* Re: [PATCH net 0/3] Few mvneta fixes
From: Arnd Bergmann @ 2017-12-19 20:18 UTC (permalink / raw)
To: Gregory CLEMENT
Cc: David S. Miller, Linux Kernel Mailing List, Networking,
Thomas Petazzoni, Andrew Lunn, Jason Cooper, Antoine Tenart,
Nadav Haklai, Lior Amsalem, Miquèl Raynal, Marcin Wojtas,
Dmitri Epshtein, Linux ARM, Sebastian Hesselbarth
In-Reply-To: <20171219165947.28516-1-gregory.clement@free-electrons.com>
On Tue, Dec 19, 2017 at 5:59 PM, Gregory CLEMENT
<gregory.clement@free-electrons.com> wrote:
> Hello,
>
> here it is a small series of fixes found on the mvneta driver. They
> had been already used in the vendor kernel and are now ported to
> mainline.
Does one of the patches look like it addresses the rare Oops we discussed on
#kernelci this morning?
https://storage.kernelci.org/stable/linux-4.9.y/v4.9.70/arm/mvebu_v7_defconfig/lab-free-electrons/boot-armada-375-db.html
Arnd
^ permalink raw reply
* Re: [PATCH net-next 4/4] sfc: expose CTPIO stats on NICs that support them
From: Jakub Kicinski @ 2017-12-19 20:20 UTC (permalink / raw)
To: Edward Cree; +Cc: linux-net-drivers, davem, netdev
In-Reply-To: <d0b118c3-380b-d7f4-de32-60841852809d@solarflare.com>
On Tue, 19 Dec 2017 17:06:36 +0000, Edward Cree wrote:
> From: Bert Kenward <bkenward@solarflare.com>
>
> While the Linux driver doesn't use CTPIO ('cut-through programmed I/O'),
> other drivers on the same port might, so if we're responsible for
> reporting per-port stats we need to include the CTPIO stats.
>
> Signed-off-by: Edward Cree <ecree@solarflare.com>
FWIW this is missing a sign-off from Bert.
^ permalink raw reply
* [PATCH bpf]: Fix tools and testing build.
From: David Miller @ 2017-12-19 20:22 UTC (permalink / raw)
To: daniel; +Cc: netdev
I'm getting various build failures on sparc64. The key is
usually that the userland tools get built 32-bit.
1) clock_gettime() is in librt, so that must be added to the link
libraries.
2) "sizeof(x)" must be printed with "%Z" printf prefix.
Signed-off-by: David S. Miller <davem@davemloft.net>
---
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 792af7c3b74f..05fc4e2e7b3a 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -11,7 +11,7 @@ ifneq ($(wildcard $(GENHDR)),)
endif
CFLAGS += -Wall -O2 -I$(APIDIR) -I$(LIBDIR) -I$(GENDIR) $(GENFLAGS) -I../../../include
-LDLIBS += -lcap -lelf
+LDLIBS += -lcap -lelf -lrt
TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
test_align test_verifier_log test_dev_cgroup
diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index 69427531408d..6761be18a91f 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -351,7 +351,7 @@ static void test_bpf_obj_id(void)
info_len != sizeof(struct bpf_map_info) ||
strcmp((char *)map_infos[i].name, expected_map_name),
"get-map-info(fd)",
- "err %d errno %d type %d(%d) info_len %u(%lu) key_size %u value_size %u max_entries %u map_flags %X name %s(%s)\n",
+ "err %d errno %d type %d(%d) info_len %u(%Zu) key_size %u value_size %u max_entries %u map_flags %X name %s(%s)\n",
err, errno,
map_infos[i].type, BPF_MAP_TYPE_ARRAY,
info_len, sizeof(struct bpf_map_info),
@@ -395,7 +395,7 @@ static void test_bpf_obj_id(void)
*(int *)prog_infos[i].map_ids != map_infos[i].id ||
strcmp((char *)prog_infos[i].name, expected_prog_name),
"get-prog-info(fd)",
- "err %d errno %d i %d type %d(%d) info_len %u(%lu) jit_enabled %d jited_prog_len %u xlated_prog_len %u jited_prog %d xlated_prog %d load_time %lu(%lu) uid %u(%u) nr_map_ids %u(%u) map_id %u(%u) name %s(%s)\n",
+ "err %d errno %d i %d type %d(%d) info_len %u(%Zu) jit_enabled %d jited_prog_len %u xlated_prog_len %u jited_prog %d xlated_prog %d load_time %lu(%lu) uid %u(%u) nr_map_ids %u(%u) map_id %u(%u) name %s(%s)\n",
err, errno, i,
prog_infos[i].type, BPF_PROG_TYPE_SOCKET_FILTER,
info_len, sizeof(struct bpf_prog_info),
@@ -463,7 +463,7 @@ static void test_bpf_obj_id(void)
memcmp(&prog_info, &prog_infos[i], info_len) ||
*(int *)prog_info.map_ids != saved_map_id,
"get-prog-info(next_id->fd)",
- "err %d errno %d info_len %u(%lu) memcmp %d map_id %u(%u)\n",
+ "err %d errno %d info_len %u(%Zu) memcmp %d map_id %u(%u)\n",
err, errno, info_len, sizeof(struct bpf_prog_info),
memcmp(&prog_info, &prog_infos[i], info_len),
*(int *)prog_info.map_ids, saved_map_id);
@@ -509,7 +509,7 @@ static void test_bpf_obj_id(void)
memcmp(&map_info, &map_infos[i], info_len) ||
array_value != array_magic_value,
"check get-map-info(next_id->fd)",
- "err %d errno %d info_len %u(%lu) memcmp %d array_value %llu(%llu)\n",
+ "err %d errno %d info_len %u(%Zu) memcmp %d array_value %llu(%llu)\n",
err, errno, info_len, sizeof(struct bpf_map_info),
memcmp(&map_info, &map_infos[i], info_len),
array_value, array_magic_value);
^ permalink raw reply related
* Re: [PATCH v5 0/6] enable creating [k,u]probe with perf_event_open
From: Peter Zijlstra @ 2017-12-19 20:25 UTC (permalink / raw)
To: Song Liu
Cc: Daniel Borkmann, Steven Rostedt, mingo@redhat.com, David Miller,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Kernel Team
In-Reply-To: <3D2EBCDD-9023-4969-940E-BC03230361E2@fb.com>
Took 1-4, Thanks!
^ permalink raw reply
* Re: [RFC] hv_netvsc: automatically name slave VF network device
From: Jakub Kicinski @ 2017-12-19 20:32 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev, Stephen Hemminger
In-Reply-To: <20171219193537.22587-1-sthemmin@microsoft.com>
On Tue, 19 Dec 2017 11:35:37 -0800, Stephen Hemminger wrote:
> Rename the VF device to ethX_vf based on the ethX as the
> synthetic device. This eliminates the need for delay on setup,
> and the PCI (udev based) naming is not reproducible on Hyper-V
> anyway. The name of the VF does not matter since all control
> operations take place the primary device. It does make the
> user experience better to associate the names.
>
> Based on feedback from all.systems.go talk.
> The downside is that it requires exporting a symbol from netdev
> core which makes it harder to backport.
>
> Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Why do you have to name the devices in the kernel space in the first
place? :/ Why don't upstream the correct change to biosdevname like
hardware vendors do?
Your VF setup is really _not_ special, I don't understand why we are
OK with ignoring the standard practices. Real enterprise distroes
are very careful never to break the naming of interfaces and they keep
the naming policy in user space. Playing tricks in the kernel has every
chance of breaking existing user setups.
^ permalink raw reply
* [for-next 01/11] net/mlx5: E-Switch, Refactor vport representors initialization
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-rdma-u79uwXL29TY76Z2rM5mHXA,
Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Refactor the init stage of vport representors registration.
vport number and hw id can be assigned by the E-Switch driver and not by
the netdevice driver. While here, make the error path of mlx5_eswitch_init()
a reverse order of the good path, also use kcalloc to allocate an array
instead of kzalloc.
Signed-off-by: Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Saeed Mahameed <saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 7 ----
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 12 +++----
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 2 ++
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 40 +++++++++++++++++++---
4 files changed, 42 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 2c43606c26b5..4661ef12c18c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1100,17 +1100,12 @@ static void mlx5e_rep_register_vf_vports(struct mlx5e_priv *priv)
struct mlx5_eswitch *esw = mdev->priv.eswitch;
int total_vfs = MLX5_TOTAL_VPORTS(mdev);
int vport;
- u8 mac[ETH_ALEN];
-
- mlx5_query_nic_vport_mac_address(mdev, 0, mac);
for (vport = 1; vport < total_vfs; vport++) {
struct mlx5_eswitch_rep rep;
rep.load = mlx5e_vport_rep_load;
rep.unload = mlx5e_vport_rep_unload;
- rep.vport = vport;
- ether_addr_copy(rep.hw_id, mac);
mlx5_eswitch_register_vport_rep(esw, vport, &rep);
}
}
@@ -1132,10 +1127,8 @@ void mlx5e_register_vport_reps(struct mlx5e_priv *priv)
struct mlx5_eswitch *esw = mdev->priv.eswitch;
struct mlx5_eswitch_rep rep;
- mlx5_query_nic_vport_mac_address(mdev, 0, rep.hw_id);
rep.load = mlx5e_nic_rep_load;
rep.unload = mlx5e_nic_rep_unload;
- rep.vport = FDB_UPLINK_VPORT;
rep.netdev = priv->netdev;
mlx5_eswitch_register_vport_rep(esw, 0, &rep); /* UPLINK PF vport*/
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index bbb140f517c4..6d4cbdb69823 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1644,13 +1644,9 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
goto abort;
}
- esw->offloads.vport_reps =
- kzalloc(total_vports * sizeof(struct mlx5_eswitch_rep),
- GFP_KERNEL);
- if (!esw->offloads.vport_reps) {
- err = -ENOMEM;
+ err = esw_offloads_init_reps(esw);
+ if (err)
goto abort;
- }
hash_init(esw->offloads.encap_tbl);
hash_init(esw->offloads.mod_hdr_tbl);
@@ -1681,8 +1677,8 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
abort:
if (esw->work_queue)
destroy_workqueue(esw->work_queue);
+ esw_offloads_cleanup_reps(esw);
kfree(esw->vports);
- kfree(esw->offloads.vport_reps);
kfree(esw);
return err;
}
@@ -1696,7 +1692,7 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw)
esw->dev->priv.eswitch = NULL;
destroy_workqueue(esw->work_queue);
- kfree(esw->offloads.vport_reps);
+ esw_offloads_cleanup_reps(esw);
kfree(esw->vports);
kfree(esw);
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 565c8b7a399a..9722c2a96090 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -197,6 +197,8 @@ struct mlx5_eswitch {
void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports);
int esw_offloads_init(struct mlx5_eswitch *esw, int nvports);
+void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw);
+int esw_offloads_init_reps(struct mlx5_eswitch *esw);
/* E-Switch API */
int mlx5_eswitch_init(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 1143d80119bd..230b05842cb0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -732,6 +732,42 @@ static int esw_offloads_start(struct mlx5_eswitch *esw)
return err;
}
+void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw)
+{
+ kfree(esw->offloads.vport_reps);
+}
+
+int esw_offloads_init_reps(struct mlx5_eswitch *esw)
+{
+ struct mlx5_core_dev *dev = esw->dev;
+ struct mlx5_esw_offload *offloads;
+ struct mlx5_eswitch_rep *rep;
+ int total_vfs = MLX5_TOTAL_VPORTS(dev);
+ u8 hw_id[ETH_ALEN];
+ int vport;
+
+ esw->offloads.vport_reps =
+ kcalloc(total_vfs, sizeof(struct mlx5_eswitch_rep),
+ GFP_KERNEL);
+
+ if (!esw->offloads.vport_reps)
+ return -ENOMEM;
+
+ offloads = &esw->offloads;
+ mlx5_query_nic_vport_mac_address(dev, 0, hw_id);
+
+ for (vport = 0; vport < total_vfs; vport++) {
+ rep = &offloads->vport_reps[vport];
+
+ rep->vport = vport;
+ ether_addr_copy(rep->hw_id, hw_id);
+ }
+
+ offloads->vport_reps[0].vport = FDB_UPLINK_VPORT;
+
+ return 0;
+}
+
int esw_offloads_init(struct mlx5_eswitch *esw, int nvports)
{
struct mlx5_eswitch_rep *rep;
@@ -1127,13 +1163,9 @@ void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw,
rep = &offloads->vport_reps[vport_index];
- memset(rep, 0, sizeof(*rep));
-
rep->load = __rep->load;
rep->unload = __rep->unload;
- rep->vport = __rep->vport;
rep->netdev = __rep->netdev;
- ether_addr_copy(rep->hw_id, __rep->hw_id);
INIT_LIST_HEAD(&rep->vport_sqs_list);
rep->valid = true;
--
2.14.2
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [for-next 02/11] net/mlx5: E-Switch, Refactor load/unload of representors
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-rdma-u79uwXL29TY76Z2rM5mHXA,
Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Refactor the load/unload stages for better code reuse.
Signed-off-by: Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Saeed Mahameed <saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 66 +++++++++++++---------
1 file changed, 40 insertions(+), 26 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 230b05842cb0..139da9f3cc70 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -768,12 +768,47 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
return 0;
}
-int esw_offloads_init(struct mlx5_eswitch *esw, int nvports)
+static void esw_offloads_unload_reps(struct mlx5_eswitch *esw, int nvports)
+{
+ struct mlx5_eswitch_rep *rep;
+ int vport;
+
+ for (vport = nvports - 1; vport >= 0; vport--) {
+ rep = &esw->offloads.vport_reps[vport];
+ if (!rep->valid)
+ continue;
+
+ rep->unload(esw, rep);
+ }
+}
+
+static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
{
struct mlx5_eswitch_rep *rep;
int vport;
int err;
+ for (vport = 0; vport < nvports; vport++) {
+ rep = &esw->offloads.vport_reps[vport];
+ if (!rep->valid)
+ continue;
+
+ err = rep->load(esw, rep);
+ if (err)
+ goto err_reps;
+ }
+
+ return 0;
+
+err_reps:
+ esw_offloads_unload_reps(esw, vport);
+ return err;
+}
+
+int esw_offloads_init(struct mlx5_eswitch *esw, int nvports)
+{
+ int err;
+
/* disable PF RoCE so missed packets don't go through RoCE steering */
mlx5_dev_list_lock();
mlx5_remove_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB);
@@ -791,25 +826,13 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports)
if (err)
goto create_fg_err;
- for (vport = 0; vport < nvports; vport++) {
- rep = &esw->offloads.vport_reps[vport];
- if (!rep->valid)
- continue;
-
- err = rep->load(esw, rep);
- if (err)
- goto err_reps;
- }
+ err = esw_offloads_load_reps(esw, nvports);
+ if (err)
+ goto err_reps;
return 0;
err_reps:
- for (vport--; vport >= 0; vport--) {
- rep = &esw->offloads.vport_reps[vport];
- if (!rep->valid)
- continue;
- rep->unload(esw, rep);
- }
esw_destroy_vport_rx_group(esw);
create_fg_err:
@@ -850,16 +873,7 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw)
void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports)
{
- struct mlx5_eswitch_rep *rep;
- int vport;
-
- for (vport = nvports - 1; vport >= 0; vport--) {
- rep = &esw->offloads.vport_reps[vport];
- if (!rep->valid)
- continue;
- rep->unload(esw, rep);
- }
-
+ esw_offloads_unload_reps(esw, nvports);
esw_destroy_vport_rx_group(esw);
esw_destroy_offloads_table(esw);
esw_destroy_offloads_fdb_tables(esw);
--
2.14.2
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [pull request][for-next 00/11] Mellanox, mlx5 E-Switch updates 2017-12-19
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Saeed Mahameed
Hi Dave and Doug,
==============
This series includes updates for mlx5 E-Switch infrastructures,
to be merged into net-next and rdma-next trees.
Mark's patches provide E-Switch refactoring that generalize the mlx5
E-Switch vf representors interfaces and data structures. The serious is
mainly focused on moving ethernet (netdev) specific representors logic out
of E-Switch (eswitch.c) into mlx5e representor module (en_rep.c), which
provides better separation and allows future support for other types of vf
representors (e.g. RDMA).
Gal's patches at the end of this serious, provide a simple syntax fix and
two other patches that handles vport ingress/egress ACL steering name
spaces to be aligned with the Firmware/Hardware specs.
===============
This patchset is based on rc4 and I see that net-next is still on rc3, i hope
this is not a problem.
Please pull and let me know if there's any problem.
Thanks,
Saeed.
---
The following changes since commit 1291a0d5049dbc06baaaf66a9ff3f53db493b19b:
Linux 4.15-rc4 (2017-12-17 18:59:59 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux.git tags/mlx5-shared-4.16-1
for you to fetch changes up to e39cdfc72337490405b9382843b2fbdd09d7a49e:
net/mlx5: Separate ingress/egress namespaces for each vport (2017-12-18 22:49:54 -0800)
----------------------------------------------------------------
mlx5-shared-4.16-1
mlx5 shared code for both rdma-next and net-next trees.
----------------------------------------------------------------
Gal Pressman (3):
net/mlx5e: E-Switch, Use the name of static array instead of its address
net/mlx5: Fix ingress/egress naming mistake
net/mlx5: Separate ingress/egress namespaces for each vport
Mark Bloch (8):
net/mlx5: E-Switch, Refactor vport representors initialization
net/mlx5: E-Switch, Refactor load/unload of representors
net/mlx5: E-Switch, Simplify representor load/unload callback API
net/mlx5: E-Switch, Move mlx5e only logic outside E-Switch
net/mlx5: E-Switch, Create a dedicated send to vport rule deletion function
net/mlx5e: Move ethernet representors data into separate struct
net/mlx5: E-Switch, Create generic header struct to be used by representors
net/mlx5e: E-Switch, Move send-to-vport rule struct to en_rep
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 147 ++++++++++----
drivers/net/ethernet/mellanox/mlx5/core/en_rep.h | 14 ++
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 15 +-
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 48 +++--
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 45 +++--
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 219 ++++++++++++---------
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 145 +++++++++++---
drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 4 +-
include/linux/mlx5/fs.h | 4 +
9 files changed, 427 insertions(+), 214 deletions(-)
^ permalink raw reply
* [for-next 04/11] net/mlx5: E-Switch, Move mlx5e only logic outside E-Switch
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
In our pursuit to cleanup e-switch sub-module from mlx5e specific code,
we move the functions that insert/remove the flow steering rules that
allow mlx5e representors to send packets directly to VFs into the EN
driver code.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 57 +++++++++++++++++++++-
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 9 ++--
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 55 +--------------------
3 files changed, 59 insertions(+), 62 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 6d2219f3acf6..19edaa155062 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -190,6 +190,59 @@ int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr)
return 0;
}
+static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw,
+ struct mlx5_eswitch_rep *rep)
+{
+ struct mlx5_esw_sq *esw_sq, *tmp;
+
+ if (esw->mode != SRIOV_OFFLOADS)
+ return;
+
+ list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) {
+ mlx5_del_flow_rules(esw_sq->send_to_vport_rule);
+ list_del(&esw_sq->list);
+ kfree(esw_sq);
+ }
+}
+
+static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
+ struct mlx5_eswitch_rep *rep,
+ u16 *sqns_array, int sqns_num)
+{
+ struct mlx5_flow_handle *flow_rule;
+ struct mlx5_esw_sq *esw_sq;
+ int err;
+ int i;
+
+ if (esw->mode != SRIOV_OFFLOADS)
+ return 0;
+
+ for (i = 0; i < sqns_num; i++) {
+ esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL);
+ if (!esw_sq) {
+ err = -ENOMEM;
+ goto out_err;
+ }
+
+ /* Add re-inject rule to the PF/representor sqs */
+ flow_rule = mlx5_eswitch_add_send_to_vport_rule(esw,
+ rep->vport,
+ sqns_array[i]);
+ if (IS_ERR(flow_rule)) {
+ err = PTR_ERR(flow_rule);
+ kfree(esw_sq);
+ goto out_err;
+ }
+ esw_sq->send_to_vport_rule = flow_rule;
+ list_add(&esw_sq->list, &rep->vport_sqs_list);
+ }
+ return 0;
+
+out_err:
+ mlx5e_sqs2vport_stop(esw, rep);
+ return err;
+}
+
int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv)
{
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
@@ -210,7 +263,7 @@ int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv)
sqs[num_sqs++] = c->sq[tc].sqn;
}
- err = mlx5_eswitch_sqs2vport_start(esw, rep, sqs, num_sqs);
+ err = mlx5e_sqs2vport_start(esw, rep, sqs, num_sqs);
kfree(sqs);
out:
@@ -225,7 +278,7 @@ void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv)
struct mlx5e_rep_priv *rpriv = priv->ppriv;
struct mlx5_eswitch_rep *rep = rpriv->rep;
- mlx5_eswitch_sqs2vport_stop(esw, rep);
+ mlx5e_sqs2vport_stop(esw, rep);
}
static void mlx5e_rep_neigh_update_init_interval(struct mlx5e_rep_priv *rpriv)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 23808a65889c..21b506fd2b67 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -222,6 +222,9 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
int vport,
struct ifla_vf_stats *vf_stats);
+struct mlx5_flow_handle *
+mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport,
+ u32 sqn);
struct mlx5_flow_spec;
struct mlx5_esw_flow_attr;
@@ -258,12 +261,6 @@ struct mlx5_esw_flow_attr {
struct mlx5e_tc_flow_parse_attr *parse_attr;
};
-int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw,
- struct mlx5_eswitch_rep *rep,
- u16 *sqns_array, int sqns_num);
-void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw,
- struct mlx5_eswitch_rep *rep);
-
int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode);
int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode);
int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 7ba33814c96d..3e412d609c25 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -302,7 +302,7 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
return err;
}
-static struct mlx5_flow_handle *
+struct mlx5_flow_handle *
mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn)
{
struct mlx5_flow_act flow_act = {0};
@@ -339,59 +339,6 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn
return flow_rule;
}
-void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw,
- struct mlx5_eswitch_rep *rep)
-{
- struct mlx5_esw_sq *esw_sq, *tmp;
-
- if (esw->mode != SRIOV_OFFLOADS)
- return;
-
- list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) {
- mlx5_del_flow_rules(esw_sq->send_to_vport_rule);
- list_del(&esw_sq->list);
- kfree(esw_sq);
- }
-}
-
-int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw,
- struct mlx5_eswitch_rep *rep,
- u16 *sqns_array, int sqns_num)
-{
- struct mlx5_flow_handle *flow_rule;
- struct mlx5_esw_sq *esw_sq;
- int err;
- int i;
-
- if (esw->mode != SRIOV_OFFLOADS)
- return 0;
-
- for (i = 0; i < sqns_num; i++) {
- esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL);
- if (!esw_sq) {
- err = -ENOMEM;
- goto out_err;
- }
-
- /* Add re-inject rule to the PF/representor sqs */
- flow_rule = mlx5_eswitch_add_send_to_vport_rule(esw,
- rep->vport,
- sqns_array[i]);
- if (IS_ERR(flow_rule)) {
- err = PTR_ERR(flow_rule);
- kfree(esw_sq);
- goto out_err;
- }
- esw_sq->send_to_vport_rule = flow_rule;
- list_add(&esw_sq->list, &rep->vport_sqs_list);
- }
- return 0;
-
-out_err:
- mlx5_eswitch_sqs2vport_stop(esw, rep);
- return err;
-}
-
static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
{
struct mlx5_flow_act flow_act = {0};
--
2.14.2
^ permalink raw reply related
* [for-next 03/11] net/mlx5: E-Switch, Simplify representor load/unload callback API
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
In the load() callback for loading representors we don't really need
struct mlx5_eswitch but struct mlx5_core_dev, pass it directly.
In the unload() callback for unloading representors we don't need the
struct mlx5_eswitch argument, remove it.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 14 +++++++-------
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 5 ++---
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 6 +++---
3 files changed, 12 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 4661ef12c18c..6d2219f3acf6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -967,7 +967,7 @@ static const struct mlx5e_profile mlx5e_rep_profile = {
/* e-Switch vport representors */
static int
-mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
+mlx5e_nic_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
struct mlx5e_priv *priv = netdev_priv(rep->netdev);
struct mlx5e_rep_priv *rpriv = priv->ppriv;
@@ -992,7 +992,7 @@ mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
}
static void
-mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
+mlx5e_nic_rep_unload(struct mlx5_eswitch_rep *rep)
{
struct mlx5e_priv *priv = netdev_priv(rep->netdev);
struct mlx5e_rep_priv *rpriv = priv->ppriv;
@@ -1008,7 +1008,7 @@ mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
}
static int
-mlx5e_vport_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
+mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
struct mlx5e_rep_priv *rpriv;
struct net_device *netdev;
@@ -1019,7 +1019,7 @@ mlx5e_vport_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
if (!rpriv)
return -ENOMEM;
- netdev = mlx5e_create_netdev(esw->dev, &mlx5e_rep_profile, rpriv);
+ netdev = mlx5e_create_netdev(dev, &mlx5e_rep_profile, rpriv);
if (!netdev) {
pr_warn("Failed to create representor netdev for vport %d\n",
rep->vport);
@@ -1044,7 +1044,7 @@ mlx5e_vport_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
goto err_detach_netdev;
}
- upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(esw));
+ upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(dev->priv.eswitch));
err = tc_setup_cb_egdev_register(netdev, mlx5e_setup_tc_block_cb,
upriv);
if (err)
@@ -1076,7 +1076,7 @@ mlx5e_vport_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
}
static void
-mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
+mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep)
{
struct net_device *netdev = rep->netdev;
struct mlx5e_priv *priv = netdev_priv(netdev);
@@ -1085,7 +1085,7 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep)
struct mlx5e_priv *upriv;
unregister_netdev(rep->netdev);
- upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(esw));
+ upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(priv->mdev->priv.eswitch));
tc_setup_cb_egdev_unregister(netdev, mlx5e_setup_tc_block_cb,
upriv);
mlx5e_rep_neigh_cleanup(rpriv);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9722c2a96090..23808a65889c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -139,10 +139,9 @@ struct mlx5_esw_sq {
};
struct mlx5_eswitch_rep {
- int (*load)(struct mlx5_eswitch *esw,
+ int (*load)(struct mlx5_core_dev *dev,
struct mlx5_eswitch_rep *rep);
- void (*unload)(struct mlx5_eswitch *esw,
- struct mlx5_eswitch_rep *rep);
+ void (*unload)(struct mlx5_eswitch_rep *rep);
u16 vport;
u8 hw_id[ETH_ALEN];
struct net_device *netdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 139da9f3cc70..7ba33814c96d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -778,7 +778,7 @@ static void esw_offloads_unload_reps(struct mlx5_eswitch *esw, int nvports)
if (!rep->valid)
continue;
- rep->unload(esw, rep);
+ rep->unload(rep);
}
}
@@ -793,7 +793,7 @@ static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
if (!rep->valid)
continue;
- err = rep->load(esw, rep);
+ err = rep->load(esw->dev, rep);
if (err)
goto err_reps;
}
@@ -1194,7 +1194,7 @@ void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
rep = &offloads->vport_reps[vport_index];
if (esw->mode == SRIOV_OFFLOADS && esw->vports[vport_index].enabled)
- rep->unload(esw, rep);
+ rep->unload(rep);
rep->valid = false;
}
--
2.14.2
^ permalink raw reply related
* [for-next 05/11] net/mlx5: E-Switch, Create a dedicated send to vport rule deletion function
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
In order for representors to send packets directly to VFs we use an
E-Switch function which insert special rules into the HW. For symmetry
create an E-Switch function that deletes these rules as well.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 1 +
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 5 +++++
3 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 19edaa155062..01bf4e3c8afa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -199,7 +199,7 @@ static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw,
return;
list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) {
- mlx5_del_flow_rules(esw_sq->send_to_vport_rule);
+ mlx5_eswitch_del_send_to_vport_rule(esw_sq->send_to_vport_rule);
list_del(&esw_sq->list);
kfree(esw_sq);
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 21b506fd2b67..9ed401225225 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -225,6 +225,7 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
struct mlx5_flow_handle *
mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport,
u32 sqn);
+void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule);
struct mlx5_flow_spec;
struct mlx5_esw_flow_attr;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 3e412d609c25..b37f8ac01e43 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -339,6 +339,11 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn
return flow_rule;
}
+void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule)
+{
+ mlx5_del_flow_rules(rule);
+}
+
static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
{
struct mlx5_flow_act flow_act = {0};
--
2.14.2
^ permalink raw reply related
* [for-next 06/11] net/mlx5e: Move ethernet representors data into separate struct
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
Ethernet representors have a need to store data which is applicable
only for them. Create a priv void pointer in struct mlx5_eswitch_rep
and move mlx5e to store the relevant data there. As part of this change
we also initialize rep_if in mlx5e_rep_register_vf_vports() as otherwise the
E-Switch code will copy a priv value which is garbage.
We also rename mlx5_eswitch_get_uplink_netdev() to
mlx5_eswitch_get_uplink_priv() and make it return void *.
This way E-Switch code doesn't need to deal with net devices and
we leave the task of getting it to mlx5e.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 58 ++++++++++++++--------
drivers/net/ethernet/mellanox/mlx5/core/en_rep.h | 9 ++++
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 14 ++++--
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 7 +--
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 7 ++-
5 files changed, 60 insertions(+), 35 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 01bf4e3c8afa..3c74f0599ad3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -194,11 +194,13 @@ static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw,
struct mlx5_eswitch_rep *rep)
{
struct mlx5_esw_sq *esw_sq, *tmp;
+ struct mlx5e_rep_priv *rpriv;
if (esw->mode != SRIOV_OFFLOADS)
return;
- list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) {
+ rpriv = mlx5e_rep_to_rep_priv(rep);
+ list_for_each_entry_safe(esw_sq, tmp, &rpriv->vport_sqs_list, list) {
mlx5_eswitch_del_send_to_vport_rule(esw_sq->send_to_vport_rule);
list_del(&esw_sq->list);
kfree(esw_sq);
@@ -210,6 +212,7 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
u16 *sqns_array, int sqns_num)
{
struct mlx5_flow_handle *flow_rule;
+ struct mlx5e_rep_priv *rpriv;
struct mlx5_esw_sq *esw_sq;
int err;
int i;
@@ -217,6 +220,7 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
if (esw->mode != SRIOV_OFFLOADS)
return 0;
+ rpriv = mlx5e_rep_to_rep_priv(rep);
for (i = 0; i < sqns_num; i++) {
esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL);
if (!esw_sq) {
@@ -234,7 +238,7 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
goto out_err;
}
esw_sq->send_to_vport_rule = flow_rule;
- list_add(&esw_sq->list, &rep->vport_sqs_list);
+ list_add(&esw_sq->list, &rpriv->vport_sqs_list);
}
return 0;
@@ -291,7 +295,7 @@ static void mlx5e_rep_neigh_update_init_interval(struct mlx5e_rep_priv *rpriv)
#endif
unsigned long ipv4_interval = NEIGH_VAR(&arp_tbl.parms,
DELAY_PROBE_TIME);
- struct net_device *netdev = rpriv->rep->netdev;
+ struct net_device *netdev = rpriv->netdev;
struct mlx5e_priv *priv = netdev_priv(netdev);
rpriv->neigh_update.min_interval = min_t(unsigned long, ipv6_interval, ipv4_interval);
@@ -312,7 +316,7 @@ static void mlx5e_rep_neigh_stats_work(struct work_struct *work)
{
struct mlx5e_rep_priv *rpriv = container_of(work, struct mlx5e_rep_priv,
neigh_update.neigh_stats_work.work);
- struct net_device *netdev = rpriv->rep->netdev;
+ struct net_device *netdev = rpriv->netdev;
struct mlx5e_priv *priv = netdev_priv(netdev);
struct mlx5e_neigh_hash_entry *nhe;
@@ -408,7 +412,7 @@ static int mlx5e_rep_netevent_event(struct notifier_block *nb,
struct mlx5e_rep_priv *rpriv = container_of(nb, struct mlx5e_rep_priv,
neigh_update.netevent_nb);
struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update;
- struct net_device *netdev = rpriv->rep->netdev;
+ struct net_device *netdev = rpriv->netdev;
struct mlx5e_priv *priv = netdev_priv(netdev);
struct mlx5e_neigh_hash_entry *nhe = NULL;
struct mlx5e_neigh m_neigh = {};
@@ -536,7 +540,7 @@ static int mlx5e_rep_neigh_init(struct mlx5e_rep_priv *rpriv)
static void mlx5e_rep_neigh_cleanup(struct mlx5e_rep_priv *rpriv)
{
struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update;
- struct mlx5e_priv *priv = netdev_priv(rpriv->rep->netdev);
+ struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
unregister_netevent_notifier(&neigh_update->netevent_nb);
@@ -957,7 +961,7 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv)
err = PTR_ERR(flow_rule);
goto err_destroy_direct_tirs;
}
- rep->vport_rx_rule = flow_rule;
+ rpriv->vport_rx_rule = flow_rule;
err = mlx5e_tc_init(priv);
if (err)
@@ -966,7 +970,7 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv)
return 0;
err_del_flow_rule:
- mlx5_del_flow_rules(rep->vport_rx_rule);
+ mlx5_del_flow_rules(rpriv->vport_rx_rule);
err_destroy_direct_tirs:
mlx5e_destroy_direct_tirs(priv);
err_destroy_direct_rqts:
@@ -977,10 +981,9 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv)
static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv)
{
struct mlx5e_rep_priv *rpriv = priv->ppriv;
- struct mlx5_eswitch_rep *rep = rpriv->rep;
mlx5e_tc_cleanup(priv);
- mlx5_del_flow_rules(rep->vport_rx_rule);
+ mlx5_del_flow_rules(rpriv->vport_rx_rule);
mlx5e_destroy_direct_tirs(priv);
mlx5e_destroy_direct_rqts(priv);
}
@@ -1022,8 +1025,8 @@ static const struct mlx5e_profile mlx5e_rep_profile = {
static int
mlx5e_nic_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
- struct mlx5e_priv *priv = netdev_priv(rep->netdev);
- struct mlx5e_rep_priv *rpriv = priv->ppriv;
+ struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep);
+ struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
int err;
@@ -1047,8 +1050,8 @@ mlx5e_nic_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
static void
mlx5e_nic_rep_unload(struct mlx5_eswitch_rep *rep)
{
- struct mlx5e_priv *priv = netdev_priv(rep->netdev);
- struct mlx5e_rep_priv *rpriv = priv->ppriv;
+ struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep);
+ struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
if (test_bit(MLX5E_STATE_OPENED, &priv->state))
mlx5e_remove_sqs_fwd_rules(priv);
@@ -1063,6 +1066,7 @@ mlx5e_nic_rep_unload(struct mlx5_eswitch_rep *rep)
static int
mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
{
+ struct mlx5e_rep_priv *uplink_rpriv;
struct mlx5e_rep_priv *rpriv;
struct net_device *netdev;
struct mlx5e_priv *upriv;
@@ -1080,8 +1084,10 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
return -EINVAL;
}
- rep->netdev = netdev;
+ rpriv->netdev = netdev;
rpriv->rep = rep;
+ rep->priv = rpriv;
+ INIT_LIST_HEAD(&rpriv->vport_sqs_list);
err = mlx5e_attach_netdev(netdev_priv(netdev));
if (err) {
@@ -1097,7 +1103,8 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
goto err_detach_netdev;
}
- upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(dev->priv.eswitch));
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(dev->priv.eswitch);
+ upriv = netdev_priv(uplink_rpriv->netdev);
err = tc_setup_cb_egdev_register(netdev, mlx5e_setup_tc_block_cb,
upriv);
if (err)
@@ -1131,14 +1138,16 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
static void
mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep)
{
- struct net_device *netdev = rep->netdev;
+ struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep);
+ struct net_device *netdev = rpriv->netdev;
struct mlx5e_priv *priv = netdev_priv(netdev);
- struct mlx5e_rep_priv *rpriv = priv->ppriv;
+ struct mlx5e_rep_priv *uplink_rpriv;
void *ppriv = priv->ppriv;
struct mlx5e_priv *upriv;
- unregister_netdev(rep->netdev);
- upriv = netdev_priv(mlx5_eswitch_get_uplink_netdev(priv->mdev->priv.eswitch));
+ unregister_netdev(netdev);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(priv->mdev->priv.eswitch);
+ upriv = netdev_priv(uplink_rpriv->netdev);
tc_setup_cb_egdev_unregister(netdev, mlx5e_setup_tc_block_cb,
upriv);
mlx5e_rep_neigh_cleanup(rpriv);
@@ -1155,7 +1164,7 @@ static void mlx5e_rep_register_vf_vports(struct mlx5e_priv *priv)
int vport;
for (vport = 1; vport < total_vfs; vport++) {
- struct mlx5_eswitch_rep rep;
+ struct mlx5_eswitch_rep rep = {};
rep.load = mlx5e_vport_rep_load;
rep.unload = mlx5e_vport_rep_unload;
@@ -1178,11 +1187,16 @@ void mlx5e_register_vport_reps(struct mlx5e_priv *priv)
{
struct mlx5_core_dev *mdev = priv->mdev;
struct mlx5_eswitch *esw = mdev->priv.eswitch;
+ struct mlx5e_rep_priv *rpriv;
struct mlx5_eswitch_rep rep;
+ rpriv = priv->ppriv;
+ rpriv->netdev = priv->netdev;
+
rep.load = mlx5e_nic_rep_load;
rep.unload = mlx5e_nic_rep_unload;
- rep.netdev = priv->netdev;
+ rep.priv = rpriv;
+ INIT_LIST_HEAD(&rpriv->vport_sqs_list);
mlx5_eswitch_register_vport_rep(esw, 0, &rep); /* UPLINK PF vport*/
mlx5e_rep_register_vf_vports(priv); /* VFs vports */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index 5659ed9f51e6..8db68369367e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -56,8 +56,17 @@ struct mlx5e_neigh_update_table {
struct mlx5e_rep_priv {
struct mlx5_eswitch_rep *rep;
struct mlx5e_neigh_update_table neigh_update;
+ struct net_device *netdev;
+ struct mlx5_flow_handle *vport_rx_rule;
+ struct list_head vport_sqs_list;
};
+static inline
+struct mlx5e_rep_priv *mlx5e_rep_to_rep_priv(struct mlx5_eswitch_rep *rep)
+{
+ return (struct mlx5e_rep_priv *)rep->priv;
+}
+
struct mlx5e_neigh {
struct net_device *dev;
union {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 55979ec2e88a..f462496cce7a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -617,7 +617,8 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv,
FLOW_DISSECTOR_KEY_ENC_PORTS,
f->mask);
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
- struct net_device *up_dev = mlx5_eswitch_get_uplink_netdev(esw);
+ struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ struct net_device *up_dev = uplink_rpriv->netdev;
struct mlx5e_priv *up_priv = netdev_priv(up_dev);
/* Full udp dst port must be given */
@@ -1507,6 +1508,7 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv,
int *out_ttl)
{
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
+ struct mlx5e_rep_priv *uplink_rpriv;
struct rtable *rt;
struct neighbour *n = NULL;
@@ -1520,9 +1522,10 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv,
#else
return -EOPNOTSUPP;
#endif
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
/* if the egress device isn't on the same HW e-switch, we use the uplink */
if (!switchdev_port_same_parent_id(priv->netdev, rt->dst.dev))
- *out_dev = mlx5_eswitch_get_uplink_netdev(esw);
+ *out_dev = uplink_rpriv->netdev;
else
*out_dev = rt->dst.dev;
@@ -1543,6 +1546,7 @@ static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
struct neighbour **out_n,
int *out_ttl)
{
+ struct mlx5e_rep_priv *uplink_rpriv;
struct neighbour *n = NULL;
struct dst_entry *dst;
@@ -1557,9 +1561,10 @@ static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
*out_ttl = ip6_dst_hoplimit(dst);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
/* if the egress device isn't on the same HW e-switch, we use the uplink */
if (!switchdev_port_same_parent_id(priv->netdev, dst->dev))
- *out_dev = mlx5_eswitch_get_uplink_netdev(esw);
+ *out_dev = uplink_rpriv->netdev;
else
*out_dev = dst->dev;
#else
@@ -1859,7 +1864,8 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
struct mlx5e_tc_flow *flow)
{
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
- struct net_device *up_dev = mlx5_eswitch_get_uplink_netdev(esw);
+ struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ struct net_device *up_dev = uplink_rpriv->netdev;
unsigned short family = ip_tunnel_info_af(tun_info);
struct mlx5e_priv *up_priv = netdev_priv(up_dev);
struct mlx5_esw_flow_attr *attr = flow->esw_attr;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9ed401225225..3a21ea4e4d24 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -142,12 +142,9 @@ struct mlx5_eswitch_rep {
int (*load)(struct mlx5_core_dev *dev,
struct mlx5_eswitch_rep *rep);
void (*unload)(struct mlx5_eswitch_rep *rep);
+ void *priv;
u16 vport;
u8 hw_id[ETH_ALEN];
- struct net_device *netdev;
-
- struct mlx5_flow_handle *vport_rx_rule;
- struct list_head vport_sqs_list;
u16 vlan;
u32 vlan_refcount;
bool valid;
@@ -274,7 +271,7 @@ void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw,
struct mlx5_eswitch_rep *rep);
void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
int vport_index);
-struct net_device *mlx5_eswitch_get_uplink_netdev(struct mlx5_eswitch *esw);
+void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw);
int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
struct mlx5_esw_flow_attr *attr);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index b37f8ac01e43..6560e5943e1e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1131,9 +1131,8 @@ void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw,
rep->load = __rep->load;
rep->unload = __rep->unload;
- rep->netdev = __rep->netdev;
+ rep->priv = __rep->priv;
- INIT_LIST_HEAD(&rep->vport_sqs_list);
rep->valid = true;
}
@@ -1151,12 +1150,12 @@ void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
rep->valid = false;
}
-struct net_device *mlx5_eswitch_get_uplink_netdev(struct mlx5_eswitch *esw)
+void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw)
{
#define UPLINK_REP_INDEX 0
struct mlx5_esw_offload *offloads = &esw->offloads;
struct mlx5_eswitch_rep *rep;
rep = &offloads->vport_reps[UPLINK_REP_INDEX];
- return rep->netdev;
+ return rep->priv;
}
--
2.14.2
^ permalink raw reply related
* [for-next 08/11] net/mlx5e: E-Switch, Move send-to-vport rule struct to en_rep
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
Move struct mlx5_esw_sq which keeps send-to-vport rule to from the eswitch
code to mlx5e and rename it to better reflect where it belongs
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 22 +++++++++++-----------
drivers/net/ethernet/mellanox/mlx5/core/en_rep.h | 5 +++++
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 5 -----
3 files changed, 16 insertions(+), 16 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 5b2b673c0b13..c6a77f8e99a4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -193,17 +193,17 @@ int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr)
static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw,
struct mlx5_eswitch_rep *rep)
{
- struct mlx5_esw_sq *esw_sq, *tmp;
+ struct mlx5e_rep_sq *rep_sq, *tmp;
struct mlx5e_rep_priv *rpriv;
if (esw->mode != SRIOV_OFFLOADS)
return;
rpriv = mlx5e_rep_to_rep_priv(rep);
- list_for_each_entry_safe(esw_sq, tmp, &rpriv->vport_sqs_list, list) {
- mlx5_eswitch_del_send_to_vport_rule(esw_sq->send_to_vport_rule);
- list_del(&esw_sq->list);
- kfree(esw_sq);
+ list_for_each_entry_safe(rep_sq, tmp, &rpriv->vport_sqs_list, list) {
+ mlx5_eswitch_del_send_to_vport_rule(rep_sq->send_to_vport_rule);
+ list_del(&rep_sq->list);
+ kfree(rep_sq);
}
}
@@ -213,7 +213,7 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
{
struct mlx5_flow_handle *flow_rule;
struct mlx5e_rep_priv *rpriv;
- struct mlx5_esw_sq *esw_sq;
+ struct mlx5e_rep_sq *rep_sq;
int err;
int i;
@@ -222,8 +222,8 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
rpriv = mlx5e_rep_to_rep_priv(rep);
for (i = 0; i < sqns_num; i++) {
- esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL);
- if (!esw_sq) {
+ rep_sq = kzalloc(sizeof(*rep_sq), GFP_KERNEL);
+ if (!rep_sq) {
err = -ENOMEM;
goto out_err;
}
@@ -234,11 +234,11 @@ static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
sqns_array[i]);
if (IS_ERR(flow_rule)) {
err = PTR_ERR(flow_rule);
- kfree(esw_sq);
+ kfree(rep_sq);
goto out_err;
}
- esw_sq->send_to_vport_rule = flow_rule;
- list_add(&esw_sq->list, &rpriv->vport_sqs_list);
+ rep_sq->send_to_vport_rule = flow_rule;
+ list_add(&rep_sq->list, &rpriv->vport_sqs_list);
}
return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index e4473a9ebd50..b9b481f2833a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -133,6 +133,11 @@ struct mlx5e_encap_entry {
int encap_size;
};
+struct mlx5e_rep_sq {
+ struct mlx5_flow_handle *send_to_vport_rule;
+ struct list_head list;
+};
+
void *mlx5e_alloc_nic_rep_priv(struct mlx5_core_dev *mdev);
void mlx5e_register_vport_reps(struct mlx5e_priv *priv);
void mlx5e_unregister_vport_reps(struct mlx5e_priv *priv);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 91175965df7f..3b481182f13a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -138,11 +138,6 @@ struct mlx5_eswitch_fdb {
};
};
-struct mlx5_esw_sq {
- struct mlx5_flow_handle *send_to_vport_rule;
- struct list_head list;
-};
-
struct mlx5_eswitch_rep;
struct mlx5_eswitch_rep_if {
int (*load)(struct mlx5_core_dev *dev,
--
2.14.2
^ permalink raw reply related
* [for-next 07/11] net/mlx5: E-Switch, Create generic header struct to be used by representors
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Mark Bloch, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Mark Bloch <markb@mellanox.com>
Now that we don't store type dependent data in struct mlx5_eswitch_rep
we can create a generic interface, and representor type.
struct mlx5_eswitch_rep will store an array of interfaces, each
interface is used by a different representor type.
Once we moved to a more generic interface, rdma driver representors can
be added and utilize the same mechanism as the Ethernet driver
representors use.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 29 ++++-----
drivers/net/ethernet/mellanox/mlx5/core/en_rep.h | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 9 +--
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 22 +++++--
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 72 ++++++++++++++++------
5 files changed, 90 insertions(+), 44 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 3c74f0599ad3..5b2b673c0b13 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -1086,7 +1086,7 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
rpriv->netdev = netdev;
rpriv->rep = rep;
- rep->priv = rpriv;
+ rep->rep_if[REP_ETH].priv = rpriv;
INIT_LIST_HEAD(&rpriv->vport_sqs_list);
err = mlx5e_attach_netdev(netdev_priv(netdev));
@@ -1103,7 +1103,7 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
goto err_detach_netdev;
}
- uplink_rpriv = mlx5_eswitch_get_uplink_priv(dev->priv.eswitch);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(dev->priv.eswitch, REP_ETH);
upriv = netdev_priv(uplink_rpriv->netdev);
err = tc_setup_cb_egdev_register(netdev, mlx5e_setup_tc_block_cb,
upriv);
@@ -1146,7 +1146,8 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep)
struct mlx5e_priv *upriv;
unregister_netdev(netdev);
- uplink_rpriv = mlx5_eswitch_get_uplink_priv(priv->mdev->priv.eswitch);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(priv->mdev->priv.eswitch,
+ REP_ETH);
upriv = netdev_priv(uplink_rpriv->netdev);
tc_setup_cb_egdev_unregister(netdev, mlx5e_setup_tc_block_cb,
upriv);
@@ -1164,11 +1165,11 @@ static void mlx5e_rep_register_vf_vports(struct mlx5e_priv *priv)
int vport;
for (vport = 1; vport < total_vfs; vport++) {
- struct mlx5_eswitch_rep rep = {};
+ struct mlx5_eswitch_rep_if rep_if = {};
- rep.load = mlx5e_vport_rep_load;
- rep.unload = mlx5e_vport_rep_unload;
- mlx5_eswitch_register_vport_rep(esw, vport, &rep);
+ rep_if.load = mlx5e_vport_rep_load;
+ rep_if.unload = mlx5e_vport_rep_unload;
+ mlx5_eswitch_register_vport_rep(esw, vport, &rep_if, REP_ETH);
}
}
@@ -1180,24 +1181,24 @@ static void mlx5e_rep_unregister_vf_vports(struct mlx5e_priv *priv)
int vport;
for (vport = 1; vport < total_vfs; vport++)
- mlx5_eswitch_unregister_vport_rep(esw, vport);
+ mlx5_eswitch_unregister_vport_rep(esw, vport, REP_ETH);
}
void mlx5e_register_vport_reps(struct mlx5e_priv *priv)
{
struct mlx5_core_dev *mdev = priv->mdev;
struct mlx5_eswitch *esw = mdev->priv.eswitch;
+ struct mlx5_eswitch_rep_if rep_if;
struct mlx5e_rep_priv *rpriv;
- struct mlx5_eswitch_rep rep;
rpriv = priv->ppriv;
rpriv->netdev = priv->netdev;
- rep.load = mlx5e_nic_rep_load;
- rep.unload = mlx5e_nic_rep_unload;
- rep.priv = rpriv;
+ rep_if.load = mlx5e_nic_rep_load;
+ rep_if.unload = mlx5e_nic_rep_unload;
+ rep_if.priv = rpriv;
INIT_LIST_HEAD(&rpriv->vport_sqs_list);
- mlx5_eswitch_register_vport_rep(esw, 0, &rep); /* UPLINK PF vport*/
+ mlx5_eswitch_register_vport_rep(esw, 0, &rep_if, REP_ETH); /* UPLINK PF vport*/
mlx5e_rep_register_vf_vports(priv); /* VFs vports */
}
@@ -1208,7 +1209,7 @@ void mlx5e_unregister_vport_reps(struct mlx5e_priv *priv)
struct mlx5_eswitch *esw = mdev->priv.eswitch;
mlx5e_rep_unregister_vf_vports(priv); /* VFs vports */
- mlx5_eswitch_unregister_vport_rep(esw, 0); /* UPLINK PF*/
+ mlx5_eswitch_unregister_vport_rep(esw, 0, REP_ETH); /* UPLINK PF*/
}
void *mlx5e_alloc_nic_rep_priv(struct mlx5_core_dev *mdev)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
index 8db68369367e..e4473a9ebd50 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h
@@ -64,7 +64,7 @@ struct mlx5e_rep_priv {
static inline
struct mlx5e_rep_priv *mlx5e_rep_to_rep_priv(struct mlx5_eswitch_rep *rep)
{
- return (struct mlx5e_rep_priv *)rep->priv;
+ return (struct mlx5e_rep_priv *)rep->rep_if[REP_ETH].priv;
}
struct mlx5e_neigh {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index f462496cce7a..259e91e2d09a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -617,7 +617,7 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv,
FLOW_DISSECTOR_KEY_ENC_PORTS,
f->mask);
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
- struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH);
struct net_device *up_dev = uplink_rpriv->netdev;
struct mlx5e_priv *up_priv = netdev_priv(up_dev);
@@ -1522,7 +1522,7 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv,
#else
return -EOPNOTSUPP;
#endif
- uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH);
/* if the egress device isn't on the same HW e-switch, we use the uplink */
if (!switchdev_port_same_parent_id(priv->netdev, rt->dst.dev))
*out_dev = uplink_rpriv->netdev;
@@ -1561,7 +1561,7 @@ static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
*out_ttl = ip6_dst_hoplimit(dst);
- uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH);
/* if the egress device isn't on the same HW e-switch, we use the uplink */
if (!switchdev_port_same_parent_id(priv->netdev, dst->dev))
*out_dev = uplink_rpriv->netdev;
@@ -1864,7 +1864,8 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
struct mlx5e_tc_flow *flow)
{
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
- struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw);
+ struct mlx5e_rep_priv *uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw,
+ REP_ETH);
struct net_device *up_dev = uplink_rpriv->netdev;
unsigned short family = ip_tunnel_info_af(tun_info);
struct mlx5e_priv *up_priv = netdev_priv(up_dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 3a21ea4e4d24..91175965df7f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -45,6 +45,11 @@ enum {
SRIOV_OFFLOADS
};
+enum {
+ REP_ETH,
+ NUM_REP_TYPES,
+};
+
#ifdef CONFIG_MLX5_ESWITCH
#define MLX5_MAX_UC_PER_VPORT(dev) \
@@ -138,16 +143,21 @@ struct mlx5_esw_sq {
struct list_head list;
};
-struct mlx5_eswitch_rep {
+struct mlx5_eswitch_rep;
+struct mlx5_eswitch_rep_if {
int (*load)(struct mlx5_core_dev *dev,
struct mlx5_eswitch_rep *rep);
void (*unload)(struct mlx5_eswitch_rep *rep);
void *priv;
+ bool valid;
+};
+
+struct mlx5_eswitch_rep {
+ struct mlx5_eswitch_rep_if rep_if[NUM_REP_TYPES];
u16 vport;
u8 hw_id[ETH_ALEN];
u16 vlan;
u32 vlan_refcount;
- bool valid;
};
struct mlx5_esw_offload {
@@ -268,10 +278,12 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, u8 encap);
int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink, u8 *encap);
void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw,
int vport_index,
- struct mlx5_eswitch_rep *rep);
+ struct mlx5_eswitch_rep_if *rep_if,
+ u8 rep_type);
void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
- int vport_index);
-void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw);
+ int vport_index,
+ u8 rep_type);
+void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type);
int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
struct mlx5_esw_flow_attr *attr);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 6560e5943e1e..ee617f7d3e84 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -130,7 +130,7 @@ static int esw_set_global_vlan_pop(struct mlx5_eswitch *esw, u8 val)
esw_debug(esw->dev, "%s applying global %s policy\n", __func__, val ? "pop" : "none");
for (vf_vport = 1; vf_vport < esw->enabled_vports; vf_vport++) {
rep = &esw->offloads.vport_reps[vf_vport];
- if (!rep->valid)
+ if (!rep->rep_if[REP_ETH].valid)
continue;
err = __mlx5_eswitch_set_vport_vlan(esw, rep->vport, 0, 0, val);
@@ -720,21 +720,31 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
return 0;
}
-static void esw_offloads_unload_reps(struct mlx5_eswitch *esw, int nvports)
+static void esw_offloads_unload_reps_type(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
{
struct mlx5_eswitch_rep *rep;
int vport;
for (vport = nvports - 1; vport >= 0; vport--) {
rep = &esw->offloads.vport_reps[vport];
- if (!rep->valid)
+ if (!rep->rep_if[rep_type].valid)
continue;
- rep->unload(rep);
+ rep->rep_if[rep_type].unload(rep);
}
}
-static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
+static void esw_offloads_unload_reps(struct mlx5_eswitch *esw, int nvports)
+{
+ u8 rep_type = NUM_REP_TYPES;
+
+ while (rep_type-- > 0)
+ esw_offloads_unload_reps_type(esw, nvports, rep_type);
+}
+
+static int esw_offloads_load_reps_type(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
{
struct mlx5_eswitch_rep *rep;
int vport;
@@ -742,10 +752,10 @@ static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
for (vport = 0; vport < nvports; vport++) {
rep = &esw->offloads.vport_reps[vport];
- if (!rep->valid)
+ if (!rep->rep_if[rep_type].valid)
continue;
- err = rep->load(esw->dev, rep);
+ err = rep->rep_if[rep_type].load(esw->dev, rep);
if (err)
goto err_reps;
}
@@ -753,7 +763,28 @@ static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
return 0;
err_reps:
- esw_offloads_unload_reps(esw, vport);
+ esw_offloads_unload_reps_type(esw, vport, rep_type);
+ return err;
+}
+
+static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
+{
+ u8 rep_type = 0;
+ int err;
+
+ while (rep_type < NUM_REP_TYPES) {
+ err = esw_offloads_load_reps_type(esw, nvports,
+ rep_type);
+ if (err)
+ goto err_reps;
+ rep_type++;
+ }
+
+ return err;
+
+err_reps:
+ while (rep_type-- > 0)
+ esw_offloads_unload_reps_type(esw, nvports, rep_type);
return err;
}
@@ -1122,22 +1153,23 @@ int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink, u8 *encap)
void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw,
int vport_index,
- struct mlx5_eswitch_rep *__rep)
+ struct mlx5_eswitch_rep_if *__rep_if,
+ u8 rep_type)
{
struct mlx5_esw_offload *offloads = &esw->offloads;
- struct mlx5_eswitch_rep *rep;
+ struct mlx5_eswitch_rep_if *rep_if;
- rep = &offloads->vport_reps[vport_index];
+ rep_if = &offloads->vport_reps[vport_index].rep_if[rep_type];
- rep->load = __rep->load;
- rep->unload = __rep->unload;
- rep->priv = __rep->priv;
+ rep_if->load = __rep_if->load;
+ rep_if->unload = __rep_if->unload;
+ rep_if->priv = __rep_if->priv;
- rep->valid = true;
+ rep_if->valid = true;
}
void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
- int vport_index)
+ int vport_index, u8 rep_type)
{
struct mlx5_esw_offload *offloads = &esw->offloads;
struct mlx5_eswitch_rep *rep;
@@ -1145,17 +1177,17 @@ void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw,
rep = &offloads->vport_reps[vport_index];
if (esw->mode == SRIOV_OFFLOADS && esw->vports[vport_index].enabled)
- rep->unload(rep);
+ rep->rep_if[rep_type].unload(rep);
- rep->valid = false;
+ rep->rep_if[rep_type].valid = false;
}
-void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw)
+void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type)
{
#define UPLINK_REP_INDEX 0
struct mlx5_esw_offload *offloads = &esw->offloads;
struct mlx5_eswitch_rep *rep;
rep = &offloads->vport_reps[UPLINK_REP_INDEX];
- return rep->priv;
+ return rep->rep_if[rep_type].priv;
}
--
2.14.2
^ permalink raw reply related
* [for-next 09/11] net/mlx5e: E-Switch, Use the name of static array instead of its address
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Using the address of a static array is the same as using its name (in
this specific use-case), but it's confusing and makes the code less
readable.
Fixes: 1bd27b11c1df ("net/mlx5: Introduce E-switch QoS management")
Fixes: bd77bf1cb595 ("net/mlx5: Add SRIOV VF max rate configuration support")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 26 +++++++++++------------
1 file changed, 13 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 6d4cbdb69823..cdf65ed8714c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1290,7 +1290,7 @@ static int esw_create_tsar(struct mlx5_eswitch *esw)
err = mlx5_create_scheduling_element_cmd(dev,
SCHEDULING_HIERARCHY_E_SWITCH,
- &tsar_ctx,
+ tsar_ctx,
&esw->qos.root_tsar_id);
if (err) {
esw_warn(esw->dev, "E-Switch create TSAR failed (%d)\n", err);
@@ -1333,20 +1333,20 @@ static int esw_vport_enable_qos(struct mlx5_eswitch *esw, int vport_num,
if (vport->qos.enabled)
return -EEXIST;
- MLX5_SET(scheduling_context, &sched_ctx, element_type,
+ MLX5_SET(scheduling_context, sched_ctx, element_type,
SCHEDULING_CONTEXT_ELEMENT_TYPE_VPORT);
- vport_elem = MLX5_ADDR_OF(scheduling_context, &sched_ctx,
+ vport_elem = MLX5_ADDR_OF(scheduling_context, sched_ctx,
element_attributes);
MLX5_SET(vport_element, vport_elem, vport_number, vport_num);
- MLX5_SET(scheduling_context, &sched_ctx, parent_element_id,
+ MLX5_SET(scheduling_context, sched_ctx, parent_element_id,
esw->qos.root_tsar_id);
- MLX5_SET(scheduling_context, &sched_ctx, max_average_bw,
+ MLX5_SET(scheduling_context, sched_ctx, max_average_bw,
initial_max_rate);
- MLX5_SET(scheduling_context, &sched_ctx, bw_share, initial_bw_share);
+ MLX5_SET(scheduling_context, sched_ctx, bw_share, initial_bw_share);
err = mlx5_create_scheduling_element_cmd(dev,
SCHEDULING_HIERARCHY_E_SWITCH,
- &sched_ctx,
+ sched_ctx,
&vport->qos.esw_tsar_ix);
if (err) {
esw_warn(esw->dev, "E-Switch create TSAR vport element failed (vport=%d,err=%d)\n",
@@ -1392,22 +1392,22 @@ static int esw_vport_qos_config(struct mlx5_eswitch *esw, int vport_num,
if (!vport->qos.enabled)
return -EIO;
- MLX5_SET(scheduling_context, &sched_ctx, element_type,
+ MLX5_SET(scheduling_context, sched_ctx, element_type,
SCHEDULING_CONTEXT_ELEMENT_TYPE_VPORT);
- vport_elem = MLX5_ADDR_OF(scheduling_context, &sched_ctx,
+ vport_elem = MLX5_ADDR_OF(scheduling_context, sched_ctx,
element_attributes);
MLX5_SET(vport_element, vport_elem, vport_number, vport_num);
- MLX5_SET(scheduling_context, &sched_ctx, parent_element_id,
+ MLX5_SET(scheduling_context, sched_ctx, parent_element_id,
esw->qos.root_tsar_id);
- MLX5_SET(scheduling_context, &sched_ctx, max_average_bw,
+ MLX5_SET(scheduling_context, sched_ctx, max_average_bw,
max_rate);
- MLX5_SET(scheduling_context, &sched_ctx, bw_share, bw_share);
+ MLX5_SET(scheduling_context, sched_ctx, bw_share, bw_share);
bitmask |= MODIFY_SCHEDULING_ELEMENT_IN_MODIFY_BITMASK_MAX_AVERAGE_BW;
bitmask |= MODIFY_SCHEDULING_ELEMENT_IN_MODIFY_BITMASK_BW_SHARE;
err = mlx5_modify_scheduling_element_cmd(dev,
SCHEDULING_HIERARCHY_E_SWITCH,
- &sched_ctx,
+ sched_ctx,
vport->qos.esw_tsar_ix,
bitmask);
if (err) {
--
2.14.2
^ permalink raw reply related
* [for-next 10/11] net/mlx5: Fix ingress/egress naming mistake
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
The functions names do not represent their actions, switch the mistaken
ingress/egress naming.
Fixes: fba53f7b5719 ("net/mlx5: Introduce mlx5_flow_steering structure")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index c70fd663a633..5e786e29f93a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2406,7 +2406,7 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
return PTR_ERR(prio);
}
-static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering)
+static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering)
{
struct fs_prio *prio;
@@ -2420,7 +2420,7 @@ static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering)
return PTR_ERR_OR_ZERO(prio);
}
-static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering)
+static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering)
{
struct fs_prio *prio;
--
2.14.2
^ permalink raw reply related
* [for-next 11/11] net/mlx5: Separate ingress/egress namespaces for each vport
From: Saeed Mahameed @ 2017-12-19 20:33 UTC (permalink / raw)
To: David S. Miller, Doug Ledford
Cc: netdev, linux-rdma, Leon Romanovsky, Gal Pressman, Saeed Mahameed
In-Reply-To: <20171219203340.2600-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Each vport has its own root flow table for the ACL flow tables and root
flow table is per namespace, therefore we should create a namespace for
each vport.
Fixes: efdc810ba39d ("net/mlx5: Flow steering, Add vport ACL support")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 10 +-
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 145 ++++++++++++++++++----
drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 4 +-
include/linux/mlx5/fs.h | 4 +
4 files changed, 133 insertions(+), 30 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index cdf65ed8714c..7649e36653d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -867,9 +867,10 @@ static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
esw_debug(dev, "Create vport[%d] egress ACL log_max_size(%d)\n",
vport->vport, MLX5_CAP_ESW_EGRESS_ACL(dev, log_max_ft_size));
- root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS);
+ root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS,
+ vport->vport);
if (!root_ns) {
- esw_warn(dev, "Failed to get E-Switch egress flow namespace\n");
+ esw_warn(dev, "Failed to get E-Switch egress flow namespace for vport (%d)\n", vport->vport);
return -EOPNOTSUPP;
}
@@ -984,9 +985,10 @@ static int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
esw_debug(dev, "Create vport[%d] ingress ACL log_max_size(%d)\n",
vport->vport, MLX5_CAP_ESW_INGRESS_ACL(dev, log_max_ft_size));
- root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS);
+ root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS,
+ vport->vport);
if (!root_ns) {
- esw_warn(dev, "Failed to get E-Switch ingress flow namespace\n");
+ esw_warn(dev, "Failed to get E-Switch ingress flow namespace for vport (%d)\n", vport->vport);
return -EOPNOTSUPP;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 5e786e29f93a..45e75b1010f7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2014,16 +2014,6 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
return &steering->fdb_root_ns->ns;
else
return NULL;
- case MLX5_FLOW_NAMESPACE_ESW_EGRESS:
- if (steering->esw_egress_root_ns)
- return &steering->esw_egress_root_ns->ns;
- else
- return NULL;
- case MLX5_FLOW_NAMESPACE_ESW_INGRESS:
- if (steering->esw_ingress_root_ns)
- return &steering->esw_ingress_root_ns->ns;
- else
- return NULL;
case MLX5_FLOW_NAMESPACE_SNIFFER_RX:
if (steering->sniffer_rx_root_ns)
return &steering->sniffer_rx_root_ns->ns;
@@ -2054,6 +2044,33 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
}
EXPORT_SYMBOL(mlx5_get_flow_namespace);
+struct mlx5_flow_namespace *mlx5_get_flow_vport_acl_namespace(struct mlx5_core_dev *dev,
+ enum mlx5_flow_namespace_type type,
+ int vport)
+{
+ struct mlx5_flow_steering *steering = dev->priv.steering;
+
+ if (!steering || vport >= MLX5_TOTAL_VPORTS(dev))
+ return NULL;
+
+ switch (type) {
+ case MLX5_FLOW_NAMESPACE_ESW_EGRESS:
+ if (steering->esw_egress_root_ns &&
+ steering->esw_egress_root_ns[vport])
+ return &steering->esw_egress_root_ns[vport]->ns;
+ else
+ return NULL;
+ case MLX5_FLOW_NAMESPACE_ESW_INGRESS:
+ if (steering->esw_ingress_root_ns &&
+ steering->esw_ingress_root_ns[vport])
+ return &steering->esw_ingress_root_ns[vport]->ns;
+ else
+ return NULL;
+ default:
+ return NULL;
+ }
+}
+
static struct fs_prio *fs_create_prio(struct mlx5_flow_namespace *ns,
unsigned int prio, int num_levels)
{
@@ -2331,13 +2348,41 @@ static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns)
clean_tree(&root_ns->ns.node);
}
+static void cleanup_egress_acls_root_ns(struct mlx5_core_dev *dev)
+{
+ struct mlx5_flow_steering *steering = dev->priv.steering;
+ int i;
+
+ if (!steering->esw_egress_root_ns)
+ return;
+
+ for (i = 0; i < MLX5_TOTAL_VPORTS(dev); i++)
+ cleanup_root_ns(steering->esw_egress_root_ns[i]);
+
+ kfree(steering->esw_egress_root_ns);
+}
+
+static void cleanup_ingress_acls_root_ns(struct mlx5_core_dev *dev)
+{
+ struct mlx5_flow_steering *steering = dev->priv.steering;
+ int i;
+
+ if (!steering->esw_ingress_root_ns)
+ return;
+
+ for (i = 0; i < MLX5_TOTAL_VPORTS(dev); i++)
+ cleanup_root_ns(steering->esw_ingress_root_ns[i]);
+
+ kfree(steering->esw_ingress_root_ns);
+}
+
void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
{
struct mlx5_flow_steering *steering = dev->priv.steering;
cleanup_root_ns(steering->root_ns);
- cleanup_root_ns(steering->esw_egress_root_ns);
- cleanup_root_ns(steering->esw_ingress_root_ns);
+ cleanup_egress_acls_root_ns(dev);
+ cleanup_ingress_acls_root_ns(dev);
cleanup_root_ns(steering->fdb_root_ns);
cleanup_root_ns(steering->sniffer_rx_root_ns);
cleanup_root_ns(steering->sniffer_tx_root_ns);
@@ -2406,34 +2451,86 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
return PTR_ERR(prio);
}
-static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering)
+static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering, int vport)
{
struct fs_prio *prio;
- steering->esw_egress_root_ns = create_root_ns(steering, FS_FT_ESW_EGRESS_ACL);
- if (!steering->esw_egress_root_ns)
+ steering->esw_egress_root_ns[vport] = create_root_ns(steering, FS_FT_ESW_EGRESS_ACL);
+ if (!steering->esw_egress_root_ns[vport])
return -ENOMEM;
/* create 1 prio*/
- prio = fs_create_prio(&steering->esw_egress_root_ns->ns, 0,
- MLX5_TOTAL_VPORTS(steering->dev));
+ prio = fs_create_prio(&steering->esw_egress_root_ns[vport]->ns, 0, 1);
return PTR_ERR_OR_ZERO(prio);
}
-static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering)
+static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering, int vport)
{
struct fs_prio *prio;
- steering->esw_ingress_root_ns = create_root_ns(steering, FS_FT_ESW_INGRESS_ACL);
- if (!steering->esw_ingress_root_ns)
+ steering->esw_ingress_root_ns[vport] = create_root_ns(steering, FS_FT_ESW_INGRESS_ACL);
+ if (!steering->esw_ingress_root_ns[vport])
return -ENOMEM;
/* create 1 prio*/
- prio = fs_create_prio(&steering->esw_ingress_root_ns->ns, 0,
- MLX5_TOTAL_VPORTS(steering->dev));
+ prio = fs_create_prio(&steering->esw_ingress_root_ns[vport]->ns, 0, 1);
return PTR_ERR_OR_ZERO(prio);
}
+static int init_egress_acls_root_ns(struct mlx5_core_dev *dev)
+{
+ struct mlx5_flow_steering *steering = dev->priv.steering;
+ int err;
+ int i;
+
+ steering->esw_egress_root_ns = kcalloc(MLX5_TOTAL_VPORTS(dev),
+ sizeof(*steering->esw_egress_root_ns),
+ GFP_KERNEL);
+ if (!steering->esw_egress_root_ns)
+ return -ENOMEM;
+
+ for (i = 0; i < MLX5_TOTAL_VPORTS(dev); i++) {
+ err = init_egress_acl_root_ns(steering, i);
+ if (err)
+ goto cleanup_root_ns;
+ }
+
+ return 0;
+
+cleanup_root_ns:
+ for (i--; i >= 0; i--)
+ cleanup_root_ns(steering->esw_egress_root_ns[i]);
+ kfree(steering->esw_egress_root_ns);
+ return err;
+}
+
+static int init_ingress_acls_root_ns(struct mlx5_core_dev *dev)
+{
+ struct mlx5_flow_steering *steering = dev->priv.steering;
+ int err;
+ int i;
+
+ steering->esw_ingress_root_ns = kcalloc(MLX5_TOTAL_VPORTS(dev),
+ sizeof(*steering->esw_ingress_root_ns),
+ GFP_KERNEL);
+ if (!steering->esw_ingress_root_ns)
+ return -ENOMEM;
+
+ for (i = 0; i < MLX5_TOTAL_VPORTS(dev); i++) {
+ err = init_ingress_acl_root_ns(steering, i);
+ if (err)
+ goto cleanup_root_ns;
+ }
+
+ return 0;
+
+cleanup_root_ns:
+ for (i--; i >= 0; i--)
+ cleanup_root_ns(steering->esw_ingress_root_ns[i]);
+ kfree(steering->esw_ingress_root_ns);
+ return err;
+}
+
int mlx5_init_fs(struct mlx5_core_dev *dev)
{
struct mlx5_flow_steering *steering;
@@ -2476,12 +2573,12 @@ int mlx5_init_fs(struct mlx5_core_dev *dev)
goto err;
}
if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) {
- err = init_egress_acl_root_ns(steering);
+ err = init_egress_acls_root_ns(dev);
if (err)
goto err;
}
if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) {
- err = init_ingress_acl_root_ns(steering);
+ err = init_ingress_acls_root_ns(dev);
if (err)
goto err;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index 397d24a621a4..3e571045626f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -71,8 +71,8 @@ struct mlx5_flow_steering {
struct kmem_cache *ftes_cache;
struct mlx5_flow_root_namespace *root_ns;
struct mlx5_flow_root_namespace *fdb_root_ns;
- struct mlx5_flow_root_namespace *esw_egress_root_ns;
- struct mlx5_flow_root_namespace *esw_ingress_root_ns;
+ struct mlx5_flow_root_namespace **esw_egress_root_ns;
+ struct mlx5_flow_root_namespace **esw_ingress_root_ns;
struct mlx5_flow_root_namespace *sniffer_tx_root_ns;
struct mlx5_flow_root_namespace *sniffer_rx_root_ns;
};
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index b25e7baa273e..a0b48afcb422 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -95,6 +95,10 @@ struct mlx5_flow_destination {
struct mlx5_flow_namespace *
mlx5_get_flow_namespace(struct mlx5_core_dev *dev,
enum mlx5_flow_namespace_type type);
+struct mlx5_flow_namespace *
+mlx5_get_flow_vport_acl_namespace(struct mlx5_core_dev *dev,
+ enum mlx5_flow_namespace_type type,
+ int vport);
struct mlx5_flow_table *
mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
--
2.14.2
^ permalink raw reply related
* sparc64 verifier failures..
From: David Miller @ 2017-12-19 20:36 UTC (permalink / raw)
To: daniel; +Cc: netdev
I'm getting about 100 verifier failures on sparc64.
The vast majority of them seem to be due to misaligned packet
accesses. Here is a sample of some of the failures.
#32/p invalid fp arithmetic FAIL
Unexpected error message!
0: (b7) r0 = 0
1: (bf) r1 = r10
2: (17) r1 -= 8
3: (7b) *(u64 *)(r1 +0) = r0
misaligned access off (0x0; 0xffffffffffffffff)+0+0 size 8
...
#160/p raw_stack: skb_load_bytes, spilled regs corruption FAIL
Unexpected error message!
0: (b7) r2 = 4
1: (bf) r6 = r10
2: (07) r6 += -8
3: (7b) *(u64 *)(r6 +0) = r1
4: (bf) r3 = r6
5: (b7) r4 = 8
6: (85) call bpf_skb_load_bytes#26
7: (79) r0 = *(u64 *)(r6 +0)
8: (61) r0 = *(u32 *)(r0 +8)
misaligned access off (0x0; 0xffffffffffffffff)+0+8 size 4
#161/p raw_stack: skb_load_bytes, spilled regs corruption 2 FAIL
Unexpected error message!
0: (b7) r2 = 4
1: (bf) r6 = r10
2: (07) r6 += -16
3: (7b) *(u64 *)(r6 -8) = r1
4: (7b) *(u64 *)(r6 +0) = r1
5: (7b) *(u64 *)(r6 +8) = r1
6: (bf) r3 = r6
7: (b7) r4 = 8
8: (85) call bpf_skb_load_bytes#26
9: (79) r0 = *(u64 *)(r6 -8)
10: (79) r2 = *(u64 *)(r6 +8)
11: (79) r3 = *(u64 *)(r6 +0)
12: (61) r0 = *(u32 *)(r0 +8)
13: (61) r2 = *(u32 *)(r2 +32)
14: (0f) r0 += r2
15: (61) r3 = *(u32 *)(r3 +4)
misaligned access off (0x0; 0xffffffffffffffff)+0+4 size 4
...
#184/p direct packet access: test15 (spill with xadd) FAIL
Unexpected error message!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (bf) r0 = r2
3: (07) r0 += 8
4: (2d) if r0 > r3 goto pc+8
R0=pkt(id=0,off=8,r=8,imm=0) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=8,imm=0) R3=pkt_end(id=0,off=0,imm=0) R10=fp0,call_-1
5: (b7) r5 = 4096
6: (bf) r4 = r10
7: (07) r4 += -8
8: (7b) *(u64 *)(r4 +0) = r2
9: (db) lock *(u64 *)(r4 +0) += r5
10: (79) r2 = *(u64 *)(r4 +0)
11: (63) *(u32 *)(r2 +0) = r5
misaligned access off (0x0; 0xffffffffffffffff)+0+0 size 4
...
#189/p direct packet access: test20 (x += pkt_ptr, 1) FAIL
Failed to load prog 'Permission denied'!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (b7) r0 = -1
3: (7b) *(u64 *)(r10 -8) = r0
4: (79) r0 = *(u64 *)(r10 -8)
5: (57) r0 &= 32767
6: (bf) r4 = r0
7: (0f) r4 += r2
8: (bf) r5 = r4
9: (07) r4 += 32766
10: (2d) if r4 > r3 goto pc+1
R0=inv(id=0,umax_value=32767,var_off=(0x0; 0x7fff)) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=0,imm=0) R3=pkt_end(id=0,off=0,imm=0) R4=pkt(id=1,off=32766,r=32766,umax_value=32767,var_off=(0x0; 0x7fff)) R5=pkt(id=1,off=0,r=32766,umax_value=32767,var_off=(0x0; 0x7fff)) R10=fp0,call_-1
11: (7b) *(u64 *)(r5 +0) = r4
misaligned packet access off 2+(0x0; 0x7fff)+0+0 size 8
#190/p direct packet access: test21 (x += pkt_ptr, 2) FAIL
Failed to load prog 'Permission denied'!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (bf) r0 = r2
3: (07) r0 += 8
4: (2d) if r0 > r3 goto pc+9
R0=pkt(id=0,off=8,r=8,imm=0) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=8,imm=0) R3=pkt_end(id=0,off=0,imm=0) R10=fp0,call_-1
5: (b7) r4 = -1
6: (7b) *(u64 *)(r10 -8) = r4
7: (79) r4 = *(u64 *)(r10 -8)
8: (57) r4 &= 32767
9: (0f) r4 += r2
10: (bf) r5 = r4
11: (07) r4 += 32766
12: (2d) if r4 > r3 goto pc+1
R0=pkt(id=0,off=8,r=8,imm=0) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=8,imm=0) R3=pkt_end(id=0,off=0,imm=0) R4=pkt(id=1,off=32766,r=32766,umax_value=32767,var_off=(0x0; 0x7fff)) R5=pkt(id=1,off=0,r=32766,umax_value=32767,var_off=(0x0; 0x7fff)) R10=fp0,call_-1
13: (7b) *(u64 *)(r5 +0) = r4
misaligned packet access off 2+(0x0; 0x7fff)+0+0 size 8
#191/p direct packet access: test22 (x += pkt_ptr, 3) FAIL
Failed to load prog 'Permission denied'!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (bf) r0 = r2
3: (07) r0 += 8
4: (7b) *(u64 *)(r10 -8) = r2
5: (7b) *(u64 *)(r10 -16) = r3
6: (79) r3 = *(u64 *)(r10 -16)
7: (2d) if r0 > r3 goto pc+11
R0=pkt(id=0,off=8,r=8,imm=0) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=8,imm=0) R3=pkt_end(id=0,off=0,imm=0) R10=fp0,call_-1 fp-8=pkt fp-16=pkt_end
8: (79) r2 = *(u64 *)(r10 -8)
9: (b7) r4 = -1
10: (db) lock *(u64 *)(r10 -8) += r4
11: (79) r4 = *(u64 *)(r10 -8)
12: (77) r4 >>= 49
13: (0f) r4 += r2
14: (bf) r0 = r4
15: (07) r0 += 2
16: (2d) if r0 > r3 goto pc+2
R0=pkt(id=1,off=2,r=2,umax_value=32767,var_off=(0x0; 0x7fff)) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=8,imm=0) R3=pkt_end(id=0,off=0,imm=0) R4=pkt(id=1,off=0,r=2,umax_value=32767,var_off=(0x0; 0x7fff)) R10=fp0,call_-1 fp-16=pkt_end
17: (b7) r2 = 1
18: (6b) *(u16 *)(r4 +0) = r2
misaligned packet access off 2+(0x0; 0x7fff)+0+0 size 2
#192/p direct packet access: test23 (x += pkt_ptr, 4) FAIL
Unexpected error message!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (b7) r0 = -1
3: (7b) *(u64 *)(r10 -8) = r0
4: (79) r0 = *(u64 *)(r10 -8)
5: (57) r0 &= 65535
6: (bf) r4 = r0
7: (b7) r0 = 31
8: (0f) r0 += r4
9: (0f) r0 += r2
10: (bf) r5 = r0
11: (07) r0 += 65534
12: (2d) if r0 > r3 goto pc+1
R0=pkt(id=1,off=65534,r=0,umin_value=31,umax_value=65566,var_off=(0x0; 0x1ffff)) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=0,imm=0) R3=pkt_end(id=0,off=0,imm=0) R4=inv(id=0,umax_value=65535,var_off=(0x0; 0xffff)) R5=pkt(id=1,off=0,r=0,umin_value=31,umax_value=65566,var_off=(0x0; 0x1ffff)) R10=fp0,call_-1
13: (7b) *(u64 *)(r5 +0) = r0
misaligned packet access off 2+(0x0; 0x1ffff)+0+0 size 8
#193/p direct packet access: test24 (x += pkt_ptr, 5) FAIL
Failed to load prog 'Permission denied'!
0: (61) r2 = *(u32 *)(r1 +76)
1: (61) r3 = *(u32 *)(r1 +80)
2: (b7) r0 = -1
3: (7b) *(u64 *)(r10 -8) = r0
4: (79) r0 = *(u64 *)(r10 -8)
5: (57) r0 &= 255
6: (bf) r4 = r0
7: (b7) r0 = 64
8: (0f) r0 += r4
9: (0f) r0 += r2
10: (bf) r5 = r0
11: (07) r0 += 32766
12: (2d) if r0 > r3 goto pc+1
R0=pkt(id=1,off=32766,r=32766,umin_value=64,umax_value=319,var_off=(0x0; 0x1ff)) R1=ctx(id=0,off=0,imm=0) R2=pkt(id=0,off=0,r=0,imm=0) R3=pkt_end(id=0,off=0,imm=0) R4=inv(id=0,umax_value=255,var_off=(0x0; 0xff)) R5=pkt(id=1,off=0,r=32766,umin_value=64,umax_value=319,var_off=(0x0; 0x1ff)) R10=fp0,call_-1
13: (7b) *(u64 *)(r5 +0) = r0
misaligned packet access off 2+(0x0; 0x1ff)+0+0 size 8
...
#221/u valid map access into an array with a variable FAIL
Unexpected error message!
0: (7a) *(u64 *)(r10 -8) = 0
1: (bf) r2 = r10
2: (07) r2 += -8
3: (18) r1 = 0x0
5: (85) call bpf_map_lookup_elem#1
6: (15) if r0 == 0x0 goto pc+5
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
7: (61) r1 = *(u32 *)(r0 +0)
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
8: (35) if r1 >= 0xb goto pc+3
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umax_value=10,var_off=(0x0; 0xf)) R10=fp0,call_-1
9: (67) r1 <<= 2
10: (0f) r0 += r1
11: (7a) *(u64 *)(r0 +0) = 4
misaligned value access off (0x0; 0x3c)+0+0 size 8
#221/p valid map access into an array with a variable FAIL
Failed to load prog 'Permission denied'!
0: (7a) *(u64 *)(r10 -8) = 0
1: (bf) r2 = r10
2: (07) r2 += -8
3: (18) r1 = 0xfff8003fb5788e00
5: (85) call bpf_map_lookup_elem#1
6: (15) if r0 == 0x0 goto pc+5
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
7: (61) r1 = *(u32 *)(r0 +0)
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
8: (35) if r1 >= 0xb goto pc+3
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umax_value=10,var_off=(0x0; 0xf)) R10=fp0,call_-1
9: (67) r1 <<= 2
10: (0f) r0 += r1
11: (7a) *(u64 *)(r0 +0) = 4
misaligned value access off (0x0; 0x3c)+0+0 size 8
...
#222/p valid map access into an array with a signed variable FAIL
Failed to load prog 'Permission denied'!
0: (7a) *(u64 *)(r10 -8) = 0
1: (bf) r2 = r10
2: (07) r2 += -8
3: (18) r1 = 0xfff8003fb5789200
5: (85) call bpf_map_lookup_elem#1
6: (15) if r0 == 0x0 goto pc+9
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
7: (61) r1 = *(u32 *)(r0 +0)
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R10=fp0,call_-1
8: (65) if r1 s> 0xffffffff goto pc+1
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R10=fp0,call_-1
9: (b4) (u32) r1 = (u32) 0
10: (b4) (u32) r2 = (u32) 11
11: (6d) if r2 s> r1 goto pc+1
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv0 R2=inv11 R10=fp0,call_-1
12: (b4) (u32) r1 = (u32) 0
13: (64) (u32) r1 <<= (u32) 2
14: (0f) r0 += r1
15: (7a) *(u64 *)(r0 +0) = 4
R0_w=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1_w=inv0 R2=inv11 R10=fp0,call_-1
16: (95) exit
from 11 to 13: safe
from 8 to 10: R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R10=fp0,call_-1
10: (b4) (u32) r2 = (u32) 11
11: (6d) if r2 s> r1 goto pc+1
R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umin_value=11,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2=inv11 R10=fp0,call_-1
12: (b4) (u32) r1 = (u32) 0
13: safe
from 11 to 13: R0=map_value(id=0,off=0,ks=8,vs=48,imm=0) R1=inv(id=0,umax_value=10,var_off=(0x0; 0xf)) R2=inv11 R10=fp0,call_-1
13: (64) (u32) r1 <<= (u32) 2
14: (0f) r0 += r1
15: (7a) *(u64 *)(r0 +0) = 4
misaligned value access off (0x0; 0x3c)+0+0 size 8
...
Summary: 603 PASSED, 100 FAILED
^ permalink raw reply
* Re: [PATCH net] openvswitch: Fix pop_vlan action for double tagged frames
From: Eric Garver @ 2017-12-19 20:42 UTC (permalink / raw)
To: Jiri Benc; +Cc: netdev, ovs-dev
In-Reply-To: <20171219203929.12c1cb93@redhat.com>
On Tue, Dec 19, 2017 at 08:39:29PM +0100, Jiri Benc wrote:
> On Tue, 19 Dec 2017 13:57:53 -0500, Eric Garver wrote:
> > --- a/net/openvswitch/flow.c
> > +++ b/net/openvswitch/flow.c
> > @@ -559,8 +559,9 @@ static int parse_nsh(struct sk_buff *skb, struct sw_flow_key *key)
> > * of a correct length, otherwise the same as skb->network_header.
> > * For other key->eth.type values it is left untouched.
> > *
> > - * - skb->protocol: the type of the data starting at skb->network_header.
> > - * Equals to key->eth.type.
> > + * - skb->protocol: For Ethernet, the ethertype or VLAN TPID.
> > + * For non-Ethernet, the type of the data starting at skb->network_header
> > + * (also equal to key->eth.type).
> > */
> > static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > {
> > @@ -579,6 +580,7 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > return -EINVAL;
> >
> > skb_reset_network_header(skb);
> > + key->eth.type = skb->protocol;
> > } else {
> > eth = eth_hdr(skb);
> > ether_addr_copy(key->eth.src, eth->h_source);
> > @@ -592,15 +594,14 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > if (unlikely(parse_vlan(skb, key)))
> > return -ENOMEM;
> >
> > - skb->protocol = parse_ethertype(skb);
> > - if (unlikely(skb->protocol == htons(0)))
> > + key->eth.type = parse_ethertype(skb);
> > + if (unlikely(key->eth.type == htons(0)))
> > return -ENOMEM;
> >
> > skb_reset_network_header(skb);
> > __skb_push(skb, skb->data - skb_mac_header(skb));
> > }
> > skb_reset_mac_len(skb);
> > - key->eth.type = skb->protocol;
> >
> > /* Network layer. */
> > if (key->eth.type == htons(ETH_P_IP)) {
>
> Unfortunately, this does not work. key_extract must set skb->protocol
> even for Ethernet frames that come from a mixed L2/L3 tunnel. Such
> packets will have key->mac_proto set to MAC_PROTO_ETHERNET and
> skb->protocol set to ETH_P_TEB (see key_extract_mac_proto). In
> key_extract, skb->protocol has to be correctly set to the dissected
> value.
AFAICS, it's always overridden to ETH_P_TEB on output by
ovs_vport_send() and that's the sole reason it works today.
For dissecting, the L2 case is currently setting skb->protocol to the
real ethertype (e.g. 0x800) not ETH_P_TEB. For RX from tunnel case it'll
indeed be ETH_P_TEB.
>
> Which means that we have to check for the existence of inner vlan tag
> (by checking key->eth.cvlan.tci or, perhaps better, by returning it
> from parse_vlan) and set skb->protocol accordingly.
>
> Jiri
^ permalink raw reply
* Re: [PATCH net] openvswitch: Fix pop_vlan action for double tagged frames
From: Eric Garver @ 2017-12-19 20:42 UTC (permalink / raw)
To: Jiri Benc; +Cc: netdev, ovs-dev
In-Reply-To: <20171219203929.12c1cb93@redhat.com>
On Tue, Dec 19, 2017 at 08:39:29PM +0100, Jiri Benc wrote:
> On Tue, 19 Dec 2017 13:57:53 -0500, Eric Garver wrote:
> > --- a/net/openvswitch/flow.c
> > +++ b/net/openvswitch/flow.c
> > @@ -559,8 +559,9 @@ static int parse_nsh(struct sk_buff *skb, struct sw_flow_key *key)
> > * of a correct length, otherwise the same as skb->network_header.
> > * For other key->eth.type values it is left untouched.
> > *
> > - * - skb->protocol: the type of the data starting at skb->network_header.
> > - * Equals to key->eth.type.
> > + * - skb->protocol: For Ethernet, the ethertype or VLAN TPID.
> > + * For non-Ethernet, the type of the data starting at skb->network_header
> > + * (also equal to key->eth.type).
> > */
> > static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > {
> > @@ -579,6 +580,7 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > return -EINVAL;
> >
> > skb_reset_network_header(skb);
> > + key->eth.type = skb->protocol;
> > } else {
> > eth = eth_hdr(skb);
> > ether_addr_copy(key->eth.src, eth->h_source);
> > @@ -592,15 +594,14 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> > if (unlikely(parse_vlan(skb, key)))
> > return -ENOMEM;
> >
> > - skb->protocol = parse_ethertype(skb);
> > - if (unlikely(skb->protocol == htons(0)))
> > + key->eth.type = parse_ethertype(skb);
> > + if (unlikely(key->eth.type == htons(0)))
> > return -ENOMEM;
> >
> > skb_reset_network_header(skb);
> > __skb_push(skb, skb->data - skb_mac_header(skb));
> > }
> > skb_reset_mac_len(skb);
> > - key->eth.type = skb->protocol;
> >
> > /* Network layer. */
> > if (key->eth.type == htons(ETH_P_IP)) {
>
> Unfortunately, this does not work. key_extract must set skb->protocol
> even for Ethernet frames that come from a mixed L2/L3 tunnel. Such
> packets will have key->mac_proto set to MAC_PROTO_ETHERNET and
> skb->protocol set to ETH_P_TEB (see key_extract_mac_proto). In
> key_extract, skb->protocol has to be correctly set to the dissected
> value.
AFAICS, it's always overridden to ETH_P_TEB on output by
ovs_vport_send() and that's the sole reason it works today.
For dissecting, the L2 case is currently setting skb->protocol to the
real ethertype (e.g. 0x800) not ETH_P_TEB. For RX from tunnel case it'll
indeed be ETH_P_TEB.
>
> Which means that we have to check for the existence of inner vlan tag
> (by checking key->eth.cvlan.tci or, perhaps better, by returning it
> from parse_vlan) and set skb->protocol accordingly.
>
> Jiri
^ permalink raw reply
* Re: [PATCH net-next 4/4] sfc: expose CTPIO stats on NICs that support them
From: David Miller @ 2017-12-19 20:42 UTC (permalink / raw)
To: jakub.kicinski; +Cc: ecree, linux-net-drivers, netdev
In-Reply-To: <20171219122019.7b3ac5d8@cakuba.netronome.com>
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Tue, 19 Dec 2017 12:20:36 -0800
> On Tue, 19 Dec 2017 17:06:36 +0000, Edward Cree wrote:
>> From: Bert Kenward <bkenward@solarflare.com>
>>
>> While the Linux driver doesn't use CTPIO ('cut-through programmed I/O'),
>> other drivers on the same port might, so if we're responsible for
>> reporting per-port stats we need to include the CTPIO stats.
>>
>> Signed-off-by: Edward Cree <ecree@solarflare.com>
>
> FWIW this is missing a sign-off from Bert.
Edward, please fix this and resubmit.
^ permalink raw reply
* Re: [RFC] hv_netvsc: automatically name slave VF network device
From: Stephen Hemminger @ 2017-12-19 20:44 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, Stephen Hemminger
In-Reply-To: <20171219123234.683f9b8d@cakuba.netronome.com>
On Tue, 19 Dec 2017 12:32:34 -0800
Jakub Kicinski <kubakici@wp.pl> wrote:
> On Tue, 19 Dec 2017 11:35:37 -0800, Stephen Hemminger wrote:
> > Rename the VF device to ethX_vf based on the ethX as the
> > synthetic device. This eliminates the need for delay on setup,
> > and the PCI (udev based) naming is not reproducible on Hyper-V
> > anyway. The name of the VF does not matter since all control
> > operations take place the primary device. It does make the
> > user experience better to associate the names.
> >
> > Based on feedback from all.systems.go talk.
> > The downside is that it requires exporting a symbol from netdev
> > core which makes it harder to backport.
> >
> > Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
>
> Why do you have to name the devices in the kernel space in the first
> place? :/ Why don't upstream the correct change to biosdevname like
> hardware vendors do?
biosdevname is dead, gone and wouldn't work on Azure (it dumpster dives in /dev/mem).
I assume you mean the modern application is udev, and it works but the name is meaningless
because it based of synthetic PCI information. The PCI host adapter is simulated
for pass through devices. Names like enp12s0.
Since every passthrough VF device on Hyper-V/Azure has a matching synthetic
network device with same mac address. It is best to have the relationship
shown in the name.
>
> Your VF setup is really _not_ special, I don't understand why we are
> OK with ignoring the standard practices. Real enterprise distroes
> are very careful never to break the naming of interfaces and they keep
> the naming policy in user space. Playing tricks in the kernel has every
> chance of breaking existing user setups.
Actually, Systemd folks said "naming policy is in userspace only because
kernel can't get it right". Also there is no uniformity in userspace
there are at least 5 systems trying to do network setup. And most of
them depend on eth0 (yes still). Fixing userspace is impossible.
^ permalink raw reply
* Re: [PATCH bpf]: Fix tools and testing build.
From: Daniel Borkmann @ 2017-12-19 20:44 UTC (permalink / raw)
To: David Miller; +Cc: netdev
In-Reply-To: <20171219.152203.371422797348974116.davem@davemloft.net>
On 12/19/2017 09:22 PM, David Miller wrote:
>
> I'm getting various build failures on sparc64. The key is
> usually that the userland tools get built 32-bit.
>
> 1) clock_gettime() is in librt, so that must be added to the link
> libraries.
>
> 2) "sizeof(x)" must be printed with "%Z" printf prefix.
>
> Signed-off-by: David S. Miller <davem@davemloft.net>
Applied to bpf tree, thanks David!
^ permalink raw reply
* Re: [net-next: PATCH 0/8] Armada 7k/8k PP2 ACPI support
From: Andrew Lunn @ 2017-12-19 20:46 UTC (permalink / raw)
To: Marcin Wojtas
Cc: David Miller, linux-kernel, linux-arm-kernel, netdev,
Russell King - ARM Linux, Rafael J. Wysocki, Florian Fainelli,
Antoine Ténart, Thomas Petazzoni, Gregory Clément,
Ezequiel Garcia, nadavh, Neta Zur Hershkovits, Ard Biesheuvel,
Grzegorz Jaszczyk, Tomasz Nowicki
In-Reply-To: <CAPv3WKddg82waiapD8JzH0pCsWVTE2iT0LFz7PCYrrDDpXqmbw@mail.gmail.com>
> Of course! v2 will not have such problem, I've been waiting however
> for the feedback about the ACPI representation. Anyway, I'm strongly
> leaning towards using _ADR/_CID objects in PHY's nodes for ACPI, so
> maybe I'll just issue the v2 in order to push the discussion a bit
> forward.
Hi Marcin
I know ~0 about ACPI. But what seems to be missing for me is
documentation. You are defining a ABI here, which all future MDIO
busses, PHYs, and to some extent Ethernet switches need to follow. So
i would expect this to be documented somewhere.
How does documentation work in the ACPI world?
Andrew
^ permalink raw reply
* Re: thunderx sgmii interface hang
From: Andrew Lunn @ 2017-12-19 20:52 UTC (permalink / raw)
To: Tim Harvey; +Cc: Sunil Goutham, netdev
In-Reply-To: <CAJ+vNU0yYkoAGZ6sU9VGfA6A04KU4s3rzZSb=1uxsqwKHioePQ@mail.gmail.com>
On Mon, Dec 18, 2017 at 01:53:47PM -0800, Tim Harvey wrote:
> On Wed, Dec 13, 2017 at 11:43 AM, Andrew Lunn <andrew@lunn.ch> wrote:
> >> The nic appears to work fine (pings, TCP etc) up until a performance
> >> test is attempted.
> >> When an iperf bandwidth test is attempted the nic ends up in a state
> >> where truncated-ip packets are being sent out (per a tcpdump from
> >> another board):
> >
> > Hi Tim
> >
> > Are pause frames supported? Have you tried turning them off?
> >
> > Can you reproduce the issue with UDP? Or is it TCP only?
> >
>
> Andrew,
>
> Pause frames don't appear to be supported yet and the issue occurs
> when using UDP as well as TCP. I'm not clear what the best way to
> troubleshoot this is.
Hi Tim
Is pause being negotiated? In theory, it should not be. The PHY should
not offer it, if the MAC has not enabled it. But some PHY drivers are
probably broken and offer pause when they should not.
Also, can you trigger the issue using UDP at say 75% the maximum
bandwidth. That should be low enough that the peer never even tries to
use pause.
All this pause stuff is just a stab in the dark. Something else to try
is to turn off various forms off acceleration, ethtook -K, and see if
that makes a difference.
Andrew
^ 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