti_rule(struct mlx5_eswitch *esw, u32 vport)
}
static struct mlx5_flow_handle *
-esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u32 vport)
+esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u16 vport)
{
u8 mac_c[ETH_ALEN];
u8 mac_v[ETH_ALEN];
@@ -377,7 +377,7 @@ typedef int (*vport_addr_action)(struct mlx5_eswitch *esw,
static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
{
u8 *mac = vaddr->node.addr;
- u32 vport = vaddr->vport;
+ u16 vport = vaddr->vport;
int err;
/* Skip mlx5_mpfs_add_mac for PFs,
@@ -409,7 +409,7 @@ static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
{
u8 *mac = vaddr->node.addr;
- u32 vport = vaddr->vport;
+ u16 vport = vaddr->vport;
int err = 0;
/* Skip mlx5_mpfs_del_mac for PFs,
@@ -438,7 +438,7 @@ static void update_allmulti_vports(struct mlx5_eswitch *esw,
struct esw_mc_addr *esw_mc)
{
u8 *mac = vaddr->node.addr;
- u32 vport_idx = 0;
+ u16 vport_idx = 0;
for (vport_idx = 0; vport_idx < esw->total_vports; vport_idx++) {
struct mlx5_vport *vport = &esw->vports[vport_idx];
@@ -485,7 +485,7 @@ static int esw_add_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
struct hlist_head *hash = esw->mc_table;
struct esw_mc_addr *esw_mc;
u8 *mac = vaddr->node.addr;
- u32 vport = vaddr->vport;
+ u16 vport = vaddr->vport;
if (!esw->fdb_table.legacy.fdb)
return 0;
@@ -525,7 +525,7 @@ static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
struct hlist_head *hash = esw->mc_table;
struct esw_mc_addr *esw_mc;
u8 *mac = vaddr->node.addr;
- u32 vport = vaddr->vport;
+ u16 vport = vaddr->vport;
if (!esw->fdb_table.legacy.fdb)
return 0;
@@ -564,7 +564,7 @@ static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
/* Apply vport UC/MC list to HW l2 table and FDB table */
static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
- u32 vport_num, int list_type)
+ u16 vport_num, int list_type)
{
struct mlx5_vport *vport = &esw->vports[vport_num];
bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
@@ -599,7 +599,7 @@ static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
/* Sync vport UC/MC list from vport context */
static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
- u32 vport_num, int list_type)
+ u16 vport_num, int list_type)
{
struct mlx5_vport *vport = &esw->vports[vport_num];
bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
@@ -686,7 +686,7 @@ static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
/* Sync vport UC/MC list from vport context
* Must be called after esw_update_vport_addr_list
*/
-static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u32 vport_num)
+static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u16 vport_num)
{
struct mlx5_vport *vport = &esw->vports[vport_num];
struct l2addr_node *node;
@@ -721,7 +721,7 @@ static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u32 vport_num)
}
/* Apply vport rx mode to HW FDB table */
-static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u32 vport_num,
+static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
bool promisc, bool mc_promisc)
{
struct esw_mc_addr *allmulti_addr = &esw->mc_promisc;
@@ -764,7 +764,7 @@ static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u32 vport_num,
}
/* Sync vport rx mode from vport context */
-static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw, u32 vport_num)
+static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num)
{
struct mlx5_vport *vport = &esw->vports[vport_num];
int promisc_all = 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
index 9b150ce9d315..9a928eb48522 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
@@ -255,7 +255,7 @@ int mlx5_modify_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 mtu)
EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mtu);
int mlx5_query_nic_vport_mac_list(struct mlx5_core_dev *dev,
- u32 vport,
+ u16 vport,
enum mlx5_list_type list_type,
u8 addr_list[][ETH_ALEN],
int *list_size)
@@ -373,7 +373,7 @@ int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev,
EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_list);
int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev,
- u32 vport,
+ u16 vport,
u16 vlans[],
int *size)
{
@@ -526,7 +526,7 @@ int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid)
EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_node_guid);
int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev,
- u32 vport, u64 node_guid)
+ u16 vport, u64 node_guid)
{
int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
void *nic_vport_context;
@@ -827,7 +827,7 @@ int mlx5_query_hca_vport_node_guid(struct mlx5_core_dev *dev,
EXPORT_SYMBOL_GPL(mlx5_query_hca_vport_node_guid);
int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev,
- u32 vport,
+ u16 vport,
int *promisc_uc,
int *promisc_mc,
int *promisc_all)
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 9c694808c212..1654b911cdb2 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -60,7 +60,7 @@ int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev,
u64 *system_image_guid);
int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid);
int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev,
- u32 vport, u64 node_guid);
+ u16 vport, u64 node_guid);
int mlx5_query_nic_vport_qkey_viol_cntr(struct mlx5_core_dev *mdev,
u16 *qkey_viol_cntr);
int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport,
@@ -78,7 +78,7 @@ int mlx5_query_hca_vport_system_image_guid(struct mlx5_core_dev *dev,
int mlx5_query_hca_vport_node_guid(struct mlx5_core_dev *dev,
u64 *node_guid);
int mlx5_query_nic_vport_mac_list(struct mlx5_core_dev *dev,
- u32 vport,
+ u16 vport,
enum mlx5_list_type list_type,
u8 addr_list[][ETH_ALEN],
int *list_size);
@@ -87,7 +87,7 @@ int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev,
u8 addr_list[][ETH_ALEN],
int list_size);
int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev,
- u32 vport,
+ u16 vport,
int *promisc_uc,
int *promisc_mc,
int *promisc_all);
@@ -96,7 +96,7 @@ int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev,
int promisc_mc,
int promisc_all);
int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev,
- u32 vport,
+ u16 vport,
u16 vlans[],
int *size);
int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev,
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 04/15] net/mlx5: Introduce Mellanox SmartNIC and modify page management logic
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Mellanox's SmartNIC combines embedded CPU(e.g, ARM) processing power
with advanced network offloads to accelerate a multitude of security,
networking and storage applications.
With the introduction of the SmartNIC, there is a new PCI function
called Embedded CPU Physical Function(ECPF). And it's possible for a
PF to get its ICM pages from the ECPF PCI function. Driver shall
identify if it is running on such a function by reading a bit in
the initialization segment.
When firmware asks for pages, it would issue a page request event
specifying how many pages it requests and for which function. That
driver responds with a manage_pages command providing the requested
pages along with an indication for which function it is providing these
pages.
The encoding before this patch was as follows:
function_id == 0: pages are requested for the function receiving
the EQE.
function_id != 0: pages are requested for VF identified by the
function_id value
A new one bit field in the EQE identifies that pages are requested for
the ECPF.
The notion of page_supplier can be introduced here and to support that,
manage pages and query pages were modified so firmware can distinguish
the following cases:
1. Function provides pages for itself
2. PF provides pages for its VF
3. ECPF provides pages to itself
4. ECPF provides pages for another function
This distinction is possible through the introduction of the bit
"embedded_cpu_function" in query_pages, manage_pages and page request
EQE.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
.../net/ethernet/mellanox/mlx5/core/ecpf.c | 9 ++++
.../net/ethernet/mellanox/mlx5/core/ecpf.h | 25 +++++++++
.../net/ethernet/mellanox/mlx5/core/main.c | 2 +
.../ethernet/mellanox/mlx5/core/mlx5_core.h | 2 +-
.../ethernet/mellanox/mlx5/core/pagealloc.c | 54 +++++++++++++------
.../net/ethernet/mellanox/mlx5/core/sriov.c | 2 +-
include/linux/mlx5/device.h | 2 +-
include/linux/mlx5/driver.h | 9 +++-
include/linux/mlx5/mlx5_ifc.h | 12 +++--
10 files changed, 94 insertions(+), 25 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 0257731e6d42..07965350b903 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -35,7 +35,7 @@ mlx5_core-$(CONFIG_MLX5_ESWITCH) += en_rep.o en_tc.o en/tc_tun.o
#
# Core extra
#
-mlx5_core-$(CONFIG_MLX5_ESWITCH) += eswitch.o eswitch_offloads.o
+mlx5_core-$(CONFIG_MLX5_ESWITCH) += eswitch.o eswitch_offloads.o ecpf.o
mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o
mlx5_core-$(CONFIG_VXLAN) += lib/vxlan.o
mlx5_core-$(CONFIG_PTP_1588_CLOCK) += lib/clock.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
new file mode 100644
index 000000000000..28b8c5c5c8c7
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#include "ecpf.h"
+
+bool mlx5_read_embedded_cpu(struct mlx5_core_dev *dev)
+{
+ return (ioread32be(&dev->iseg->initializing) >> MLX5_ECPU_BIT_NUM) & 1;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
new file mode 100644
index 000000000000..8b684f0ab48f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#ifndef __MLX5_ECPF_H__
+#define __MLX5_ECPF_H__
+
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+
+#ifdef CONFIG_MLX5_ESWITCH
+
+enum {
+ MLX5_ECPU_BIT_NUM = 23,
+};
+
+bool mlx5_read_embedded_cpu(struct mlx5_core_dev *dev);
+
+#else /* CONFIG_MLX5_ESWITCH */
+
+static inline bool
+mlx5_read_embedded_cpu(struct mlx5_core_dev *dev) { return false; }
+
+#endif /* CONFIG_MLX5_ESWITCH */
+
+#endif /* __MLX5_ECPF_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index e38aa206ab6d..bf810fd91454 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -65,6 +65,7 @@
#include "lib/vxlan.h"
#include "lib/devcom.h"
#include "diag/fw_tracer.h"
+#include "ecpf.h"
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) core driver");
@@ -901,6 +902,7 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
struct pci_dev *pdev = dev->pdev;
int err;
+ dev->caps.embedded_cpu = mlx5_read_embedded_cpu(dev);
mutex_lock(&dev->intf_state_mutex);
if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) {
dev_warn(&dev->pdev->dev, "%s: interface is up, NOP\n",
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index c68dcea5985b..b127044293b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -121,7 +121,7 @@ int mlx5_modify_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy,
u32 modify_bitmask);
int mlx5_destroy_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy,
u32 element_id);
-int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
+int mlx5_wait_for_pages(struct mlx5_core_dev *dev, int *pages);
u64 mlx5_read_internal_timer(struct mlx5_core_dev *dev,
struct ptp_system_timestamp *sts);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
index a83b517b0714..41025387ff2c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
@@ -48,6 +48,7 @@ enum {
struct mlx5_pages_req {
struct mlx5_core_dev *dev;
u16 func_id;
+ u8 ec_function;
s32 npages;
struct work_struct work;
};
@@ -143,6 +144,7 @@ static int mlx5_cmd_query_pages(struct mlx5_core_dev *dev, u16 *func_id,
MLX5_SET(query_pages_in, in, op_mod, boot ?
MLX5_QUERY_PAGES_IN_OP_MOD_BOOT_PAGES :
MLX5_QUERY_PAGES_IN_OP_MOD_INIT_PAGES);
+ MLX5_SET(query_pages_in, in, embedded_cpu_function, mlx5_core_is_ecpf(dev));
err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
if (err)
@@ -253,7 +255,8 @@ static int alloc_system_page(struct mlx5_core_dev *dev, u16 func_id)
return err;
}
-static void page_notify_fail(struct mlx5_core_dev *dev, u16 func_id)
+static void page_notify_fail(struct mlx5_core_dev *dev, u16 func_id,
+ bool ec_function)
{
u32 out[MLX5_ST_SZ_DW(manage_pages_out)] = {0};
u32 in[MLX5_ST_SZ_DW(manage_pages_in)] = {0};
@@ -262,6 +265,7 @@ static void page_notify_fail(struct mlx5_core_dev *dev, u16 func_id)
MLX5_SET(manage_pages_in, in, opcode, MLX5_CMD_OP_MANAGE_PAGES);
MLX5_SET(manage_pages_in, in, op_mod, MLX5_PAGES_CANT_GIVE);
MLX5_SET(manage_pages_in, in, function_id, func_id);
+ MLX5_SET(manage_pages_in, in, embedded_cpu_function, ec_function);
err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
if (err)
@@ -270,7 +274,7 @@ static void page_notify_fail(struct mlx5_core_dev *dev, u16 func_id)
}
static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
- int notify_fail)
+ int notify_fail, bool ec_function)
{
u32 out[MLX5_ST_SZ_DW(manage_pages_out)] = {0};
int inlen = MLX5_ST_SZ_BYTES(manage_pages_in);
@@ -305,6 +309,7 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
MLX5_SET(manage_pages_in, in, op_mod, MLX5_PAGES_GIVE);
MLX5_SET(manage_pages_in, in, function_id, func_id);
MLX5_SET(manage_pages_in, in, input_num_entries, npages);
+ MLX5_SET(manage_pages_in, in, embedded_cpu_function, ec_function);
err = mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
if (err) {
@@ -316,8 +321,11 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
dev->priv.fw_pages += npages;
if (func_id)
dev->priv.vfs_pages += npages;
+ else if (mlx5_core_is_ecpf(dev) && !ec_function)
+ dev->priv.peer_pf_pages += npages;
- mlx5_core_dbg(dev, "err %d\n", err);
+ mlx5_core_dbg(dev, "npages %d, ec_function %d, func_id 0x%x, err %d\n",
+ npages, ec_function, func_id, err);
kvfree(in);
return 0;
@@ -328,7 +336,7 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
out_free:
kvfree(in);
if (notify_fail)
- page_notify_fail(dev, func_id);
+ page_notify_fail(dev, func_id, ec_function);
return err;
}
@@ -364,7 +372,7 @@ static int reclaim_pages_cmd(struct mlx5_core_dev *dev,
}
static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
- int *nclaimed)
+ int *nclaimed, bool ec_function)
{
int outlen = MLX5_ST_SZ_BYTES(manage_pages_out);
u32 in[MLX5_ST_SZ_DW(manage_pages_in)] = {0};
@@ -385,6 +393,7 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
MLX5_SET(manage_pages_in, in, op_mod, MLX5_PAGES_TAKE);
MLX5_SET(manage_pages_in, in, function_id, func_id);
MLX5_SET(manage_pages_in, in, input_num_entries, npages);
+ MLX5_SET(manage_pages_in, in, embedded_cpu_function, ec_function);
mlx5_core_dbg(dev, "npages %d, outlen %d\n", npages, outlen);
err = reclaim_pages_cmd(dev, in, sizeof(in), out, outlen);
@@ -410,6 +419,8 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
dev->priv.fw_pages -= num_claimed;
if (func_id)
dev->priv.vfs_pages -= num_claimed;
+ else if (mlx5_core_is_ecpf(dev) && !ec_function)
+ dev->priv.peer_pf_pages -= num_claimed;
out_free:
kvfree(out);
@@ -423,9 +434,10 @@ static void pages_work_handler(struct work_struct *work)
int err = 0;
if (req->npages < 0)
- err = reclaim_pages(dev, req->func_id, -1 * req->npages, NULL);
+ err = reclaim_pages(dev, req->func_id, -1 * req->npages, NULL,
+ req->ec_function);
else if (req->npages > 0)
- err = give_pages(dev, req->func_id, req->npages, 1);
+ err = give_pages(dev, req->func_id, req->npages, 1, req->ec_function);
if (err)
mlx5_core_warn(dev, "%s fail %d\n",
@@ -434,6 +446,10 @@ static void pages_work_handler(struct work_struct *work)
kfree(req);
}
+enum {
+ EC_FUNCTION_MASK = 0x8000,
+};
+
static int req_pages_handler(struct notifier_block *nb,
unsigned long type, void *data)
{
@@ -441,6 +457,7 @@ static int req_pages_handler(struct notifier_block *nb,
struct mlx5_core_dev *dev;
struct mlx5_priv *priv;
struct mlx5_eqe *eqe;
+ bool ec_function;
u16 func_id;
s32 npages;
@@ -450,6 +467,7 @@ static int req_pages_handler(struct notifier_block *nb,
func_id = be16_to_cpu(eqe->data.req_pages.func_id);
npages = be32_to_cpu(eqe->data.req_pages.num_pages);
+ ec_function = be16_to_cpu(eqe->data.req_pages.ec_function) & EC_FUNCTION_MASK;
mlx5_core_dbg(dev, "page request for func 0x%x, npages %d\n",
func_id, npages);
req = kzalloc(sizeof(*req), GFP_ATOMIC);
@@ -461,6 +479,7 @@ static int req_pages_handler(struct notifier_block *nb,
req->dev = dev;
req->func_id = func_id;
req->npages = npages;
+ req->ec_function = ec_function;
INIT_WORK(&req->work, pages_work_handler);
queue_work(dev->priv.pg_wq, &req->work);
return NOTIFY_OK;
@@ -479,7 +498,7 @@ int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot)
mlx5_core_dbg(dev, "requested %d %s pages for func_id 0x%x\n",
npages, boot ? "boot" : "init", func_id);
- return give_pages(dev, func_id, npages, 0);
+ return give_pages(dev, func_id, npages, 0, mlx5_core_is_ecpf(dev));
}
enum {
@@ -513,7 +532,7 @@ int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev)
fwp = rb_entry(p, struct fw_page, rb_node);
err = reclaim_pages(dev, fwp->func_id,
optimal_reclaimed_pages(),
- &nclaimed);
+ &nclaimed, mlx5_core_is_ecpf(dev));
if (err) {
mlx5_core_warn(dev, "failed reclaiming pages (%d)\n",
@@ -535,6 +554,9 @@ int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev)
WARN(dev->priv.vfs_pages,
"VFs FW pages counter is %d after reclaiming all pages\n",
dev->priv.vfs_pages);
+ WARN(dev->priv.peer_pf_pages,
+ "Peer PF FW pages counter is %d after reclaiming all pages\n",
+ dev->priv.peer_pf_pages);
return 0;
}
@@ -567,10 +589,10 @@ void mlx5_pagealloc_stop(struct mlx5_core_dev *dev)
flush_workqueue(dev->priv.pg_wq);
}
-int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev)
+int mlx5_wait_for_pages(struct mlx5_core_dev *dev, int *pages)
{
unsigned long end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS);
- int prev_vfs_pages = dev->priv.vfs_pages;
+ int prev_pages = *pages;
/* In case of internal error we will free the pages manually later */
if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) {
@@ -578,16 +600,16 @@ int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev)
return 0;
}
- mlx5_core_dbg(dev, "Waiting for %d pages from %s\n", prev_vfs_pages,
+ mlx5_core_dbg(dev, "Waiting for %d pages from %s\n", prev_pages,
dev->priv.name);
- while (dev->priv.vfs_pages) {
+ while (*pages) {
if (time_after(jiffies, end)) {
- mlx5_core_warn(dev, "aborting while there are %d pending pages\n", dev->priv.vfs_pages);
+ mlx5_core_warn(dev, "aborting while there are %d pending pages\n", *pages);
return -ETIMEDOUT;
}
- if (dev->priv.vfs_pages < prev_vfs_pages) {
+ if (*pages < prev_pages) {
end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS);
- prev_vfs_pages = dev->priv.vfs_pages;
+ prev_pages = *pages;
}
msleep(50);
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
index 6e178030d8fb..7b23fa8d2d60 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
@@ -147,7 +147,7 @@ static void mlx5_device_disable_sriov(struct mlx5_core_dev *dev)
if (MLX5_ESWITCH_MANAGER(dev))
mlx5_eswitch_disable_sriov(dev->priv.eswitch);
- if (mlx5_wait_for_vf_pages(dev))
+ if (mlx5_wait_for_pages(dev, &dev->priv.vfs_pages))
mlx5_core_warn(dev, "timeout reclaiming VFs pages\n");
}
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 46223efa1877..f2070350f60a 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -591,7 +591,7 @@ struct mlx5_eqe_cmd {
};
struct mlx5_eqe_page_req {
- u8 rsvd0[2];
+ __be16 ec_function;
__be16 func_id;
__be32 num_pages;
__be32 rsvd1[5];
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 039c9398614c..cce4e8293384 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -522,6 +522,7 @@ struct mlx5_priv {
atomic_t reg_pages;
struct list_head free_list;
int vfs_pages;
+ int peer_pf_pages;
struct mlx5_core_health health;
@@ -652,6 +653,7 @@ struct mlx5_core_dev {
u32 mcam[MLX5_ST_SZ_DW(mcam_reg)];
u32 fpga[MLX5_ST_SZ_DW(fpga_cap)];
u32 qcam[MLX5_ST_SZ_DW(qcam_reg)];
+ u8 embedded_cpu;
} caps;
u64 sys_image_guid;
phys_addr_t iseg_base;
@@ -922,7 +924,7 @@ void mlx5_pagealloc_cleanup(struct mlx5_core_dev *dev);
void mlx5_pagealloc_start(struct mlx5_core_dev *dev);
void mlx5_pagealloc_stop(struct mlx5_core_dev *dev);
void mlx5_core_req_pages_handler(struct mlx5_core_dev *dev, u16 func_id,
- s32 npages);
+ s32 npages, bool ec_function);
int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot);
int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev);
void mlx5_register_debugfs(void);
@@ -1076,6 +1078,11 @@ static inline int mlx5_core_is_pf(struct mlx5_core_dev *dev)
return !(dev->priv.pci_dev_data & MLX5_PCI_DEV_IS_VF);
}
+static inline bool mlx5_core_is_ecpf(struct mlx5_core_dev *dev)
+{
+ return dev->caps.embedded_cpu;
+}
+
#define MLX5_TOTAL_VPORTS(mdev) (1 + pci_sriov_get_totalvfs((mdev)->pdev))
#define MLX5_VPORT_MANAGER(mdev) \
(MLX5_CAP_GEN(mdev, vport_group_manager) && \
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index c5c679390fbd..46799b4c8859 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -4441,7 +4441,8 @@ struct mlx5_ifc_query_pages_out_bits {
u8 syndrome[0x20];
- u8 reserved_at_40[0x10];
+ u8 embedded_cpu_function[0x1];
+ u8 reserved_at_41[0xf];
u8 function_id[0x10];
u8 num_pages[0x20];
@@ -4460,7 +4461,8 @@ struct mlx5_ifc_query_pages_in_bits {
u8 reserved_at_20[0x10];
u8 op_mod[0x10];
- u8 reserved_at_40[0x10];
+ u8 embedded_cpu_function[0x1];
+ u8 reserved_at_41[0xf];
u8 function_id[0x10];
u8 reserved_at_60[0x20];
@@ -5880,7 +5882,8 @@ struct mlx5_ifc_manage_pages_in_bits {
u8 reserved_at_20[0x10];
u8 op_mod[0x10];
- u8 reserved_at_40[0x10];
+ u8 embedded_cpu_function[0x1];
+ u8 reserved_at_41[0xf];
u8 function_id[0x10];
u8 input_num_entries[0x20];
@@ -8749,7 +8752,8 @@ struct mlx5_ifc_initial_seg_bits {
u8 initializing[0x1];
u8 reserved_at_fe1[0x4];
u8 nic_interface_supported[0x3];
- u8 reserved_at_fe8[0x18];
+ u8 embedded_cpu[0x1];
+ u8 reserved_at_fe9[0x17];
struct mlx5_ifc_health_buffer_bits health_buffer;
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 03/15] IB/mlx5: Use unified register/load function for uplink and VF vports
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Mark Bloch
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
IB driver maintains different registration and load function calls
for uplink and VF vports. This is not necessary as they only differ
with each other on their profiles.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/infiniband/hw/mlx5/ib_rep.c | 73 +++++++----------------------
drivers/infiniband/hw/mlx5/ib_rep.h | 10 ++--
drivers/infiniband/hw/mlx5/main.c | 26 +++++-----
3 files changed, 37 insertions(+), 72 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c
index 46a9ddc8ca56..3fb22967c098 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.c
+++ b/drivers/infiniband/hw/mlx5/ib_rep.c
@@ -6,7 +6,7 @@
#include "ib_rep.h"
#include "srq.h"
-static const struct mlx5_ib_profile rep_profile = {
+static const struct mlx5_ib_profile vf_rep_profile = {
STAGE_CREATE(MLX5_IB_STAGE_INIT,
mlx5_ib_stage_init_init,
mlx5_ib_stage_init_cleanup),
@@ -45,31 +45,18 @@ static const struct mlx5_ib_profile rep_profile = {
NULL),
};
-static int
-mlx5_ib_nic_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
-{
- struct mlx5_ib_dev *ibdev;
-
- ibdev = mlx5_ib_rep_to_dev(rep);
- if (!__mlx5_ib_add(ibdev, ibdev->profile))
- return -EINVAL;
- return 0;
-}
-
-static void
-mlx5_ib_nic_rep_unload(struct mlx5_eswitch_rep *rep)
-{
- struct mlx5_ib_dev *ibdev;
-
- ibdev = mlx5_ib_rep_to_dev(rep);
- __mlx5_ib_remove(ibdev, ibdev->profile, MLX5_IB_STAGE_MAX);
-}
-
static int
mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
+#define FDB_UPLINK_VPORT 0xffff
+ const struct mlx5_ib_profile *profile;
struct mlx5_ib_dev *ibdev;
+ if (rep->vport == FDB_UPLINK_VPORT)
+ profile = &uplink_rep_profile;
+ else
+ profile = &vf_rep_profile;
+
ibdev = (struct mlx5_ib_dev *)ib_alloc_device(sizeof(*ibdev));
if (!ibdev)
return -ENOMEM;
@@ -78,7 +65,7 @@ mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
ibdev->mdev = dev;
ibdev->num_ports = max(MLX5_CAP_GEN(dev, num_ports),
MLX5_CAP_GEN(dev, num_vhca_ports));
- if (!__mlx5_ib_add(ibdev, &rep_profile))
+ if (!__mlx5_ib_add(ibdev, profile))
return -EINVAL;
rep->rep_if[REP_IB].priv = ibdev;
@@ -105,15 +92,14 @@ static void *mlx5_ib_vport_get_proto_dev(struct mlx5_eswitch_rep *rep)
return mlx5_ib_rep_to_dev(rep);
}
-static void mlx5_ib_rep_register_vf_vports(struct mlx5_ib_dev *dev)
+void mlx5_ib_register_vport_reps(struct mlx5_core_dev *mdev)
{
- struct mlx5_eswitch *esw = dev->mdev->priv.eswitch;
- int total_vfs = MLX5_TOTAL_VPORTS(dev->mdev);
+ struct mlx5_eswitch *esw = mdev->priv.eswitch;
+ int total_vports = MLX5_TOTAL_VPORTS(mdev);
+ struct mlx5_eswitch_rep_if rep_if = {};
int vport;
- for (vport = 1; vport < total_vfs; vport++) {
- struct mlx5_eswitch_rep_if rep_if = {};
-
+ for (vport = 0; vport < total_vports; vport++) {
rep_if.load = mlx5_ib_vport_rep_load;
rep_if.unload = mlx5_ib_vport_rep_unload;
rep_if.get_proto_dev = mlx5_ib_vport_get_proto_dev;
@@ -121,39 +107,16 @@ static void mlx5_ib_rep_register_vf_vports(struct mlx5_ib_dev *dev)
}
}
-static void mlx5_ib_rep_unregister_vf_vports(struct mlx5_ib_dev *dev)
+void mlx5_ib_unregister_vport_reps(struct mlx5_core_dev *mdev)
{
- struct mlx5_eswitch *esw = dev->mdev->priv.eswitch;
- int total_vfs = MLX5_TOTAL_VPORTS(dev->mdev);
+ struct mlx5_eswitch *esw = mdev->priv.eswitch;
+ int total_vports = MLX5_TOTAL_VPORTS(mdev);
int vport;
- for (vport = 1; vport < total_vfs; vport++)
+ for (vport = total_vports - 1; vport >= 0; vport--)
mlx5_eswitch_unregister_vport_rep(esw, vport, REP_IB);
}
-void mlx5_ib_register_vport_reps(struct mlx5_ib_dev *dev)
-{
- struct mlx5_eswitch *esw = dev->mdev->priv.eswitch;
- struct mlx5_eswitch_rep_if rep_if = {};
-
- rep_if.load = mlx5_ib_nic_rep_load;
- rep_if.unload = mlx5_ib_nic_rep_unload;
- rep_if.get_proto_dev = mlx5_ib_vport_get_proto_dev;
- rep_if.priv = dev;
-
- mlx5_eswitch_register_vport_rep(esw, 0, &rep_if, REP_IB);
-
- mlx5_ib_rep_register_vf_vports(dev);
-}
-
-void mlx5_ib_unregister_vport_reps(struct mlx5_ib_dev *dev)
-{
- struct mlx5_eswitch *esw = dev->mdev->priv.eswitch;
-
- mlx5_ib_rep_unregister_vf_vports(dev); /* VFs vports */
- mlx5_eswitch_unregister_vport_rep(esw, 0, REP_IB); /* UPLINK PF*/
-}
-
u8 mlx5_ib_eswitch_mode(struct mlx5_eswitch *esw)
{
return mlx5_eswitch_mode(esw);
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.h b/drivers/infiniband/hw/mlx5/ib_rep.h
index 2ba73636a2fb..798d41e61fb4 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.h
+++ b/drivers/infiniband/hw/mlx5/ib_rep.h
@@ -10,14 +10,16 @@
#include "mlx5_ib.h"
#ifdef CONFIG_MLX5_ESWITCH
+extern const struct mlx5_ib_profile uplink_rep_profile;
+
u8 mlx5_ib_eswitch_mode(struct mlx5_eswitch *esw);
struct mlx5_ib_dev *mlx5_ib_get_rep_ibdev(struct mlx5_eswitch *esw,
int vport_index);
struct mlx5_ib_dev *mlx5_ib_get_uplink_ibdev(struct mlx5_eswitch *esw);
struct mlx5_eswitch_rep *mlx5_ib_vport_rep(struct mlx5_eswitch *esw,
int vport_index);
-void mlx5_ib_register_vport_reps(struct mlx5_ib_dev *dev);
-void mlx5_ib_unregister_vport_reps(struct mlx5_ib_dev *dev);
+void mlx5_ib_register_vport_reps(struct mlx5_core_dev *mdev);
+void mlx5_ib_unregister_vport_reps(struct mlx5_core_dev *mdev);
int create_flow_rule_vport_sq(struct mlx5_ib_dev *dev,
struct mlx5_ib_sq *sq);
struct net_device *mlx5_ib_get_rep_netdev(struct mlx5_eswitch *esw,
@@ -48,8 +50,8 @@ struct mlx5_eswitch_rep *mlx5_ib_vport_rep(struct mlx5_eswitch *esw,
return NULL;
}
-static inline void mlx5_ib_register_vport_reps(struct mlx5_ib_dev *dev) {}
-static inline void mlx5_ib_unregister_vport_reps(struct mlx5_ib_dev *dev) {}
+static inline void mlx5_ib_register_vport_reps(struct mlx5_core_dev *mdev) {}
+static inline void mlx5_ib_unregister_vport_reps(struct mlx5_core_dev *mdev) {}
static inline int create_flow_rule_vport_sq(struct mlx5_ib_dev *dev,
struct mlx5_ib_sq *sq)
{
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 94fe253d4956..87ce62e44898 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -6386,7 +6386,7 @@ static const struct mlx5_ib_profile pf_profile = {
mlx5_ib_stage_delay_drop_cleanup),
};
-static const struct mlx5_ib_profile nic_rep_profile = {
+const struct mlx5_ib_profile uplink_rep_profile = {
STAGE_CREATE(MLX5_IB_STAGE_INIT,
mlx5_ib_stage_init_init,
mlx5_ib_stage_init_cleanup),
@@ -6479,6 +6479,12 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
printk_once(KERN_INFO "%s", mlx5_version);
+ if (MLX5_ESWITCH_MANAGER(mdev) &&
+ mlx5_ib_eswitch_mode(mdev->priv.eswitch) == SRIOV_OFFLOADS) {
+ mlx5_ib_register_vport_reps(mdev);
+ return mdev;
+ }
+
port_type_cap = MLX5_CAP_GEN(mdev, port_type);
ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap);
@@ -6493,14 +6499,6 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
dev->num_ports = max(MLX5_CAP_GEN(mdev, num_ports),
MLX5_CAP_GEN(mdev, num_vhca_ports));
- if (MLX5_ESWITCH_MANAGER(mdev) &&
- mlx5_ib_eswitch_mode(mdev->priv.eswitch) == SRIOV_OFFLOADS) {
- dev->rep = mlx5_ib_vport_rep(mdev->priv.eswitch, 0);
- dev->profile = &nic_rep_profile;
- mlx5_ib_register_vport_reps(dev);
- return dev;
- }
-
return __mlx5_ib_add(dev, &pf_profile);
}
@@ -6509,6 +6507,11 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context)
struct mlx5_ib_multiport_info *mpi;
struct mlx5_ib_dev *dev;
+ if (MLX5_ESWITCH_MANAGER(mdev) && context == mdev) {
+ mlx5_ib_unregister_vport_reps(mdev);
+ return;
+ }
+
if (mlx5_core_is_mp_slave(mdev)) {
mpi = context;
mutex_lock(&mlx5_ib_multiport_mutex);
@@ -6520,10 +6523,7 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context)
}
dev = context;
- if (dev->profile == &nic_rep_profile)
- mlx5_ib_unregister_vport_reps(dev);
- else
- __mlx5_ib_remove(dev, dev->profile, MLX5_IB_STAGE_MAX);
+ __mlx5_ib_remove(dev, dev->profile, MLX5_IB_STAGE_MAX);
ib_dealloc_device((struct ib_device *)dev);
}
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 05/15] net/mlx5: Update enable HCA dependency
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
With the introduction of ECPF, we require that the ECPF driver will
aways call enable/disable HCA for that PF in the same way a PF does
this for its VFs. The PF is still responsible for calling enable and
disable HCA for its VFs.
To distinguish between the ECPF executing enable/disable HCA for
itself or for the PF, it sets the embedded CPU function bit in the
input params struct of these commands. When the bit is cleared and
function ID is zero, it refers to the peer PF.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/ecpf.c | 76 +++++++++++++++++++
.../net/ethernet/mellanox/mlx5/core/ecpf.h | 4 +
.../net/ethernet/mellanox/mlx5/core/main.c | 14 ++++
include/linux/mlx5/mlx5_ifc.h | 6 +-
4 files changed, 98 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
index 28b8c5c5c8c7..1bcf8b8f9713 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
@@ -7,3 +7,79 @@ bool mlx5_read_embedded_cpu(struct mlx5_core_dev *dev)
{
return (ioread32be(&dev->iseg->initializing) >> MLX5_ECPU_BIT_NUM) & 1;
}
+
+static int mlx5_peer_pf_enable_hca(struct mlx5_core_dev *dev)
+{
+ u32 out[MLX5_ST_SZ_DW(enable_hca_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(enable_hca_in)] = {};
+
+ MLX5_SET(enable_hca_in, in, opcode, MLX5_CMD_OP_ENABLE_HCA);
+ MLX5_SET(enable_hca_in, in, function_id, 0);
+ MLX5_SET(enable_hca_in, in, embedded_cpu_function, 0);
+ return mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out));
+}
+
+static int mlx5_peer_pf_disable_hca(struct mlx5_core_dev *dev)
+{
+ u32 out[MLX5_ST_SZ_DW(disable_hca_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(disable_hca_in)] = {};
+
+ MLX5_SET(disable_hca_in, in, opcode, MLX5_CMD_OP_DISABLE_HCA);
+ MLX5_SET(disable_hca_in, in, function_id, 0);
+ MLX5_SET(enable_hca_in, in, embedded_cpu_function, 0);
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+static int mlx5_peer_pf_init(struct mlx5_core_dev *dev)
+{
+ int err;
+
+ err = mlx5_peer_pf_enable_hca(dev);
+ if (err)
+ mlx5_core_err(dev, "Failed to enable peer PF HCA err(%d)\n",
+ err);
+
+ return err;
+}
+
+static void mlx5_peer_pf_cleanup(struct mlx5_core_dev *dev)
+{
+ int err;
+
+ err = mlx5_peer_pf_disable_hca(dev);
+ if (err) {
+ mlx5_core_err(dev, "Failed to disable peer PF HCA err(%d)\n",
+ err);
+ return;
+ }
+
+ err = mlx5_wait_for_pages(dev, &dev->priv.peer_pf_pages);
+ if (err)
+ mlx5_core_warn(dev, "Timeout reclaiming peer PF pages err(%d)\n",
+ err);
+}
+
+int mlx5_ec_init(struct mlx5_core_dev *dev)
+{
+ int err = 0;
+
+ if (!mlx5_core_is_ecpf(dev))
+ return 0;
+
+ /* ECPF shall enable HCA for peer PF in the same way a PF
+ * does this for its VFs.
+ */
+ err = mlx5_peer_pf_init(dev);
+ if (err)
+ return err;
+
+ return 0;
+}
+
+void mlx5_ec_cleanup(struct mlx5_core_dev *dev)
+{
+ if (!mlx5_core_is_ecpf(dev))
+ return;
+
+ mlx5_peer_pf_cleanup(dev);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
index 8b684f0ab48f..d3d7a00a02ac 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
@@ -14,11 +14,15 @@ enum {
};
bool mlx5_read_embedded_cpu(struct mlx5_core_dev *dev);
+int mlx5_ec_init(struct mlx5_core_dev *dev);
+void mlx5_ec_cleanup(struct mlx5_core_dev *dev);
#else /* CONFIG_MLX5_ESWITCH */
static inline bool
mlx5_read_embedded_cpu(struct mlx5_core_dev *dev) { return false; }
+static inline int mlx5_ec_init(struct mlx5_core_dev *dev) { return 0; }
+static inline void mlx5_ec_cleanup(struct mlx5_core_dev *dev) {}
#endif /* CONFIG_MLX5_ESWITCH */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index bf810fd91454..772dc460cfe6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -615,6 +615,8 @@ int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id)
MLX5_SET(enable_hca_in, in, opcode, MLX5_CMD_OP_ENABLE_HCA);
MLX5_SET(enable_hca_in, in, function_id, func_id);
+ MLX5_SET(enable_hca_in, in, embedded_cpu_function,
+ dev->caps.embedded_cpu);
return mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out));
}
@@ -625,6 +627,8 @@ int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id)
MLX5_SET(disable_hca_in, in, opcode, MLX5_CMD_OP_DISABLE_HCA);
MLX5_SET(disable_hca_in, in, function_id, func_id);
+ MLX5_SET(enable_hca_in, in, embedded_cpu_function,
+ dev->caps.embedded_cpu);
return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
}
@@ -1074,6 +1078,12 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
goto err_sriov;
}
+ err = mlx5_ec_init(dev);
+ if (err) {
+ dev_err(&pdev->dev, "Failed to init embedded CPU\n");
+ goto err_ec;
+ }
+
if (mlx5_device_registered(dev)) {
mlx5_attach_device(dev);
} else {
@@ -1091,6 +1101,9 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
return 0;
err_reg_dev:
+ mlx5_ec_cleanup(dev);
+
+err_ec:
mlx5_sriov_detach(dev);
err_sriov:
@@ -1165,6 +1178,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
if (mlx5_device_registered(dev))
mlx5_detach_device(dev);
+ mlx5_ec_cleanup(dev);
mlx5_sriov_detach(dev);
mlx5_cleanup_fs(dev);
mlx5_accel_ipsec_cleanup(dev);
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 46799b4c8859..1b6d5a563a3a 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -6061,7 +6061,8 @@ struct mlx5_ifc_enable_hca_in_bits {
u8 reserved_at_20[0x10];
u8 op_mod[0x10];
- u8 reserved_at_40[0x10];
+ u8 embedded_cpu_function[0x1];
+ u8 reserved_at_41[0xf];
u8 function_id[0x10];
u8 reserved_at_60[0x20];
@@ -6105,7 +6106,8 @@ struct mlx5_ifc_disable_hca_in_bits {
u8 reserved_at_20[0x10];
u8 op_mod[0x10];
- u8 reserved_at_40[0x10];
+ u8 embedded_cpu_function[0x1];
+ u8 reserved_at_41[0xf];
u8 function_id[0x10];
u8 reserved_at_60[0x20];
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 07/15] net/mlx5: Add host params change event
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
In Embedded CPU (EC) configurations, the EC driver needs to know when
the number of virtual functions change on the corresponding PF at the
host side. This is required so the EC driver can create or destroy
representor net devices that represent the VFs ports.
Whenever a change in the number of VFs occurs, firmware will generate an
event towards the EC which will trigger a work to complete the rest of
the handling. The specifics of the handling will be introduced in a
downstream patch.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 3 +++
drivers/net/ethernet/mellanox/mlx5/core/events.c | 2 ++
include/linux/mlx5/device.h | 2 ++
include/linux/mlx5/driver.h | 5 +++++
4 files changed, 12 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 7092457705a2..5c02f9291799 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -530,6 +530,9 @@ static u64 gather_async_events_mask(struct mlx5_core_dev *dev)
if (MLX5_CAP_GEN(dev, max_num_of_monitor_counters))
async_event_mask |= (1ull << MLX5_EVENT_TYPE_MONITOR_COUNTER);
+ if (mlx5_core_is_ecpf_esw_manager(dev))
+ async_event_mask |= (1ull << MLX5_EVENT_TYPE_HOST_PARAMS_CHANGE);
+
return async_event_mask;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c
index fbc42b7252a9..4f7f776d6332 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/events.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c
@@ -103,6 +103,8 @@ static const char *eqe_type_str(u8 type)
return "MLX5_EVENT_TYPE_STALL_EVENT";
case MLX5_EVENT_TYPE_CMD:
return "MLX5_EVENT_TYPE_CMD";
+ case MLX5_EVENT_TYPE_HOST_PARAMS_CHANGE:
+ return "MLX5_EVENT_TYPE_HOST_PARAMS_CHANGE";
case MLX5_EVENT_TYPE_PAGE_REQUEST:
return "MLX5_EVENT_TYPE_PAGE_REQUEST";
case MLX5_EVENT_TYPE_PAGE_FAULT:
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index f2070350f60a..f93a5598b942 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -342,6 +342,8 @@ enum mlx5_event {
MLX5_EVENT_TYPE_PAGE_FAULT = 0xc,
MLX5_EVENT_TYPE_NIC_VPORT_CHANGE = 0xd,
+ MLX5_EVENT_TYPE_HOST_PARAMS_CHANGE = 0xe,
+
MLX5_EVENT_TYPE_DCT_DRAINED = 0x1c,
MLX5_EVENT_TYPE_FPGA_ERROR = 0x20,
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index cce4e8293384..151563a12fc2 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1083,6 +1083,11 @@ static inline bool mlx5_core_is_ecpf(struct mlx5_core_dev *dev)
return dev->caps.embedded_cpu;
}
+static inline bool mlx5_core_is_ecpf_esw_manager(struct mlx5_core_dev *dev)
+{
+ return dev->caps.embedded_cpu && MLX5_CAP_GEN(dev, eswitch_manager);
+}
+
#define MLX5_TOTAL_VPORTS(mdev) (1 + pci_sriov_get_totalvfs((mdev)->pdev))
#define MLX5_VPORT_MANAGER(mdev) \
(MLX5_CAP_GEN(mdev, vport_group_manager) && \
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 06/15] net/mlx5: Add query host params command
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Eli Cohen, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
The QUERY_HOST_PARAMS command is used by an Embedded CPU Physical
Function (ECPF) driver to identify and retrieve information about the
PF on the host side. E.g, number of virtual functions and PCI BDF.
The number of VFs can be changed on the fly, a function is added to
query current number of VFs and will be used in downstream patches.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 +
.../net/ethernet/mellanox/mlx5/core/ecpf.c | 27 ++++++++++++
.../net/ethernet/mellanox/mlx5/core/ecpf.h | 4 ++
include/linux/mlx5/mlx5_ifc.h | 41 +++++++++++++++++++
4 files changed, 74 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index a25a8c6f938e..46d70eb2d2f7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -316,6 +316,7 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op,
case MLX5_CMD_OP_DESTROY_GENERAL_OBJECT:
case MLX5_CMD_OP_DEALLOC_MEMIC:
case MLX5_CMD_OP_PAGE_FAULT_RESUME:
+ case MLX5_CMD_OP_QUERY_HOST_PARAMS:
return MLX5_CMD_STAT_OK;
case MLX5_CMD_OP_QUERY_HCA_CAP:
@@ -627,6 +628,7 @@ const char *mlx5_command_str(int command)
MLX5_COMMAND_STR_CASE(QUERY_MODIFY_HEADER_CONTEXT);
MLX5_COMMAND_STR_CASE(ALLOC_MEMIC);
MLX5_COMMAND_STR_CASE(DEALLOC_MEMIC);
+ MLX5_COMMAND_STR_CASE(QUERY_HOST_PARAMS);
default: return "unknown command opcode";
}
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
index 1bcf8b8f9713..4746f2d28fb6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c
@@ -83,3 +83,30 @@ void mlx5_ec_cleanup(struct mlx5_core_dev *dev)
mlx5_peer_pf_cleanup(dev);
}
+
+static int mlx5_query_host_params_context(struct mlx5_core_dev *dev,
+ u32 *out, int outlen)
+{
+ u32 in[MLX5_ST_SZ_DW(query_host_params_in)] = {};
+
+ MLX5_SET(query_host_params_in, in, opcode,
+ MLX5_CMD_OP_QUERY_HOST_PARAMS);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, outlen);
+}
+
+int mlx5_query_host_params_num_vfs(struct mlx5_core_dev *dev, int *num_vf)
+{
+ u32 out[MLX5_ST_SZ_DW(query_host_params_out)] = {};
+ int err;
+
+ err = mlx5_query_host_params_context(dev, out, sizeof(out));
+ if (err)
+ return err;
+
+ *num_vf = MLX5_GET(query_host_params_out, out,
+ host_params_context.host_num_of_vfs);
+ mlx5_core_dbg(dev, "host_num_of_vfs %d\n", *num_vf);
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
index d3d7a00a02ac..346372df218f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.h
@@ -16,6 +16,7 @@ enum {
bool mlx5_read_embedded_cpu(struct mlx5_core_dev *dev);
int mlx5_ec_init(struct mlx5_core_dev *dev);
void mlx5_ec_cleanup(struct mlx5_core_dev *dev);
+int mlx5_query_host_params_num_vfs(struct mlx5_core_dev *dev, int *num_vf);
#else /* CONFIG_MLX5_ESWITCH */
@@ -23,6 +24,9 @@ static inline bool
mlx5_read_embedded_cpu(struct mlx5_core_dev *dev) { return false; }
static inline int mlx5_ec_init(struct mlx5_core_dev *dev) { return 0; }
static inline void mlx5_ec_cleanup(struct mlx5_core_dev *dev) {}
+static inline int
+mlx5_query_host_params_num_vfs(struct mlx5_core_dev *dev, int *num_vf)
+{ return -EOPNOTSUPP; }
#endif /* CONFIG_MLX5_ESWITCH */
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 1b6d5a563a3a..565046830559 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -142,6 +142,7 @@ enum {
MLX5_CMD_OP_QUERY_XRQ_DC_PARAMS_ENTRY = 0x725,
MLX5_CMD_OP_SET_XRQ_DC_PARAMS_ENTRY = 0x726,
MLX5_CMD_OP_QUERY_XRQ_ERROR_PARAMS = 0x727,
+ MLX5_CMD_OP_QUERY_HOST_PARAMS = 0x740,
MLX5_CMD_OP_QUERY_VPORT_STATE = 0x750,
MLX5_CMD_OP_MODIFY_VPORT_STATE = 0x751,
MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT = 0x752,
@@ -9522,4 +9523,44 @@ struct mlx5_ifc_mtrc_ctrl_bits {
u8 reserved_at_80[0x180];
};
+struct mlx5_ifc_host_params_context_bits {
+ u8 host_number[0x8];
+ u8 reserved_at_8[0x8];
+ u8 host_num_of_vfs[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 host_pci_bus[0x10];
+
+ u8 reserved_at_40[0x10];
+ u8 host_pci_device[0x10];
+
+ u8 reserved_at_60[0x10];
+ u8 host_pci_function[0x10];
+
+ u8 reserved_at_80[0x180];
+};
+
+struct mlx5_ifc_query_host_params_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_query_host_params_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+
+ struct mlx5_ifc_host_params_context_bits host_params_context;
+
+ u8 reserved_at_280[0x180];
+};
+
#endif /* MLX5_IFC_H */
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 08/15] net/mlx5: Provide an alternative VF upper bound for ECPF
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
ECPF doesn't support SR-IOV, but an ECPF E-Switch manager shall know
the max VFs supported by its peer host PF in order to control those
VF vports.
The current driver implementation uses the total vfs quantity as
provided by the pci sub-system for an upper bound of the VF vports
the e-switch code needs to deal with. This obviously can't work as
is on ECPF e-switch manager. For now, we use a hard coded value of
128 on such systems.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/linux/mlx5/driver.h | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 151563a12fc2..46e0aa52a58a 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1088,7 +1088,16 @@ static inline bool mlx5_core_is_ecpf_esw_manager(struct mlx5_core_dev *dev)
return dev->caps.embedded_cpu && MLX5_CAP_GEN(dev, eswitch_manager);
}
-#define MLX5_TOTAL_VPORTS(mdev) (1 + pci_sriov_get_totalvfs((mdev)->pdev))
+#define MLX5_HOST_PF_MAX_VFS (127u)
+static inline u16 mlx5_core_max_vfs(struct mlx5_core_dev *dev)
+{
+ if (mlx5_core_is_ecpf_esw_manager(dev))
+ return MLX5_HOST_PF_MAX_VFS;
+ else
+ return pci_sriov_get_totalvfs(dev->pdev);
+}
+
+#define MLX5_TOTAL_VPORTS(mdev) (1 + mlx5_core_max_vfs(mdev))
#define MLX5_VPORT_MANAGER(mdev) \
(MLX5_CAP_GEN(mdev, vport_group_manager) && \
(MLX5_CAP_GEN(mdev, port_type) == MLX5_CAP_PORT_TYPE_ETH) && \
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 09/15] net/mlx5: E-Switch, Normalize the name of uplink vport number
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Driver used to name uplink vport as FDB_UPLINK_VPORT, it's hard to
comply with the same naming convention along with the introduction of
other vports. Use MLX5_VPORT as the prefix for such vports and
relocate the uplink vport definition to public header file for the
benefits of both net and IB drivers.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/infiniband/hw/mlx5/ib_rep.c | 4 ++--
.../net/ethernet/mellanox/mlx5/core/en_rep.c | 22 +++++++++----------
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 6 ++---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 8 +++----
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 2 --
.../mellanox/mlx5/core/eswitch_offloads.c | 14 ++++++------
include/linux/mlx5/vport.h | 4 ++++
7 files changed, 30 insertions(+), 30 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c
index 3fb22967c098..99cae9a10195 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.c
+++ b/drivers/infiniband/hw/mlx5/ib_rep.c
@@ -3,6 +3,7 @@
* Copyright (c) 2018 Mellanox Technologies. All rights reserved.
*/
+#include <linux/mlx5/vport.h>
#include "ib_rep.h"
#include "srq.h"
@@ -48,11 +49,10 @@ static const struct mlx5_ib_profile vf_rep_profile = {
static int
mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
-#define FDB_UPLINK_VPORT 0xffff
const struct mlx5_ib_profile *profile;
struct mlx5_ib_dev *ibdev;
- if (rep->vport == FDB_UPLINK_VPORT)
+ if (rep->vport == MLX5_VPORT_UPLINK)
profile = &uplink_rep_profile;
else
profile = &vf_rep_profile;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 96cc0c6a4014..c78d21b2501e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -152,7 +152,7 @@ static void mlx5e_rep_update_hw_counters(struct mlx5e_priv *priv)
struct mlx5e_rep_priv *rpriv = priv->ppriv;
struct mlx5_eswitch_rep *rep = rpriv->rep;
- if (rep->vport == FDB_UPLINK_VPORT)
+ if (rep->vport == MLX5_VPORT_UPLINK)
mlx5e_uplink_rep_update_hw_counters(priv);
else
mlx5e_vf_rep_update_hw_counters(priv);
@@ -1207,7 +1207,7 @@ bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv)
return false;
rep = rpriv->rep;
- return (rep->vport == FDB_UPLINK_VPORT);
+ return (rep->vport == MLX5_VPORT_UPLINK);
}
static bool mlx5e_rep_has_offload_stats(const struct net_device *dev, int attr_id)
@@ -1343,7 +1343,7 @@ static void mlx5e_build_rep_params(struct net_device *netdev)
params->sw_mtu = netdev->mtu;
/* SQ */
- if (rep->vport == FDB_UPLINK_VPORT)
+ if (rep->vport == MLX5_VPORT_UPLINK)
params->log_sq_size = MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE;
else
params->log_sq_size = MLX5E_REP_PARAMS_DEF_LOG_SQ_SIZE;
@@ -1370,7 +1370,7 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev)
struct mlx5_eswitch_rep *rep = rpriv->rep;
struct mlx5_core_dev *mdev = priv->mdev;
- if (rep->vport == FDB_UPLINK_VPORT) {
+ if (rep->vport == MLX5_VPORT_UPLINK) {
SET_NETDEV_DEV(netdev, &priv->mdev->pdev->dev);
netdev->netdev_ops = &mlx5e_netdev_ops_uplink_rep;
/* we want a persistent mac for the uplink rep */
@@ -1402,7 +1402,7 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev)
netdev->hw_features |= NETIF_F_TSO6;
netdev->hw_features |= NETIF_F_RXCSUM;
- if (rep->vport != FDB_UPLINK_VPORT)
+ if (rep->vport != MLX5_VPORT_UPLINK)
netdev->features |= NETIF_F_VLAN_CHALLENGED;
netdev->features |= netdev->hw_features;
@@ -1555,7 +1555,7 @@ static int mlx5e_init_rep_tx(struct mlx5e_priv *priv)
return err;
}
- if (rpriv->rep->vport == FDB_UPLINK_VPORT) {
+ if (rpriv->rep->vport == MLX5_VPORT_UPLINK) {
uplink_priv = &rpriv->uplink_priv;
/* init shared tc flow table */
@@ -1591,7 +1591,7 @@ static void mlx5e_cleanup_rep_tx(struct mlx5e_priv *priv)
for (tc = 0; tc < priv->profile->max_tc; tc++)
mlx5e_destroy_tis(priv->mdev, priv->tisn[tc]);
- if (rpriv->rep->vport == FDB_UPLINK_VPORT) {
+ if (rpriv->rep->vport == MLX5_VPORT_UPLINK) {
/* clean indirect TC block notifications */
unregister_netdevice_notifier(&rpriv->uplink_priv.netdevice_nb);
mlx5e_rep_indr_clean_block_privs(rpriv);
@@ -1710,7 +1710,7 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
rpriv->rep = rep;
nch = mlx5e_get_max_num_channels(dev);
- profile = (rep->vport == FDB_UPLINK_VPORT) ? &mlx5e_uplink_rep_profile : &mlx5e_vf_rep_profile;
+ profile = (rep->vport == MLX5_VPORT_UPLINK) ? &mlx5e_uplink_rep_profile : &mlx5e_vf_rep_profile;
netdev = mlx5e_create_netdev(dev, profile, nch, rpriv);
if (!netdev) {
pr_warn("Failed to create representor netdev for vport %d\n",
@@ -1723,7 +1723,7 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
rep->rep_if[REP_ETH].priv = rpriv;
INIT_LIST_HEAD(&rpriv->vport_sqs_list);
- if (rep->vport == FDB_UPLINK_VPORT) {
+ if (rep->vport == MLX5_VPORT_UPLINK) {
err = mlx5e_create_mdev_resources(dev);
if (err)
goto err_destroy_netdev;
@@ -1759,7 +1759,7 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
mlx5e_detach_netdev(netdev_priv(netdev));
err_destroy_mdev_resources:
- if (rep->vport == FDB_UPLINK_VPORT)
+ if (rep->vport == MLX5_VPORT_UPLINK)
mlx5e_destroy_mdev_resources(dev);
err_destroy_netdev:
@@ -1779,7 +1779,7 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep)
unregister_netdev(netdev);
mlx5e_rep_neigh_cleanup(rpriv);
mlx5e_detach_netdev(priv);
- if (rep->vport == FDB_UPLINK_VPORT)
+ if (rep->vport == MLX5_VPORT_UPLINK)
mlx5e_destroy_mdev_resources(priv->mdev);
mlx5e_destroy_netdev(priv);
kfree(ppriv); /* mlx5e_rep_priv */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index cae6c6d48984..1a73e661056a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1834,7 +1834,7 @@ static int parse_cls_flower(struct mlx5e_priv *priv,
if (!err && (flow->flags & MLX5E_TC_FLOW_ESWITCH)) {
rep = rpriv->rep;
- if (rep->vport != FDB_UPLINK_VPORT &&
+ if (rep->vport != MLX5_VPORT_UPLINK &&
(esw->offloads.inline_mode != MLX5_INLINE_MODE_NONE &&
esw->offloads.inline_mode < match_level)) {
NL_SET_ERR_MSG_MOD(extack,
@@ -2724,7 +2724,7 @@ static struct rhashtable *get_tc_ht(struct mlx5e_priv *priv, int flags)
static bool is_peer_flow_needed(struct mlx5e_tc_flow *flow)
{
struct mlx5_esw_flow_attr *attr = flow->esw_attr;
- bool is_rep_ingress = attr->in_rep->vport != FDB_UPLINK_VPORT &&
+ bool is_rep_ingress = attr->in_rep->vport != MLX5_VPORT_UPLINK &&
flow->flags & MLX5E_TC_FLOW_INGRESS;
bool act_is_encap = !!(attr->action &
MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT);
@@ -2849,7 +2849,7 @@ static int mlx5e_tc_add_fdb_peer_flow(struct tc_cls_flower_offload *f,
* original flow and packets redirected from uplink use the
* peer mdev.
*/
- if (flow->esw_attr->in_rep->vport == FDB_UPLINK_VPORT)
+ if (flow->esw_attr->in_rep->vport == MLX5_VPORT_UPLINK)
in_mdev = peer_priv->mdev;
else
in_mdev = priv->mdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index d7382892e81c..0db56b4b7009 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -40,8 +40,6 @@
#include "eswitch.h"
#include "fs_core.h"
-#define UPLINK_VPORT 0xFFFF
-
enum {
MLX5_ACTION_NONE = 0,
MLX5_ACTION_ADD = 1,
@@ -188,7 +186,7 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u16 vport, bool rx_rule,
misc_parameters);
mc_misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
misc_parameters);
- MLX5_SET(fte_match_set_misc, mv_misc, source_port, UPLINK_VPORT);
+ MLX5_SET(fte_match_set_misc, mv_misc, source_port, MLX5_VPORT_UPLINK);
MLX5_SET_TO_ONES(fte_match_set_misc, mc_misc, source_port);
}
@@ -499,7 +497,7 @@ static int esw_add_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
return -ENOMEM;
esw_mc->uplink_rule = /* Forward MC MAC to Uplink */
- esw_fdb_set_vport_rule(esw, mac, UPLINK_VPORT);
+ esw_fdb_set_vport_rule(esw, mac, MLX5_VPORT_UPLINK);
/* Add this multicast mac to all the mc promiscuous vports */
update_allmulti_vports(esw, vaddr, esw_mc);
@@ -736,7 +734,7 @@ static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u16 vport_num,
if (!allmulti_addr->uplink_rule)
allmulti_addr->uplink_rule =
esw_fdb_set_vport_allmulti_rule(esw,
- UPLINK_VPORT);
+ MLX5_VPORT_UPLINK);
allmulti_addr->refcnt++;
} else if (vport->allmulti_rule) {
mlx5_del_flow_rules(vport->allmulti_rule);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9c89eea9b2c3..94da74b1e6ea 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -49,8 +49,6 @@
#define MLX5_MAX_MC_PER_VPORT(dev) \
(1 << MLX5_CAP_GEN(dev, log_max_current_mc_list))
-#define FDB_UPLINK_VPORT 0xffff
-
#define MLX5_MIN_BW_SHARE 1
#define MLX5_RATE_TO_BW_SHARE(rate, divider, limit) \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 53065b6ae593..b0b1267eab07 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -359,15 +359,15 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr,
in_rep = attr->in_rep;
out_rep = attr->dests[0].rep;
- if (push && in_rep->vport == FDB_UPLINK_VPORT)
+ if (push && in_rep->vport == MLX5_VPORT_UPLINK)
goto out_notsupp;
- if (pop && out_rep->vport == FDB_UPLINK_VPORT)
+ if (pop && out_rep->vport == MLX5_VPORT_UPLINK)
goto out_notsupp;
/* vport has vlan push configured, can't offload VF --> wire rules w.o it */
if (!push && !pop && fwd)
- if (in_rep->vlan && out_rep->vport == FDB_UPLINK_VPORT)
+ if (in_rep->vlan && out_rep->vport == MLX5_VPORT_UPLINK)
goto out_notsupp;
/* protects against (1) setting rules with different vlans to push and
@@ -409,7 +409,7 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
if (!push && !pop && fwd) {
/* tracks VF --> wire rules without vlan push action */
- if (attr->dests[0].rep->vport == FDB_UPLINK_VPORT) {
+ if (attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) {
vport->vlan_refcount++;
attr->vlan_handled = true;
}
@@ -469,7 +469,7 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
if (!push && !pop && fwd) {
/* tracks VF --> wire rules without vlan push action */
- if (attr->dests[0].rep->vport == FDB_UPLINK_VPORT)
+ if (attr->dests[0].rep->vport == MLX5_VPORT_UPLINK)
vport->vlan_refcount--;
return 0;
@@ -1227,7 +1227,7 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
ether_addr_copy(rep->hw_id, hw_id);
}
- offloads->vport_reps[0].vport = FDB_UPLINK_VPORT;
+ offloads->vport_reps[0].vport = MLX5_VPORT_UPLINK;
return 0;
}
@@ -1811,7 +1811,7 @@ void *mlx5_eswitch_get_proto_dev(struct mlx5_eswitch *esw,
struct mlx5_esw_offload *offloads = &esw->offloads;
struct mlx5_eswitch_rep *rep;
- if (vport == FDB_UPLINK_VPORT)
+ if (vport == MLX5_VPORT_UPLINK)
vport = UPLINK_REP_INDEX;
rep = &offloads->vport_reps[vport];
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 1654b911cdb2..2e2928eacd97 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -42,6 +42,10 @@ enum {
MLX5_CAP_INLINE_MODE_NOT_REQUIRED,
};
+enum {
+ MLX5_VPORT_UPLINK = 0xffff
+};
+
u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport);
int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod,
u16 vport, u8 state);
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 10/15] net/mlx5: Relocate vport macros to the vport header file
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
These are two macros in the driver general header which deal with the
number of total vports and if a vport is vport manager. Such macros
are vport entities, better to place them at the vport header file.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 1 +
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 1 +
include/linux/mlx5/driver.h | 6 ------
include/linux/mlx5/vport.h | 7 +++++++
4 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 5c02f9291799..bb6e5b5d9681 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -34,6 +34,7 @@
#include <linux/notifier.h>
#include <linux/module.h>
#include <linux/mlx5/driver.h>
+#include <linux/mlx5/vport.h>
#include <linux/mlx5/eq.h>
#include <linux/mlx5/cmd.h>
#ifdef CONFIG_RFS_ACCEL
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 79f122b45def..b6a7bc8f667c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -32,6 +32,7 @@
#include <linux/mutex.h>
#include <linux/mlx5/driver.h>
+#include <linux/mlx5/vport.h>
#include <linux/mlx5/eswitch.h>
#include "mlx5_core.h"
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 46e0aa52a58a..c5454f985e1d 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1097,12 +1097,6 @@ static inline u16 mlx5_core_max_vfs(struct mlx5_core_dev *dev)
return pci_sriov_get_totalvfs(dev->pdev);
}
-#define MLX5_TOTAL_VPORTS(mdev) (1 + mlx5_core_max_vfs(mdev))
-#define MLX5_VPORT_MANAGER(mdev) \
- (MLX5_CAP_GEN(mdev, vport_group_manager) && \
- (MLX5_CAP_GEN(mdev, port_type) == MLX5_CAP_PORT_TYPE_ETH) && \
- mlx5_core_is_pf(mdev))
-
static inline int mlx5_get_gid_table_len(u16 param)
{
if (param > 4) {
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 2e2928eacd97..28f47e868fbc 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -36,6 +36,13 @@
#include <linux/mlx5/driver.h>
#include <linux/mlx5/device.h>
+#define MLX5_TOTAL_VPORTS(mdev) (1 + mlx5_core_max_vfs(mdev))
+
+#define MLX5_VPORT_MANAGER(mdev) \
+ (MLX5_CAP_GEN(mdev, vport_group_manager) && \
+ (MLX5_CAP_GEN(mdev, port_type) == MLX5_CAP_PORT_TYPE_ETH) && \
+ mlx5_core_is_pf(mdev))
+
enum {
MLX5_CAP_INLINE_MODE_L2,
MLX5_CAP_INLINE_MODE_VPORT_CONTEXT,
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 11/15] net/mlx5: E-Switch, Avoid magic numbers when initializing offloads mode
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Bodong Wang, Or Gerlitz
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
When dealing with the offloads mode initialization, driver refers to
the number of VFs and add magic number one (1) to take account of the
uplink. This is not clear and will make the code less readable after
adding other vports (e.g. host PF). As these are special vports
compared to VF vports, add a helper macro to denote such special
vports and eliminate the use of magic number.
Moreover, when creating offloads flow table and groups, the driver
reserves two more slots for UC and MC miss rules. Replace this magic
number with a helper macro as well.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 2 +-
.../mellanox/mlx5/core/eswitch_offloads.c | 25 +++++++++++--------
include/linux/mlx5/vport.h | 4 ++-
3 files changed, 19 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 0db56b4b7009..49a9e3877d2c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1638,7 +1638,7 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
} else {
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_ETH);
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB);
- err = esw_offloads_init(esw, nvfs + 1);
+ err = esw_offloads_init(esw, nvfs + MLX5_SPECIAL_VPORTS);
}
if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index b0b1267eab07..1496e82b5108 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -46,6 +46,11 @@ enum {
FDB_SLOW_PATH
};
+/* There are two match-all miss flows, one for unicast dst mac and
+ * one for multicast.
+ */
+#define MLX5_ESW_MISS_FLOWS (2)
+
#define fdb_prio_table(esw, chain, prio, level) \
(esw)->fdb_table.offloads.fdb_prio[(chain)][(prio)][(level)]
@@ -904,8 +909,8 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
esw->fdb_table.offloads.fdb_left[i] =
ESW_POOLS[i] <= fdb_max ? ESW_SIZE / ESW_POOLS[i] : 0;
- table_size = nvports * MAX_SQ_NVPORTS + MAX_PF_SQ + 2 +
- esw->total_vports;
+ table_size = nvports * MAX_SQ_NVPORTS + MAX_PF_SQ +
+ MLX5_ESW_MISS_FLOWS + esw->total_vports;
/* create the slow path fdb with encap set, so further table instances
* can be created at run time while VFs are probed if the FW allows that.
@@ -999,7 +1004,8 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
dmac[0] = 0x01;
MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix);
- MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + 2);
+ MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
+ ix + MLX5_ESW_MISS_FLOWS);
g = mlx5_create_flow_group(fdb, flow_group_in);
if (IS_ERR(g)) {
@@ -1048,7 +1054,7 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw)
esw_destroy_offloads_fast_fdb_tables(esw);
}
-static int esw_create_offloads_table(struct mlx5_eswitch *esw)
+static int esw_create_offloads_table(struct mlx5_eswitch *esw, int nvports)
{
struct mlx5_flow_table_attr ft_attr = {};
struct mlx5_core_dev *dev = esw->dev;
@@ -1062,7 +1068,7 @@ static int esw_create_offloads_table(struct mlx5_eswitch *esw)
return -EOPNOTSUPP;
}
- ft_attr.max_fte = dev->priv.sriov.num_vfs + 2;
+ ft_attr.max_fte = nvports + MLX5_ESW_MISS_FLOWS;
ft_offloads = mlx5_create_flow_table(ns, &ft_attr);
if (IS_ERR(ft_offloads)) {
@@ -1082,16 +1088,15 @@ static void esw_destroy_offloads_table(struct mlx5_eswitch *esw)
mlx5_destroy_flow_table(offloads->ft_offloads);
}
-static int esw_create_vport_rx_group(struct mlx5_eswitch *esw)
+static int esw_create_vport_rx_group(struct mlx5_eswitch *esw, int nvports)
{
int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
struct mlx5_flow_group *g;
- struct mlx5_priv *priv = &esw->dev->priv;
u32 *flow_group_in;
void *match_criteria, *misc;
int err = 0;
- int nvports = priv->sriov.num_vfs + 2;
+ nvports = nvports + MLX5_ESW_MISS_FLOWS;
flow_group_in = kvzalloc(inlen, GFP_KERNEL);
if (!flow_group_in)
return -ENOMEM;
@@ -1407,11 +1412,11 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports)
if (err)
return err;
- err = esw_create_offloads_table(esw);
+ err = esw_create_offloads_table(esw, nvports);
if (err)
goto create_ft_err;
- err = esw_create_vport_rx_group(esw);
+ err = esw_create_vport_rx_group(esw, nvports);
if (err)
goto create_fg_err;
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 28f47e868fbc..3bc05449ac39 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -36,7 +36,9 @@
#include <linux/mlx5/driver.h>
#include <linux/mlx5/device.h>
-#define MLX5_TOTAL_VPORTS(mdev) (1 + mlx5_core_max_vfs(mdev))
+#define MLX5_VPORT_PF_PLACEHOLDER (1u)
+#define MLX5_SPECIAL_VPORTS (MLX5_VPORT_PF_PLACEHOLDER)
+#define MLX5_TOTAL_VPORTS(mdev) (MLX5_SPECIAL_VPORTS + mlx5_core_max_vfs(mdev))
#define MLX5_VPORT_MANAGER(mdev) \
(MLX5_CAP_GEN(mdev, vport_group_manager) && \
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 13/15] net/mlx5: Add new fields to Port Type and Speed register
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Aya Levin, Eran Ben Elisha
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Aya Levin <ayal@mellanox.com>
Register Port Type and Speed (PTYS) introduces three new fields
extending the speed/protocols the can be reported and configured.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/linux/mlx5/mlx5_ifc.h | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 565046830559..5decffe565fb 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -7826,21 +7826,23 @@ struct mlx5_ifc_ptys_reg_bits {
u8 proto_mask[0x3];
u8 an_status[0x4];
- u8 reserved_at_24[0x3c];
+ u8 reserved_at_24[0x1c];
+
+ u8 ext_eth_proto_capability[0x20];
u8 eth_proto_capability[0x20];
u8 ib_link_width_capability[0x10];
u8 ib_proto_capability[0x10];
- u8 reserved_at_a0[0x20];
+ u8 ext_eth_proto_admin[0x20];
u8 eth_proto_admin[0x20];
u8 ib_link_width_admin[0x10];
u8 ib_proto_admin[0x10];
- u8 reserved_at_100[0x20];
+ u8 ext_eth_proto_oper[0x20];
u8 eth_proto_oper[0x20];
@@ -8289,7 +8291,9 @@ struct mlx5_ifc_mpegc_reg_bits {
struct mlx5_ifc_pcam_enhanced_features_bits {
u8 reserved_at_0[0x6d];
u8 rx_icrc_encapsulated_counter[0x1];
- u8 reserved_at_6e[0x8];
+ u8 reserved_at_6e[0x4];
+ u8 ptys_extended_ethernet[0x1];
+ u8 reserved_at_73[0x3];
u8 pfcc_mask[0x1];
u8 reserved_at_77[0x3];
u8 per_lane_error_counters[0x1];
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 12/15] net/mlx5: Refactor queries to speed fields in Port Type and Speed register
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Aya Levin, Eran Ben Elisha
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Aya Levin <ayal@mellanox.com>
This patch fascicles queries to speed related fields in Port Type and
Speed register (PTYS) into a single API. I addition, this patch
refactors functions which serves only Ethernet driver: remove the
protocol type as an input parameter, move code from 'core' directory
into 'en' directory and add 'eth' prefix to the function's name. The
patch also encapsulates functions that are not used outside the Ethernet
driver removes redundant include files.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/infiniband/hw/mlx5/main.c | 6 +-
.../net/ethernet/mellanox/mlx5/core/en/port.c | 75 +++++++++++--
.../net/ethernet/mellanox/mlx5/core/en/port.h | 12 ++
.../ethernet/mellanox/mlx5/core/en_ethtool.c | 23 ++--
.../net/ethernet/mellanox/mlx5/core/port.c | 106 ------------------
include/linux/mlx5/port.h | 11 --
6 files changed, 91 insertions(+), 142 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 87ce62e44898..efd08b41126c 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -393,6 +393,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
struct ib_port_attr *props)
{
struct mlx5_ib_dev *dev = to_mdev(device);
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)] = {0};
struct mlx5_core_dev *mdev;
struct net_device *ndev, *upper;
enum ib_mtu ndev_ib_mtu;
@@ -416,10 +417,11 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
/* Possible bad flows are checked before filling out props so in case
* of an error it will still be zeroed out.
*/
- err = mlx5_query_port_eth_proto_oper(mdev, ð_prot_oper,
- mdev_port_num);
+ err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN,
+ mdev_port_num);
if (err)
goto out;
+ eth_prot_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
props->active_width = IB_WIDTH_4X;
props->active_speed = IB_SPEED_QDR;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
index 4a37713023be..9a1c2b2f87d8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
@@ -63,6 +63,67 @@ static const u32 mlx5e_link_speed[MLX5E_LINK_MODES_NUMBER] = {
[MLX5E_50GBASE_KR2] = 50000,
};
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port,
+ struct mlx5e_port_eth_proto *eproto)
+{
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+ int err;
+
+ if (!eproto)
+ return -EINVAL;
+
+ err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, port);
+ if (err)
+ return err;
+
+ eproto->cap = MLX5_GET(ptys_reg, out, eth_proto_capability);
+ eproto->admin = MLX5_GET(ptys_reg, out, eth_proto_admin);
+ eproto->oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
+ return 0;
+}
+
+void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
+ u8 *an_disable_cap, u8 *an_disable_admin)
+{
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+
+ *an_status = 0;
+ *an_disable_cap = 0;
+ *an_disable_admin = 0;
+
+ if (mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, 1))
+ return;
+
+ *an_status = MLX5_GET(ptys_reg, out, an_status);
+ *an_disable_cap = MLX5_GET(ptys_reg, out, an_disable_cap);
+ *an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin);
+}
+
+int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
+ u32 proto_admin)
+{
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+ u32 in[MLX5_ST_SZ_DW(ptys_reg)];
+ u8 an_disable_admin;
+ u8 an_disable_cap;
+ u8 an_status;
+
+ mlx5_port_query_eth_autoneg(dev, &an_status, &an_disable_cap,
+ &an_disable_admin);
+ if (!an_disable_cap && an_disable)
+ return -EPERM;
+
+ memset(in, 0, sizeof(in));
+
+ MLX5_SET(ptys_reg, in, local_port, 1);
+ MLX5_SET(ptys_reg, in, an_disable_admin, an_disable);
+ MLX5_SET(ptys_reg, in, proto_mask, MLX5_PTYS_EN);
+ MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin);
+
+ return mlx5_core_access_reg(dev, in, sizeof(in), out,
+ sizeof(out), MLX5_REG_PTYS, 0, 1);
+}
+
u32 mlx5e_port_ptys2speed(u32 eth_proto_oper)
{
unsigned long temp = eth_proto_oper;
@@ -78,16 +139,14 @@ u32 mlx5e_port_ptys2speed(u32 eth_proto_oper)
int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)] = {};
- u32 eth_proto_oper;
+ struct mlx5e_port_eth_proto eproto;
int err;
- err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN, 1);
+ err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
if (err)
return err;
- eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
- *speed = mlx5e_port_ptys2speed(eth_proto_oper);
+ *speed = mlx5e_port_ptys2speed(eproto.oper);
if (!(*speed))
err = -EINVAL;
@@ -96,17 +155,17 @@ int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
{
+ struct mlx5e_port_eth_proto eproto;
u32 max_speed = 0;
- u32 proto_cap;
int err;
int i;
- err = mlx5_query_port_proto_cap(mdev, &proto_cap, MLX5_PTYS_EN);
+ err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
if (err)
return err;
for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i)
- if (proto_cap & MLX5E_PROT_MASK(i))
+ if (eproto.cap & MLX5E_PROT_MASK(i))
max_speed = max(max_speed, mlx5e_link_speed[i]);
*speed = max_speed;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
index cd2160b8c9bf..4bdab8be10af 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
@@ -36,6 +36,18 @@
#include <linux/mlx5/driver.h>
#include "en.h"
+struct mlx5e_port_eth_proto {
+ u32 cap;
+ u32 admin;
+ u32 oper;
+};
+
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port,
+ struct mlx5e_port_eth_proto *eproto);
+void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
+ u8 *an_disable_cap, u8 *an_disable_admin);
+int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
+ u32 proto_admin);
u32 mlx5e_port_ptys2speed(u32 eth_proto_oper);
int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index c9df08133718..c29e141d72fb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -882,7 +882,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
const struct ethtool_link_ksettings *link_ksettings)
{
struct mlx5_core_dev *mdev = priv->mdev;
- u32 eth_proto_cap, eth_proto_admin;
+ struct mlx5e_port_eth_proto eproto;
bool an_changes = false;
u8 an_disable_admin;
u8 an_disable_cap;
@@ -898,14 +898,14 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
mlx5e_ethtool2ptys_adver_link(link_ksettings->link_modes.advertising) :
mlx5e_port_speed2linkmodes(speed);
- err = mlx5_query_port_proto_cap(mdev, ð_proto_cap, MLX5_PTYS_EN);
+ err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
if (err) {
- netdev_err(priv->netdev, "%s: query port eth proto cap failed: %d\n",
+ netdev_err(priv->netdev, "%s: query port eth proto failed: %d\n",
__func__, err);
goto out;
}
- link_modes = link_modes & eth_proto_cap;
+ link_modes = link_modes & eproto.cap;
if (!link_modes) {
netdev_err(priv->netdev, "%s: Not supported link mode(s) requested",
__func__);
@@ -913,24 +913,17 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
goto out;
}
- err = mlx5_query_port_proto_admin(mdev, ð_proto_admin, MLX5_PTYS_EN);
- if (err) {
- netdev_err(priv->netdev, "%s: query port eth proto admin failed: %d\n",
- __func__, err);
- goto out;
- }
-
- mlx5_query_port_autoneg(mdev, MLX5_PTYS_EN, &an_status,
- &an_disable_cap, &an_disable_admin);
+ mlx5_port_query_eth_autoneg(mdev, &an_status, &an_disable_cap,
+ &an_disable_admin);
an_disable = link_ksettings->base.autoneg == AUTONEG_DISABLE;
an_changes = ((!an_disable && an_disable_admin) ||
(an_disable && !an_disable_admin));
- if (!an_changes && link_modes == eth_proto_admin)
+ if (!an_changes && link_modes == eproto.admin)
goto out;
- mlx5_set_port_ptys(mdev, an_disable, link_modes, MLX5_PTYS_EN);
+ mlx5_port_set_eth_ptys(mdev, an_disable, link_modes);
mlx5_toggle_port_link(mdev);
out:
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index 2b82f35f4c35..b81542820528 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -30,10 +30,7 @@
* SOFTWARE.
*/
-#include <linux/module.h>
-#include <linux/mlx5/driver.h>
#include <linux/mlx5/port.h>
-#include <linux/mlx5/cmd.h>
#include "mlx5_core.h"
int mlx5_core_access_reg(struct mlx5_core_dev *dev, void *data_in,
@@ -157,44 +154,6 @@ int mlx5_set_port_beacon(struct mlx5_core_dev *dev, u16 beacon_duration)
sizeof(out), MLX5_REG_MLCR, 0, 1);
}
-int mlx5_query_port_proto_cap(struct mlx5_core_dev *dev,
- u32 *proto_cap, int proto_mask)
-{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)];
- int err;
-
- err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1);
- if (err)
- return err;
-
- if (proto_mask == MLX5_PTYS_EN)
- *proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability);
- else
- *proto_cap = MLX5_GET(ptys_reg, out, ib_proto_capability);
-
- return 0;
-}
-EXPORT_SYMBOL_GPL(mlx5_query_port_proto_cap);
-
-int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev,
- u32 *proto_admin, int proto_mask)
-{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)];
- int err;
-
- err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1);
- if (err)
- return err;
-
- if (proto_mask == MLX5_PTYS_EN)
- *proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin);
- else
- *proto_admin = MLX5_GET(ptys_reg, out, ib_proto_admin);
-
- return 0;
-}
-EXPORT_SYMBOL_GPL(mlx5_query_port_proto_admin);
-
int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
u8 *link_width_oper, u8 local_port)
{
@@ -211,23 +170,6 @@ int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
}
EXPORT_SYMBOL_GPL(mlx5_query_port_link_width_oper);
-int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
- u32 *proto_oper, u8 local_port)
-{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)];
- int err;
-
- err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN,
- local_port);
- if (err)
- return err;
-
- *proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
-
- return 0;
-}
-EXPORT_SYMBOL(mlx5_query_port_eth_proto_oper);
-
int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
u8 *proto_oper, u8 local_port)
{
@@ -245,35 +187,6 @@ int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
}
EXPORT_SYMBOL(mlx5_query_port_ib_proto_oper);
-int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
- u32 proto_admin, int proto_mask)
-{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)];
- u32 in[MLX5_ST_SZ_DW(ptys_reg)];
- u8 an_disable_admin;
- u8 an_disable_cap;
- u8 an_status;
-
- mlx5_query_port_autoneg(dev, proto_mask, &an_status,
- &an_disable_cap, &an_disable_admin);
- if (!an_disable_cap && an_disable)
- return -EPERM;
-
- memset(in, 0, sizeof(in));
-
- MLX5_SET(ptys_reg, in, local_port, 1);
- MLX5_SET(ptys_reg, in, an_disable_admin, an_disable);
- MLX5_SET(ptys_reg, in, proto_mask, proto_mask);
- if (proto_mask == MLX5_PTYS_EN)
- MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin);
- else
- MLX5_SET(ptys_reg, in, ib_proto_admin, proto_admin);
-
- return mlx5_core_access_reg(dev, in, sizeof(in), out,
- sizeof(out), MLX5_REG_PTYS, 0, 1);
-}
-EXPORT_SYMBOL_GPL(mlx5_set_port_ptys);
-
/* This function should be used after setting a port register only */
void mlx5_toggle_port_link(struct mlx5_core_dev *dev)
{
@@ -606,25 +519,6 @@ int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx)
}
EXPORT_SYMBOL_GPL(mlx5_query_port_pfc);
-void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask,
- u8 *an_status,
- u8 *an_disable_cap, u8 *an_disable_admin)
-{
- u32 out[MLX5_ST_SZ_DW(ptys_reg)];
-
- *an_status = 0;
- *an_disable_cap = 0;
- *an_disable_admin = 0;
-
- if (mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1))
- return;
-
- *an_status = MLX5_GET(ptys_reg, out, an_status);
- *an_disable_cap = MLX5_GET(ptys_reg, out, an_disable_cap);
- *an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin);
-}
-EXPORT_SYMBOL_GPL(mlx5_query_port_autoneg);
-
int mlx5_max_tc(struct mlx5_core_dev *mdev)
{
u8 num_tc = MLX5_CAP_GEN(mdev, max_tc) ? : 8;
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index bf4bc01ffb0c..5be7eefa6d75 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -110,27 +110,16 @@ enum mlx5e_connector_type {
int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
int ptys_size, int proto_mask, u8 local_port);
-int mlx5_query_port_proto_cap(struct mlx5_core_dev *dev,
- u32 *proto_cap, int proto_mask);
-int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev,
- u32 *proto_admin, int proto_mask);
int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
u8 *link_width_oper, u8 local_port);
int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
u8 *proto_oper, u8 local_port);
-int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
- u32 *proto_oper, u8 local_port);
-int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
- u32 proto_admin, int proto_mask);
void mlx5_toggle_port_link(struct mlx5_core_dev *dev);
int mlx5_set_port_admin_status(struct mlx5_core_dev *dev,
enum mlx5_port_status status);
int mlx5_query_port_admin_status(struct mlx5_core_dev *dev,
enum mlx5_port_status *status);
int mlx5_set_port_beacon(struct mlx5_core_dev *dev, u16 beacon_duration);
-void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask,
- u8 *an_status,
- u8 *an_disable_cap, u8 *an_disable_admin);
int mlx5_set_port_mtu(struct mlx5_core_dev *dev, u16 mtu, u8 port);
void mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, u16 *max_mtu, u8 port);
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 14/15] net/mlx5: Add support to ext_* fields introduced in Port Type and Speed register
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky
Cc: netdev, linux-rdma, Aya Levin, Eran Ben Elisha
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Aya Levin <ayal@mellanox.com>
This patch exposes new link modes (including 50Gbps per lane), and ext_*
fields which describes the new link modes in Port Type and Speed
register (PTYS).
Access functions, translation functions (speed <-> HW bits) and
link max speed function were modified.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/infiniband/hw/mlx5/main.c | 3 +-
.../net/ethernet/mellanox/mlx5/core/en/port.c | 85 ++++++++++++++-----
.../net/ethernet/mellanox/mlx5/core/en/port.h | 8 +-
.../ethernet/mellanox/mlx5/core/en_ethtool.c | 9 +-
include/linux/mlx5/port.h | 19 +++++
5 files changed, 94 insertions(+), 30 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index efd08b41126c..3677c00fa3bb 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -421,7 +421,8 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
mdev_port_num);
if (err)
goto out;
- eth_prot_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
+ eth_prot_oper = MLX5_GET_ETH_PROTO(ptys_reg, out, false,
+ eth_proto_oper);
props->active_width = IB_WIDTH_4X;
props->active_speed = IB_SPEED_QDR;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
index 9a1c2b2f87d8..122927f3a600 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c
@@ -63,7 +63,31 @@ static const u32 mlx5e_link_speed[MLX5E_LINK_MODES_NUMBER] = {
[MLX5E_50GBASE_KR2] = 50000,
};
-int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port,
+static const u32 mlx5e_ext_link_speed[MLX5E_EXT_LINK_MODES_NUMBER] = {
+ [MLX5E_SGMII_100M] = 100,
+ [MLX5E_1000BASE_X_SGMII] = 1000,
+ [MLX5E_5GBASE_R] = 5000,
+ [MLX5E_10GBASE_XFI_XAUI_1] = 10000,
+ [MLX5E_40GBASE_XLAUI_4_XLPPI_4] = 40000,
+ [MLX5E_25GAUI_1_25GBASE_CR_KR] = 25000,
+ [MLX5E_50GAUI_2_LAUI_2_50GBASE_CR2_KR2] = 50000,
+ [MLX5E_50GAUI_1_LAUI_1_50GBASE_CR_KR] = 50000,
+ [MLX5E_CAUI_4_100GBASE_CR4_KR4] = 100000,
+ [MLX5E_200GAUI_4_200GBASE_CR4_KR4] = 200000,
+ [MLX5E_400GAUI_8] = 400000,
+};
+
+static void mlx5e_port_get_speed_arr(struct mlx5_core_dev *mdev,
+ const u32 **arr, u32 *size)
+{
+ bool ext = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet);
+
+ *size = ext ? ARRAY_SIZE(mlx5e_ext_link_speed) :
+ ARRAY_SIZE(mlx5e_link_speed);
+ *arr = ext ? mlx5e_ext_link_speed : mlx5e_link_speed;
+}
+
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
struct mlx5e_port_eth_proto *eproto)
{
u32 out[MLX5_ST_SZ_DW(ptys_reg)];
@@ -72,13 +96,17 @@ int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port,
if (!eproto)
return -EINVAL;
+ if (ext != MLX5_CAP_PCAM_FEATURE(dev, ptys_extended_ethernet))
+ return -EOPNOTSUPP;
+
err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, port);
if (err)
return err;
- eproto->cap = MLX5_GET(ptys_reg, out, eth_proto_capability);
- eproto->admin = MLX5_GET(ptys_reg, out, eth_proto_admin);
- eproto->oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
+ eproto->cap = MLX5_GET_ETH_PROTO(ptys_reg, out, ext,
+ eth_proto_capability);
+ eproto->admin = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_admin);
+ eproto->oper = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper);
return 0;
}
@@ -100,7 +128,7 @@ void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
}
int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
- u32 proto_admin)
+ u32 proto_admin, bool ext)
{
u32 out[MLX5_ST_SZ_DW(ptys_reg)];
u32 in[MLX5_ST_SZ_DW(ptys_reg)];
@@ -118,38 +146,46 @@ int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
MLX5_SET(ptys_reg, in, local_port, 1);
MLX5_SET(ptys_reg, in, an_disable_admin, an_disable);
MLX5_SET(ptys_reg, in, proto_mask, MLX5_PTYS_EN);
- MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin);
+ if (ext)
+ MLX5_SET(ptys_reg, in, ext_eth_proto_admin, proto_admin);
+ else
+ MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin);
return mlx5_core_access_reg(dev, in, sizeof(in), out,
sizeof(out), MLX5_REG_PTYS, 0, 1);
}
-u32 mlx5e_port_ptys2speed(u32 eth_proto_oper)
+u32 mlx5e_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper)
{
unsigned long temp = eth_proto_oper;
+ const u32 *table;
u32 speed = 0;
+ u32 max_size;
int i;
- i = find_first_bit(&temp, MLX5E_LINK_MODES_NUMBER);
- if (i < MLX5E_LINK_MODES_NUMBER)
- speed = mlx5e_link_speed[i];
-
+ mlx5e_port_get_speed_arr(mdev, &table, &max_size);
+ i = find_first_bit(&temp, max_size);
+ if (i < max_size)
+ speed = table[i];
return speed;
}
int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
{
struct mlx5e_port_eth_proto eproto;
+ bool ext;
int err;
- err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
+ ext = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet);
+ err = mlx5_port_query_eth_proto(mdev, 1, ext, &eproto);
if (err)
- return err;
+ goto out;
- *speed = mlx5e_port_ptys2speed(eproto.oper);
+ *speed = mlx5e_port_ptys2speed(mdev, eproto.oper);
if (!(*speed))
err = -EINVAL;
+out:
return err;
}
@@ -157,31 +193,38 @@ int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed)
{
struct mlx5e_port_eth_proto eproto;
u32 max_speed = 0;
+ const u32 *table;
+ u32 max_size;
+ bool ext;
int err;
int i;
- err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
+ ext = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet);
+ err = mlx5_port_query_eth_proto(mdev, 1, ext, &eproto);
if (err)
return err;
- for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i)
+ mlx5e_port_get_speed_arr(mdev, &table, &max_size);
+ for (i = 0; i < max_size; ++i)
if (eproto.cap & MLX5E_PROT_MASK(i))
- max_speed = max(max_speed, mlx5e_link_speed[i]);
+ max_speed = max(max_speed, table[i]);
*speed = max_speed;
return 0;
}
-u32 mlx5e_port_speed2linkmodes(u32 speed)
+u32 mlx5e_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed)
{
u32 link_modes = 0;
+ const u32 *table;
+ u32 max_size;
int i;
- for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) {
- if (mlx5e_link_speed[i] == speed)
+ mlx5e_port_get_speed_arr(mdev, &table, &max_size);
+ for (i = 0; i < max_size; ++i) {
+ if (table[i] == speed)
link_modes |= MLX5E_PROT_MASK(i);
}
-
return link_modes;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
index 4bdab8be10af..70f536ec51c4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h
@@ -42,16 +42,16 @@ struct mlx5e_port_eth_proto {
u32 oper;
};
-int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port,
+int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext,
struct mlx5e_port_eth_proto *eproto);
void mlx5_port_query_eth_autoneg(struct mlx5_core_dev *dev, u8 *an_status,
u8 *an_disable_cap, u8 *an_disable_admin);
int mlx5_port_set_eth_ptys(struct mlx5_core_dev *dev, bool an_disable,
- u32 proto_admin);
-u32 mlx5e_port_ptys2speed(u32 eth_proto_oper);
+ u32 proto_admin, bool ext);
+u32 mlx5e_port_ptys2speed(struct mlx5_core_dev *mdev, u32 eth_proto_oper);
int mlx5e_port_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
int mlx5e_port_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed);
-u32 mlx5e_port_speed2linkmodes(u32 speed);
+u32 mlx5e_port_speed2linkmodes(struct mlx5_core_dev *mdev, u32 speed);
int mlx5e_port_query_pbmc(struct mlx5_core_dev *mdev, void *out);
int mlx5e_port_set_pbmc(struct mlx5_core_dev *mdev, void *in);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index c29e141d72fb..8343cf7d292c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -695,13 +695,14 @@ static void get_speed_duplex(struct net_device *netdev,
u32 eth_proto_oper,
struct ethtool_link_ksettings *link_ksettings)
{
+ struct mlx5e_priv *priv = netdev_priv(netdev);
u32 speed = SPEED_UNKNOWN;
u8 duplex = DUPLEX_UNKNOWN;
if (!netif_carrier_ok(netdev))
goto out;
- speed = mlx5e_port_ptys2speed(eth_proto_oper);
+ speed = mlx5e_port_ptys2speed(priv->mdev, eth_proto_oper);
if (!speed) {
speed = SPEED_UNKNOWN;
goto out;
@@ -896,9 +897,9 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
link_modes = link_ksettings->base.autoneg == AUTONEG_ENABLE ?
mlx5e_ethtool2ptys_adver_link(link_ksettings->link_modes.advertising) :
- mlx5e_port_speed2linkmodes(speed);
+ mlx5e_port_speed2linkmodes(mdev, speed);
- err = mlx5_port_query_eth_proto(mdev, 1, &eproto);
+ err = mlx5_port_query_eth_proto(mdev, 1, false, &eproto);
if (err) {
netdev_err(priv->netdev, "%s: query port eth proto failed: %d\n",
__func__, err);
@@ -923,7 +924,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv,
if (!an_changes && link_modes == eproto.admin)
goto out;
- mlx5_port_set_eth_ptys(mdev, an_disable, link_modes);
+ mlx5_port_set_eth_ptys(mdev, an_disable, link_modes, false);
mlx5_toggle_port_link(mdev);
out:
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 5be7eefa6d75..814fa194663b 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -92,6 +92,22 @@ enum mlx5e_link_mode {
MLX5E_LINK_MODES_NUMBER,
};
+enum mlx5e_ext_link_mode {
+ MLX5E_SGMII_100M = 0,
+ MLX5E_1000BASE_X_SGMII = 1,
+ MLX5E_5GBASE_R = 3,
+ MLX5E_10GBASE_XFI_XAUI_1 = 4,
+ MLX5E_40GBASE_XLAUI_4_XLPPI_4 = 5,
+ MLX5E_25GAUI_1_25GBASE_CR_KR = 6,
+ MLX5E_50GAUI_2_LAUI_2_50GBASE_CR2_KR2 = 7,
+ MLX5E_50GAUI_1_LAUI_1_50GBASE_CR_KR = 8,
+ MLX5E_CAUI_4_100GBASE_CR4_KR4 = 9,
+ MLX5E_100GAUI_2_100GBASE_CR2_KR2 = 10,
+ MLX5E_200GAUI_4_200GBASE_CR4_KR4 = 12,
+ MLX5E_400GAUI_8 = 15,
+ MLX5E_EXT_LINK_MODES_NUMBER,
+};
+
enum mlx5e_connector_type {
MLX5E_PORT_UNKNOWN = 0,
MLX5E_PORT_NONE = 1,
@@ -106,6 +122,9 @@ enum mlx5e_connector_type {
};
#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
+#define MLX5_GET_ETH_PROTO(reg, out, ext, field) \
+ (ext ? MLX5_GET(reg, out, ext_##field) : \
+ MLX5_GET(reg, out, field))
int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
--
2.20.1
^ permalink raw reply related
* [PATCH mlx5-next 15/15] IB/mlx5: Add support for 50Gbps per lane link modes
From: Saeed Mahameed @ 2019-02-13 6:55 UTC (permalink / raw)
To: Saeed Mahameed, Leon Romanovsky; +Cc: netdev, linux-rdma, Aya Levin
In-Reply-To: <20190213065546.3013-1-saeedm@mellanox.com>
From: Aya Levin <ayal@mellanox.com>
Driver now supports new link modes: 50Gbps per lane support for
50G/100G/200G. This patch reads the correct field (legacy vs. extended)
based on a FW indication bit, and adds a translation function (link
modes to IB width and speed) to the new link modes.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/infiniband/hw/mlx5/main.c | 66 ++++++++++++++++++++++++++++---
1 file changed, 61 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 3677c00fa3bb..581ae11e2fc9 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -331,8 +331,8 @@ void mlx5_ib_put_native_port_mdev(struct mlx5_ib_dev *ibdev, u8 port_num)
spin_unlock(&port->mp.mpi_lock);
}
-static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed,
- u8 *active_width)
+static int translate_eth_legacy_proto_oper(u32 eth_proto_oper, u8 *active_speed,
+ u8 *active_width)
{
switch (eth_proto_oper) {
case MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII):
@@ -389,6 +389,61 @@ static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed,
return 0;
}
+static int translate_eth_ext_proto_oper(u32 eth_proto_oper, u8 *active_speed,
+ u8 *active_width)
+{
+ switch (eth_proto_oper) {
+ case MLX5E_PROT_MASK(MLX5E_SGMII_100M):
+ case MLX5E_PROT_MASK(MLX5E_1000BASE_X_SGMII):
+ *active_width = IB_WIDTH_1X;
+ *active_speed = IB_SPEED_SDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_5GBASE_R):
+ *active_width = IB_WIDTH_1X;
+ *active_speed = IB_SPEED_DDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_10GBASE_XFI_XAUI_1):
+ *active_width = IB_WIDTH_1X;
+ *active_speed = IB_SPEED_QDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_40GBASE_XLAUI_4_XLPPI_4):
+ *active_width = IB_WIDTH_4X;
+ *active_speed = IB_SPEED_QDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_25GAUI_1_25GBASE_CR_KR):
+ *active_width = IB_WIDTH_1X;
+ *active_speed = IB_SPEED_EDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_50GAUI_2_LAUI_2_50GBASE_CR2_KR2):
+ case MLX5E_PROT_MASK(MLX5E_50GAUI_1_LAUI_1_50GBASE_CR_KR):
+ *active_width = IB_WIDTH_1X;
+ *active_speed = IB_SPEED_HDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_100GAUI_2_100GBASE_CR2_KR2):
+ *active_width = IB_WIDTH_2X;
+ *active_speed = IB_SPEED_HDR;
+ break;
+ case MLX5E_PROT_MASK(MLX5E_200GAUI_4_200GBASE_CR4_KR4):
+ *active_width = IB_WIDTH_4X;
+ *active_speed = IB_SPEED_HDR;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed,
+ u8 *active_width, bool ext)
+{
+ return ext ?
+ translate_eth_ext_proto_oper(eth_proto_oper, active_speed,
+ active_width) :
+ translate_eth_legacy_proto_oper(eth_proto_oper, active_speed,
+ active_width);
+}
+
static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
struct ib_port_attr *props)
{
@@ -401,6 +456,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
u16 qkey_viol_cntr;
u32 eth_prot_oper;
u8 mdev_port_num;
+ bool ext;
int err;
mdev = mlx5_ib_get_native_port_mdev(dev, port_num, &mdev_port_num);
@@ -421,14 +477,14 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
mdev_port_num);
if (err)
goto out;
- eth_prot_oper = MLX5_GET_ETH_PROTO(ptys_reg, out, false,
- eth_proto_oper);
+ ext = MLX5_CAP_PCAM_FEATURE(dev->mdev, ptys_extended_ethernet);
+ eth_prot_oper = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper);
props->active_width = IB_WIDTH_4X;
props->active_speed = IB_SPEED_QDR;
translate_eth_proto_oper(eth_prot_oper, &props->active_speed,
- &props->active_width);
+ &props->active_width, ext);
props->port_cap_flags |= IB_PORT_CM_SUP;
props->ip_gids = true;
--
2.20.1
^ permalink raw reply related
* Re: [PATCH] net: phy: at803x: disable delay only for RGMII mode
From: Peter Ujfalusi @ 2019-02-13 7:02 UTC (permalink / raw)
To: Vinod Koul, David S Miller
Cc: linux-arm-msm, Bjorn Andersson, netdev, Niklas Cassel,
Andrew Lunn, Florian Fainelli, Nori, Sekhar, Marc Gonzalez
In-Reply-To: <20190212141922.12849-1-vkoul@kernel.org>
Hi Vinod,
On 12/02/2019 16.19, Vinod Koul wrote:
> Per "Documentation/devicetree/bindings/net/ethernet.txt" RGMII mode
> should not have delay in PHY whereas RGMII_ID and RGMII_RXID/RGMII_TXID
> can have delay in phy.
>
> So disable the delay only for RGMII mode and disable for other modes
s/disable for other modes/enable for other modes
Works fine on am335x-evmsk:
Tested-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
and few comment
> Fixes: cd28d1d6e52e: ("net: phy: at803x: Disable phy delay for RGMII mode")
> Reported-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
> Signed-off-by: Vinod Koul <vkoul@kernel.org>
> ---
> drivers/net/phy/at803x.c | 33 ++++++++++++++++++++++++++-------
> 1 file changed, 26 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c
> index 8ff12938ab47..7b54b54e3316 100644
> --- a/drivers/net/phy/at803x.c
> +++ b/drivers/net/phy/at803x.c
> @@ -110,6 +110,18 @@ static int at803x_debug_reg_mask(struct phy_device *phydev, u16 reg,
> return phy_write(phydev, AT803X_DEBUG_DATA, val);
> }
>
> +static inline int at803x_enable_rx_delay(struct phy_device *phydev)
> +{
> + return at803x_debug_reg_mask(phydev, AT803X_DEBUG_REG_0, 0,
> + AT803X_DEBUG_RX_CLK_DLY_EN);
> +}
static inline int at803x_select_rx_delay(struct phy_device *phydev,
bool enable)
{
}
> +
> +static inline int at803x_enable_tx_delay(struct phy_device *phydev)
> +{
> + return at803x_debug_reg_mask(phydev, AT803X_DEBUG_REG_5, 0,
> + AT803X_DEBUG_TX_CLK_DLY_EN);
> +}
static inline int at803x_select_tx_delay(struct phy_device *phydev,
bool enable)
{
}
perhaps?
> +
> static inline int at803x_disable_rx_delay(struct phy_device *phydev)
> {
> return at803x_debug_reg_mask(phydev, AT803X_DEBUG_REG_0,
> @@ -255,18 +267,25 @@ static int at803x_config_init(struct phy_device *phydev)
> if (ret < 0)
> return ret;
>
> - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID ||
> - phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> - phydev->interface == PHY_INTERFACE_MODE_RGMII) {
> + if (phydev->interface == PHY_INTERFACE_MODE_RGMII) {
> ret = at803x_disable_rx_delay(phydev);
> if (ret < 0)
> return ret;
> + ret = at803x_disable_tx_delay(phydev);
> + if (ret < 0)
> + return ret;
is it a possibility to have PHY_INTERFACE_MODE_RGMII and other RGMII_ID
|| RGMII_TXID || RGMII_RXID set at the same time?
if not you can just return from here, no need to check for other RGMII
modes?
> + };
> +
> + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> + phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID) {
> + ret = at803x_enable_rx_delay(phydev);
> + if (ret < 0)
> + return ret;
> }
>
> - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID ||
> - phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> - phydev->interface == PHY_INTERFACE_MODE_RGMII) {
> - ret = at803x_disable_tx_delay(phydev);
> + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> + phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) {
> + ret = at803x_enable_tx_delay(phydev);
> if (ret < 0)
> return ret;
> }
>
- Péter
Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki
^ permalink raw reply
* Re: [PATCH] rpc: properly check debugfs dentry before using it
From: gregkh @ 2019-02-13 7:24 UTC (permalink / raw)
To: Schumaker, Anna
Cc: bfields@fieldses.org, trond.myklebust@hammerspace.com,
jlayton@kernel.org, netdev@vger.kernel.org,
linux-nfs@vger.kernel.org, dhowells@redhat.com
In-Reply-To: <2a029733e094460e2f57bb2d3fc28e7e1757b83e.camel@netapp.com>
On Tue, Feb 12, 2019 at 08:52:53PM +0000, Schumaker, Anna wrote:
> On Tue, 2019-02-12 at 19:27 +0100, Greg Kroah-Hartman wrote:
> > debugfs can now report an error code if something went wrong instead of
> > just NULL. So if the return value is to be used as a "real" dentry, it
> > needs to be checked if it is an error before dereferencing it.
> >
> > This is now happening because of ff9fb72bc077 ("debugfs: return error
> > values, not NULL"), but why debugfs files are not being created properly
> > is an older issue, probably one that has always been there and should
> > probably be looked at...
> >
> > Cc: "J. Bruce Fields" <bfields@fieldses.org>
> > Cc: Jeff Layton <jlayton@kernel.org>
> > Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
> > Cc: Anna Schumaker <anna.schumaker@netapp.com>
> > Cc: linux-nfs@vger.kernel.org
> > Cc: netdev@vger.kernel.org
> > Reported-by: David Howells <dhowells@redhat.com>
> > Tested-by: David Howells <dhowells@redhat.com>
> > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> >
> > ---
> > net/sunrpc/debugfs.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > I can take this through my tree if people don't object, or it can go
> > through the NFS tree. It does need to get merged before 5.0-final
> > though.
>
> I'm planning another bugfixes pull for 5.0, so I can take this patch and send it
> with the others this week.
Wonderful, thanks for doing this, and sorry for the regression.
greg k-h
^ permalink raw reply
* Re: linux-next: Tree for Feb 11 (wireless/80211)
From: Johannes Berg @ 2019-02-13 8:04 UTC (permalink / raw)
To: Randy Dunlap, Stephen Rothwell, Linux Next Mailing List
Cc: Linux Kernel Mailing List, linux-wireless, netdev@vger.kernel.org
In-Reply-To: <80b32344-08c9-0d8b-fdae-1db9b1003f60@infradead.org>
Thanks for the heads-up, Randy. We'd seen separate build bot reports on
this as well.
> ERROR: "__umoddi3" [net/wireless/cfg80211.ko] undefined!
> ERROR: "__umoddi3" [net/mac80211/mac80211.ko] undefined!
Should be fixed as of yesterday, let me know if it reappears.
johannes
^ permalink raw reply
* [PATCH net] net: dsa: bcm_sf2: potential array overflow in bcm_sf2_sw_suspend()
From: Dan Carpenter @ 2019-02-13 8:23 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli
Cc: Vivien Didelot, David S. Miller, netdev, kernel-janitors
The value of ->num_ports comes from bcm_sf2_sw_probe() and it is less
than or equal to DSA_MAX_PORTS. The ds->ports[] array is used inside
the dsa_is_user_port() and dsa_is_cpu_port() functions. The ds->ports[]
array is allocated in dsa_switch_alloc() and it has ds->num_ports
elements so this leads to a static checker warning about a potential out
of bounds read.
Fixes: 8cfa94984c9c ("net: dsa: bcm_sf2: add suspend/resume callbacks")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
---
drivers/net/dsa/bcm_sf2.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 5193da67dcdc..98696a88fa1c 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -690,7 +690,7 @@ static int bcm_sf2_sw_suspend(struct dsa_switch *ds)
* port, the other ones have already been disabled during
* bcm_sf2_sw_setup
*/
- for (port = 0; port < DSA_MAX_PORTS; port++) {
+ for (port = 0; port < ds->num_ports; port++) {
if (dsa_is_user_port(ds, port) || dsa_is_cpu_port(ds, port))
bcm_sf2_port_disable(ds, port, NULL);
}
--
2.17.1
^ permalink raw reply related
* Re: KASAN: use-after-free Read in sctp_outq_tail
From: Dmitry Vyukov @ 2019-02-13 8:27 UTC (permalink / raw)
To: Marcelo Ricardo Leitner
Cc: syzbot, David Miller, LKML, linux-sctp, netdev, Neil Horman,
syzkaller-bugs, Vladislav Yasevich
In-Reply-To: <20190212191920.GI10665@localhost.localdomain>
On Tue, Feb 12, 2019 at 8:19 PM Marcelo Ricardo Leitner
<marcelo.leitner@gmail.com> wrote:
>
> On Tue, Feb 12, 2019 at 11:04:27AM -0800, syzbot wrote:
> > Hello,
> >
> > syzbot found the following crash on:
> >
> > HEAD commit: d4104460aec1 Add linux-next specific files for 20190211
> > git tree: linux-next
>
> I can't find this commit. How can I know for sure which commits are in?
> Recent patches are involved with this code, but not sure what was
> included on the test.
Hi Marcelo,
linux-next is weird. Not sure why a so recent commit is not in
linux-next, maybe one needs to fetch tags or something. Fetching
git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next-history.git
helped me to get this commit:
$ git show d4104460aec1
commit d4104460aec152e23cf80ab1f950f414bf94f4ea (HEAD, tag: next-20190211)
Author: Stephen Rothwell
Date: Mon Feb 11 18:37:26 2019 +1100
Add linux-next specific files for 20190211
^ permalink raw reply
* Re: KASAN: use-after-free Read in sctp_outq_tail
From: Dmitry Vyukov @ 2019-02-13 8:29 UTC (permalink / raw)
To: Xin Long
Cc: syzbot, davem, LKML, linux-sctp, Marcelo Ricardo Leitner,
network dev, Neil Horman, syzkaller-bugs, Vlad Yasevich
In-Reply-To: <CADvbK_dndFMxyUZk=k8qEy9=PNbLc81F++2L0fbdtUdbupq4Xg@mail.gmail.com>
On Wed, Feb 13, 2019 at 5:36 AM Xin Long <lucien.xin@gmail.com> wrote:
>
> On Wed, Feb 13, 2019 at 4:00 AM syzbot
> <syzbot+7823fa3f3e2d69341ea8@syzkaller.appspotmail.com> wrote:
> >
> > Hello,
> >
> > syzbot found the following crash on:
> >
> > HEAD commit: d4104460aec1 Add linux-next specific files for 20190211
> > git tree: linux-next
> > console output: https://syzkaller.appspot.com/x/log.txt?x=14140124c00000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=c8a112d3b0d6719b
> > dashboard link: https://syzkaller.appspot.com/bug?extid=7823fa3f3e2d69341ea8
> > compiler: gcc (GCC) 9.0.0 20181231 (experimental)
> >
> > Unfortunately, I don't have any reproducer for this crash yet.
> >
> > IMPORTANT: if you fix the bug, please add the following tag to the commit:
> > Reported-by: syzbot+7823fa3f3e2d69341ea8@syzkaller.appspotmail.com
> >
> > ==================================================================
> > BUG: KASAN: use-after-free in list_add_tail include/linux/list.h:93 [inline]
> > BUG: KASAN: use-after-free in sctp_outq_tail_data net/sctp/outqueue.c:105
> > [inline]
> > BUG: KASAN: use-after-free in sctp_outq_tail+0x816/0x930
> > net/sctp/outqueue.c:313
> > Read of size 8 at addr ffff88807b19a7b8 by task syz-executor.0/30745
> I think https://patchwork.ozlabs.org/patch/1040500/ will fix this.
Let's mark it then to not return later:
#syz fix:
sctp: set stream ext to NULL after freeing it in sctp_stream_outq_migrate
> > CPU: 1 PID: 30745 Comm: syz-executor.0 Not tainted 5.0.0-rc5-next-20190211
> > #32
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
> > Google 01/01/2011
> > Call Trace:
> > __dump_stack lib/dump_stack.c:77 [inline]
> > dump_stack+0x172/0x1f0 lib/dump_stack.c:113
> > print_address_description.cold+0x7c/0x20d mm/kasan/report.c:187
> > kasan_report.cold+0x1b/0x40 mm/kasan/report.c:317
> > __asan_report_load8_noabort+0x14/0x20 mm/kasan/generic_report.c:132
> > list_add_tail include/linux/list.h:93 [inline]
> > sctp_outq_tail_data net/sctp/outqueue.c:105 [inline]
> > sctp_outq_tail+0x816/0x930 net/sctp/outqueue.c:313
> > sctp_cmd_send_msg net/sctp/sm_sideeffect.c:1109 [inline]
> > sctp_cmd_interpreter net/sctp/sm_sideeffect.c:1784 [inline]
> > sctp_side_effects net/sctp/sm_sideeffect.c:1220 [inline]
> > sctp_do_sm+0x68e/0x5380 net/sctp/sm_sideeffect.c:1191
> > sctp_primitive_SEND+0xa0/0xd0 net/sctp/primitive.c:178
> > sctp_sendmsg_to_asoc+0xa63/0x17b0 net/sctp/socket.c:1955
> > sctp_sendmsg+0x10a9/0x17e0 net/sctp/socket.c:2113
> > inet_sendmsg+0x147/0x5d0 net/ipv4/af_inet.c:798
> > sock_sendmsg_nosec net/socket.c:621 [inline]
> > sock_sendmsg+0xdd/0x130 net/socket.c:631
> > ___sys_sendmsg+0x806/0x930 net/socket.c:2136
> > __sys_sendmsg+0x105/0x1d0 net/socket.c:2174
> > __do_sys_sendmsg net/socket.c:2183 [inline]
> > __se_sys_sendmsg net/socket.c:2181 [inline]
> > __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2181
> > do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
> > entry_SYSCALL_64_after_hwframe+0x49/0xbe
> > RIP: 0033:0x457e39
> > Code: ad b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7
> > 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff
> > ff 0f 83 7b b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00
> > RSP: 002b:00007fa9b8630c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
> > RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000457e39
> > RDX: 0000000000000000 RSI: 0000000020000140 RDI: 0000000000000003
> > RBP: 000000000073bf00 R08: 0000000000000000 R09: 0000000000000000
> > R10: 0000000000000000 R11: 0000000000000246 R12: 00007fa9b86316d4
> > R13: 00000000004c4e2b R14: 00000000004d8ab8 R15: 00000000ffffffff
> >
> > Allocated by task 30745:
> > save_stack+0x45/0xd0 mm/kasan/common.c:75
> > set_track mm/kasan/common.c:87 [inline]
> > __kasan_kmalloc mm/kasan/common.c:498 [inline]
> > __kasan_kmalloc.constprop.0+0xcf/0xe0 mm/kasan/common.c:471
> > kasan_kmalloc+0x9/0x10 mm/kasan/common.c:506
> > kmem_cache_alloc_trace+0x151/0x760 mm/slab.c:3615
> > kmalloc include/linux/slab.h:548 [inline]
> > kzalloc include/linux/slab.h:743 [inline]
> > sctp_stream_init_ext+0x51/0x110 net/sctp/stream.c:172
> > sctp_sendmsg_to_asoc+0x1273/0x17b0 net/sctp/socket.c:1896
> > sctp_sendmsg+0x10a9/0x17e0 net/sctp/socket.c:2113
> > inet_sendmsg+0x147/0x5d0 net/ipv4/af_inet.c:798
> > sock_sendmsg_nosec net/socket.c:621 [inline]
> > sock_sendmsg+0xdd/0x130 net/socket.c:631
> > ___sys_sendmsg+0x806/0x930 net/socket.c:2136
> > __sys_sendmsg+0x105/0x1d0 net/socket.c:2174
> > __do_sys_sendmsg net/socket.c:2183 [inline]
> > __se_sys_sendmsg net/socket.c:2181 [inline]
> > __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2181
> > do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
> > entry_SYSCALL_64_after_hwframe+0x49/0xbe
> >
> > Freed by task 30745:
> > save_stack+0x45/0xd0 mm/kasan/common.c:75
> > set_track mm/kasan/common.c:87 [inline]
> > __kasan_slab_free+0x102/0x150 mm/kasan/common.c:460
> > kasan_slab_free+0xe/0x10 mm/kasan/common.c:468
> > __cache_free mm/slab.c:3491 [inline]
> > kfree+0xcf/0x230 mm/slab.c:3816
> > sctp_stream_outq_migrate+0x3e6/0x540 net/sctp/stream.c:88
> > sctp_stream_init+0xbc/0x410 net/sctp/stream.c:139
> > sctp_process_init+0x21c3/0x2b20 net/sctp/sm_make_chunk.c:2466
> > sctp_cmd_process_init net/sctp/sm_sideeffect.c:682 [inline]
> > sctp_cmd_interpreter net/sctp/sm_sideeffect.c:1410 [inline]
> > sctp_side_effects net/sctp/sm_sideeffect.c:1220 [inline]
> > sctp_do_sm+0x3145/0x5380 net/sctp/sm_sideeffect.c:1191
> > sctp_assoc_bh_rcv+0x343/0x660 net/sctp/associola.c:1074
> > sctp_inq_push+0x1ea/0x290 net/sctp/inqueue.c:95
> > sctp_backlog_rcv+0x196/0xbe0 net/sctp/input.c:354
> > sk_backlog_rcv include/net/sock.h:937 [inline]
> > __release_sock+0x12e/0x3a0 net/core/sock.c:2379
> > release_sock+0x59/0x1c0 net/core/sock.c:2895
> > sctp_wait_for_connect+0x316/0x540 net/sctp/socket.c:8998
> > sctp_sendmsg_to_asoc+0x13e3/0x17b0 net/sctp/socket.c:1967
> > sctp_sendmsg+0x10a9/0x17e0 net/sctp/socket.c:2113
> > inet_sendmsg+0x147/0x5d0 net/ipv4/af_inet.c:798
> > sock_sendmsg_nosec net/socket.c:621 [inline]
> > sock_sendmsg+0xdd/0x130 net/socket.c:631
> > ___sys_sendmsg+0x806/0x930 net/socket.c:2136
> > __sys_sendmsg+0x105/0x1d0 net/socket.c:2174
> > __do_sys_sendmsg net/socket.c:2183 [inline]
> > __se_sys_sendmsg net/socket.c:2181 [inline]
> > __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2181
> > do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
> > entry_SYSCALL_64_after_hwframe+0x49/0xbe
> >
> > The buggy address belongs to the object at ffff88807b19a780
> > which belongs to the cache kmalloc-96 of size 96
> > The buggy address is located 56 bytes inside of
> > 96-byte region [ffff88807b19a780, ffff88807b19a7e0)
> > The buggy address belongs to the page:
> > page:ffffea0001ec6680 count:1 mapcount:0 mapping:ffff88812c3f04c0
> > index:0xffff88807b19a800
> > flags: 0x1fffc0000000200(slab)
> > raw: 01fffc0000000200 ffffea000262acc8 ffffea0001448348 ffff88812c3f04c0
> > raw: ffff88807b19a800 ffff88807b19a000 000000010000001d 0000000000000000
> > page dumped because: kasan: bad access detected
> >
> > Memory state around the buggy address:
> > ffff88807b19a680: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
> > ffff88807b19a700: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
> > > ffff88807b19a780: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
> > ^
> > ffff88807b19a800: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
> > ffff88807b19a880: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
> > ==================================================================
> >
> >
> > ---
> > This bug is generated by a bot. It may contain errors.
> > See https://goo.gl/tpsmEJ for more information about syzbot.
> > syzbot engineers can be reached at syzkaller@googlegroups.com.
> >
> > syzbot will keep track of this bug report. See:
> > https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with
> > syzbot.
>
> --
> You received this message because you are subscribed to the Google Groups "syzkaller-bugs" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to syzkaller-bugs+unsubscribe@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/syzkaller-bugs/CADvbK_dndFMxyUZk%3Dk8qEy9%3DPNbLc81F%2B%2B2L0fbdtUdbupq4Xg%40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.
^ permalink raw reply
* Re: [PATCH iproute2 net-next v2 3/4] ss: Buffer raw fields first, then render them as a table
From: Stefano Brivio @ 2019-02-13 8:37 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Stephen Hemminger, netdev, Sabrina Dubroca
In-Reply-To: <82f1bc98-df6d-2b0a-17e5-fa057563284e@gmail.com>
On Tue, 12 Feb 2019 16:42:04 -0800
Eric Dumazet <eric.dumazet@gmail.com> wrote:
> I do not get it.
>
> "ss -emoi " uses almost 1KB per socket.
>
> 10,000,000 sockets -> we need about 10GB of memory ???
>
> This is a serious regression.
I guess this is rather subjective: the worst case I considered back then
was the output of 'ss -tei0' (less than 500 bytes) for one million
sockets, which gives 500M of memory, which should in turn be fine on a
machine handling one million sockets.
Now, if 'ss -emoi' on 10 million sockets is an actual use case (out of
curiosity: how are you going to process that output? Would JSON help?),
I see two easy options to solve this:
1. flush the output every time we reach a given buffer size (1M
perhaps). This might make the resulting blocks slightly unaligned,
with occasional loss of readability on lines occurring every 1k to
10k sockets approximately, even though after 1k sockets column sizes
won't change much (it looks anyway better than the original), and I
don't expect anybody to actually scroll that output
2. add a switch for unbuffered output, but then you need to remember to
pass it manually, and the whole output would be as bad as the
original in case you need the switch.
I'd rather go with 1., it's easy to implement (we already have partial
flushing with '--events') and it looks like a good compromise on
usability. Thoughts?
--
Stefano
^ 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