* Re: [PATCH iproute2] bpf: add btf func and func_proto kind support
From: Stephen Hemminger @ 2019-02-05 23:30 UTC (permalink / raw)
To: Yonghong Song; +Cc: ast, daniel, netdev, kernel-team
In-Reply-To: <20190125004107.3075254-1-yhs@fb.com>
On Thu, 24 Jan 2019 16:41:07 -0800
Yonghong Song <yhs@fb.com> wrote:
> The issue is discovered for bpf selftest test_skb_cgroup.sh.
> Currently we have,
> $ ./test_skb_cgroup_id.sh
> Wait for testing link-local IP to become available ... OK
> Object has unknown BTF type: 13!
> [PASS]
>
> In the above the BTF type 13 refers to BTF kind
> BTF_KIND_FUNC_PROTO.
> This patch added support of BTF_KIND_FUNC_PROTO and
> BTF_KIND_FUNC during type parsing.
> With this patch, I got
> $ ./test_skb_cgroup_id.sh
> Wait for testing link-local IP to become available ... OK
> [PASS]
>
> Signed-off-by: Yonghong Song <yhs@fb.com>
Sure applied.
^ permalink raw reply
* [PATCH net] mISDN: fix a race in dev_expire_timer()
From: Eric Dumazet @ 2019-02-05 23:38 UTC (permalink / raw)
To: David S . Miller; +Cc: netdev, Eric Dumazet, Eric Dumazet, Karsten Keil, syzbot
Since mISDN_close() uses dev->pending to iterate over active
timers, there is a chance that one timer got removed from the
->pending list in dev_expire_timer() but that the thread
has not called yet wake_up_interruptible()
So mISDN_close() could miss this and free dev before
completion of at least one dev_expire_timer()
syzbot was able to catch this race :
BUG: KASAN: use-after-free in register_lock_class+0x140c/0x1bf0 kernel/locking/lockdep.c:827
Write of size 8 at addr ffff88809fc18948 by task syz-executor1/24769
CPU: 1 PID: 24769 Comm: syz-executor1 Not tainted 5.0.0-rc5 #60
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x172/0x1f0 lib/dump_stack.c:113
print_address_description.cold+0x7c/0x20d mm/kasan/report.c:187
kasan_report.cold+0x1b/0x40 mm/kasan/report.c:317
__asan_report_store8_noabort+0x17/0x20 mm/kasan/generic_report.c:140
register_lock_class+0x140c/0x1bf0 kernel/locking/lockdep.c:827
__lock_acquire+0x11f/0x4700 kernel/locking/lockdep.c:3224
lock_acquire+0x16f/0x3f0 kernel/locking/lockdep.c:3841
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x95/0xcd kernel/locking/spinlock.c:152
__wake_up_common_lock+0xc7/0x190 kernel/sched/wait.c:120
__wake_up+0xe/0x10 kernel/sched/wait.c:145
dev_expire_timer+0xe4/0x3b0 drivers/isdn/mISDN/timerdev.c:174
call_timer_fn+0x190/0x720 kernel/time/timer.c:1325
protocol 88fb is buggy, dev hsr_slave_0
protocol 88fb is buggy, dev hsr_slave_1
expire_timers kernel/time/timer.c:1362 [inline]
__run_timers kernel/time/timer.c:1681 [inline]
__run_timers kernel/time/timer.c:1649 [inline]
run_timer_softirq+0x652/0x1700 kernel/time/timer.c:1694
__do_softirq+0x266/0x95a kernel/softirq.c:292
invoke_softirq kernel/softirq.c:373 [inline]
irq_exit+0x180/0x1d0 kernel/softirq.c:413
exiting_irq arch/x86/include/asm/apic.h:536 [inline]
smp_apic_timer_interrupt+0x14a/0x570 arch/x86/kernel/apic/apic.c:1062
apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:807
</IRQ>
RIP: 0010:__sanitizer_cov_trace_pc+0x26/0x50 kernel/kcov.c:101
Code: 90 90 90 90 55 48 89 e5 48 8b 75 08 65 48 8b 04 25 40 ee 01 00 65 8b 15 98 12 92 7e 81 e2 00 01 1f 00 75 2b 8b 90 d8 12 00 00 <83> fa 02 75 20 48 8b 88 e0 12 00 00 8b 80 dc 12 00 00 48 8b 11 48
RSP: 0018:ffff8880589b7a60 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff13
RAX: ffff888087ce25c0 RBX: 0000000000000001 RCX: ffffffff818f8ca3
RDX: 0000000000000000 RSI: ffffffff818f8b48 RDI: 0000000000000001
RBP: ffff8880589b7a60 R08: ffff888087ce25c0 R09: ffffed1015d25bd0
R10: ffffed1015d25bcf R11: ffff8880ae92de7b R12: ffffea0001ae4680
R13: ffffea0001ae4688 R14: 0000000000000000 R15: ffffea0001b41648
PageIdle include/linux/page-flags.h:398 [inline]
page_is_idle include/linux/page_idle.h:29 [inline]
mark_page_accessed+0x618/0x1140 mm/swap.c:398
touch_buffer fs/buffer.c:59 [inline]
__find_get_block+0x312/0xcc0 fs/buffer.c:1298
sb_find_get_block include/linux/buffer_head.h:338 [inline]
recently_deleted fs/ext4/ialloc.c:682 [inline]
find_inode_bit.isra.0+0x202/0x510 fs/ext4/ialloc.c:722
__ext4_new_inode+0x14ad/0x52c0 fs/ext4/ialloc.c:914
ext4_symlink+0x3f8/0xbe0 fs/ext4/namei.c:3096
vfs_symlink fs/namei.c:4126 [inline]
vfs_symlink+0x378/0x5d0 fs/namei.c:4112
do_symlinkat+0x22b/0x290 fs/namei.c:4153
__do_sys_symlink fs/namei.c:4172 [inline]
__se_sys_symlink fs/namei.c:4170 [inline]
__x64_sys_symlink+0x59/0x80 fs/namei.c:4170
do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x457b67
Code: 0f 1f 00 b8 5c 00 00 00 0f 05 48 3d 01 f0 ff ff 0f 83 6d bb fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 b8 58 00 00 00 0f 05 <48> 3d 01 f0 ff ff 0f 83 4d bb fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007fff045ce0f8 EFLAGS: 00000202 ORIG_RAX: 0000000000000058
RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 0000000000457b67
RDX: 00007fff045ce173 RSI: 00000000004bd63f RDI: 00007fff045ce160
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000013
R10: 0000000000000075 R11: 0000000000000202 R12: 0000000000000000
R13: 0000000000000001 R14: 000000000000029b R15: 0000000000000001
Allocated by task 24763:
save_stack+0x45/0xd0 mm/kasan/common.c:73
set_track mm/kasan/common.c:85 [inline]
__kasan_kmalloc mm/kasan/common.c:496 [inline]
__kasan_kmalloc.constprop.0+0xcf/0xe0 mm/kasan/common.c:469
kasan_kmalloc+0x9/0x10 mm/kasan/common.c:504
kmem_cache_alloc_trace+0x151/0x760 mm/slab.c:3609
kmalloc include/linux/slab.h:545 [inline]
mISDN_open+0x9a/0x270 drivers/isdn/mISDN/timerdev.c:59
misc_open+0x398/0x4c0 drivers/char/misc.c:141
chrdev_open+0x247/0x6b0 fs/char_dev.c:417
do_dentry_open+0x47d/0x1130 fs/open.c:771
vfs_open+0xa0/0xd0 fs/open.c:880
do_last fs/namei.c:3418 [inline]
path_openat+0x10d7/0x4690 fs/namei.c:3534
do_filp_open+0x1a1/0x280 fs/namei.c:3564
do_sys_open+0x3fe/0x5d0 fs/open.c:1063
__do_sys_openat fs/open.c:1090 [inline]
__se_sys_openat fs/open.c:1084 [inline]
__x64_sys_openat+0x9d/0x100 fs/open.c:1084
do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
entry_SYSCALL_64_after_hwframe+0x49/0xbe
Freed by task 24762:
save_stack+0x45/0xd0 mm/kasan/common.c:73
set_track mm/kasan/common.c:85 [inline]
__kasan_slab_free+0x102/0x150 mm/kasan/common.c:458
kasan_slab_free+0xe/0x10 mm/kasan/common.c:466
__cache_free mm/slab.c:3487 [inline]
kfree+0xcf/0x230 mm/slab.c:3806
mISDN_close+0x2a1/0x390 drivers/isdn/mISDN/timerdev.c:97
__fput+0x2df/0x8d0 fs/file_table.c:278
____fput+0x16/0x20 fs/file_table.c:309
task_work_run+0x14a/0x1c0 kernel/task_work.c:113
tracehook_notify_resume include/linux/tracehook.h:188 [inline]
exit_to_usermode_loop+0x273/0x2c0 arch/x86/entry/common.c:166
prepare_exit_to_usermode arch/x86/entry/common.c:197 [inline]
syscall_return_slowpath arch/x86/entry/common.c:268 [inline]
do_syscall_64+0x52d/0x610 arch/x86/entry/common.c:293
entry_SYSCALL_64_after_hwframe+0x49/0xbe
The buggy address belongs to the object at ffff88809fc18900
which belongs to the cache kmalloc-192 of size 192
The buggy address is located 72 bytes inside of
192-byte region [ffff88809fc18900, ffff88809fc189c0)
The buggy address belongs to the page:
page:ffffea00027f0600 count:1 mapcount:0 mapping:ffff88812c3f0040 index:0xffff88809fc18000
flags: 0x1fffc0000000200(slab)
raw: 01fffc0000000200 ffffea000269f648 ffffea00029f7408 ffff88812c3f0040
raw: ffff88809fc18000 ffff88809fc18000 000000010000000b 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff88809fc18800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ffff88809fc18880: 00 fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff88809fc18900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff88809fc18980: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
ffff88809fc18a00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Reported-by: syzbot <syzkaller@googlegroups.com>
---
drivers/isdn/mISDN/timerdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/isdn/mISDN/timerdev.c b/drivers/isdn/mISDN/timerdev.c
index 211ed6cffd10e1be817884a7724f3f37687afb93..57897871188706339dc9dfd140a9b55426b478af 100644
--- a/drivers/isdn/mISDN/timerdev.c
+++ b/drivers/isdn/mISDN/timerdev.c
@@ -170,8 +170,8 @@ dev_expire_timer(struct timer_list *t)
spin_lock_irqsave(&timer->dev->lock, flags);
if (timer->id >= 0)
list_move_tail(&timer->list, &timer->dev->expired);
- spin_unlock_irqrestore(&timer->dev->lock, flags);
wake_up_interruptible(&timer->dev->wait);
+ spin_unlock_irqrestore(&timer->dev->lock, flags);
}
static int
--
2.20.1.611.gfbb209baf1-goog
^ permalink raw reply related
* [PATCH net-next v3 06/12] mscc: ocelot: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
Ocelot only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID as a valid
switchdev attribute getter, convert it to use ndo_get_port_parent_id()
and get rid of the switchdev_ops::switchdev_port_attr_get altogether.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/mscc/ocelot.c | 33 ++++++++++++------------------
1 file changed, 13 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index c6a575eb0ff5..195306d05bcd 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -916,6 +916,18 @@ static int ocelot_set_features(struct net_device *dev,
return 0;
}
+static int ocelot_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
+{
+ struct ocelot_port *ocelot_port = netdev_priv(dev);
+ struct ocelot *ocelot = ocelot_port->ocelot;
+
+ ppid->id_len = sizeof(ocelot->base_mac);
+ memcpy(&ppid->id, &ocelot->base_mac, ppid->id_len);
+
+ return 0;
+}
+
static const struct net_device_ops ocelot_port_netdev_ops = {
.ndo_open = ocelot_port_open,
.ndo_stop = ocelot_port_stop,
@@ -930,6 +942,7 @@ static const struct net_device_ops ocelot_port_netdev_ops = {
.ndo_vlan_rx_add_vid = ocelot_vlan_rx_add_vid,
.ndo_vlan_rx_kill_vid = ocelot_vlan_rx_kill_vid,
.ndo_set_features = ocelot_set_features,
+ .ndo_get_port_parent_id = ocelot_get_port_parent_id,
};
static void ocelot_get_strings(struct net_device *netdev, u32 sset, u8 *data)
@@ -1013,25 +1026,6 @@ static const struct ethtool_ops ocelot_ethtool_ops = {
.set_link_ksettings = phy_ethtool_set_link_ksettings,
};
-static int ocelot_port_attr_get(struct net_device *dev,
- struct switchdev_attr *attr)
-{
- struct ocelot_port *ocelot_port = netdev_priv(dev);
- struct ocelot *ocelot = ocelot_port->ocelot;
-
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(ocelot->base_mac);
- memcpy(&attr->u.ppid.id, &ocelot->base_mac,
- attr->u.ppid.id_len);
- break;
- default:
- return -EOPNOTSUPP;
- }
-
- return 0;
-}
-
static int ocelot_port_attr_stp_state_set(struct ocelot_port *ocelot_port,
struct switchdev_trans *trans,
u8 state)
@@ -1331,7 +1325,6 @@ static int ocelot_port_obj_del(struct net_device *dev,
}
static const struct switchdev_ops ocelot_port_switchdev_ops = {
- .switchdev_port_attr_get = ocelot_port_attr_get,
.switchdev_port_attr_set = ocelot_port_attr_set,
};
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 12/12] net: Get rid of SWITCHDEV_ATTR_ID_PORT_PARENT_ID
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
Now that we have a dedicated NDO for getting a port's parent ID, get rid
of SWITCHDEV_ATTR_ID_PORT_PARENT_ID and convert all callers to use the
NDO exclusively. This is a preliminary change to getting rid of
switchdev_ops eventually.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
include/net/switchdev.h | 11 -----------
net/bridge/br_switchdev.c | 16 ++++------------
net/core/net-sysfs.c | 19 ++++---------------
net/core/rtnetlink.c | 18 ++++--------------
net/ipv4/ipmr.c | 16 ++++------------
net/switchdev/switchdev.c | 20 --------------------
6 files changed, 16 insertions(+), 84 deletions(-)
diff --git a/include/net/switchdev.h b/include/net/switchdev.h
index 63843ae5dc81..5e87b54c5dc5 100644
--- a/include/net/switchdev.h
+++ b/include/net/switchdev.h
@@ -43,7 +43,6 @@ static inline bool switchdev_trans_ph_commit(struct switchdev_trans *trans)
enum switchdev_attr_id {
SWITCHDEV_ATTR_ID_UNDEFINED,
- SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
SWITCHDEV_ATTR_ID_PORT_STP_STATE,
SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS,
SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS_SUPPORT,
@@ -61,7 +60,6 @@ struct switchdev_attr {
void *complete_priv;
void (*complete)(struct net_device *dev, int err, void *priv);
union {
- struct netdev_phys_item_id ppid; /* PORT_PARENT_ID */
u8 stp_state; /* PORT_STP_STATE */
unsigned long brport_flags; /* PORT_BRIDGE_FLAGS */
unsigned long brport_flags_support; /* PORT_BRIDGE_FLAGS_SUPPORT */
@@ -208,9 +206,6 @@ void switchdev_port_fwd_mark_set(struct net_device *dev,
struct net_device *group_dev,
bool joining);
-bool switchdev_port_same_parent_id(struct net_device *a,
- struct net_device *b);
-
int switchdev_handle_port_obj_add(struct net_device *dev,
struct switchdev_notifier_port_obj_info *port_obj_info,
bool (*check_cb)(const struct net_device *dev),
@@ -295,12 +290,6 @@ call_switchdev_blocking_notifiers(unsigned long val,
return NOTIFY_DONE;
}
-static inline bool switchdev_port_same_parent_id(struct net_device *a,
- struct net_device *b)
-{
- return false;
-}
-
static inline int
switchdev_handle_port_obj_add(struct net_device *dev,
struct switchdev_notifier_port_obj_info *port_obj_info,
diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c
index 06b0ae44585f..c86d0f4a6aa8 100644
--- a/net/bridge/br_switchdev.c
+++ b/net/bridge/br_switchdev.c
@@ -14,8 +14,7 @@ static int br_switchdev_mark_get(struct net_bridge *br, struct net_device *dev)
/* dev is yet to be added to the port list. */
list_for_each_entry(p, &br->port_list, list) {
- if (netdev_port_same_parent_id(dev, p->dev) ||
- switchdev_port_same_parent_id(dev, p->dev))
+ if (netdev_port_same_parent_id(dev, p->dev))
return p->offload_fwd_mark;
}
@@ -24,19 +23,12 @@ static int br_switchdev_mark_get(struct net_bridge *br, struct net_device *dev)
int nbp_switchdev_mark_set(struct net_bridge_port *p)
{
- const struct net_device_ops *ops = p->dev->netdev_ops;
- struct switchdev_attr attr = {
- .orig_dev = p->dev,
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- };
- int err;
+ struct netdev_phys_item_id ppid = { };
+ int err = -EOPNOTSUPP;
ASSERT_RTNL();
- if (ops->ndo_get_port_parent_id)
- err = dev_get_port_parent_id(p->dev, &attr.u.ppid, true);
- else
- err = switchdev_port_attr_get(p->dev, &attr);
+ err = dev_get_port_parent_id(p->dev, &ppid, true);
if (err) {
if (err == -EOPNOTSUPP)
return 0;
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 4eace9f1dcf9..7c5061123ead 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -12,7 +12,6 @@
#include <linux/capability.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
-#include <net/switchdev.h>
#include <linux/if_arp.h>
#include <linux/slab.h>
#include <linux/sched/signal.h>
@@ -495,27 +494,17 @@ static ssize_t phys_switch_id_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct net_device *netdev = to_net_dev(dev);
- const struct net_device_ops *ops = netdev->netdev_ops;
ssize_t ret = -EINVAL;
if (!rtnl_trylock())
return restart_syscall();
if (dev_isalive(netdev)) {
- struct switchdev_attr attr = {
- .orig_dev = netdev,
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- .flags = SWITCHDEV_F_NO_RECURSE,
- };
-
- if (ops->ndo_get_port_parent_id)
- ret = dev_get_port_parent_id(netdev, &attr.u.ppid,
- false);
- else
- ret = switchdev_port_attr_get(netdev, &attr);
+ struct netdev_phys_item_id ppid = { };
+
+ ret = dev_get_port_parent_id(netdev, &ppid, false);
if (!ret)
- ret = sprintf(buf, "%*phN\n", attr.u.ppid.id_len,
- attr.u.ppid.id);
+ ret = sprintf(buf, "%*phN\n", ppid.id_len, ppid.id);
}
rtnl_unlock();
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 90dd02c1f561..7e1d1e9ad67d 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -46,7 +46,6 @@
#include <linux/inet.h>
#include <linux/netdevice.h>
-#include <net/switchdev.h>
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
@@ -1146,26 +1145,17 @@ static int rtnl_phys_port_name_fill(struct sk_buff *skb, struct net_device *dev)
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
- const struct net_device_ops *ops = dev->netdev_ops;
- int err;
- struct switchdev_attr attr = {
- .orig_dev = dev,
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- .flags = SWITCHDEV_F_NO_RECURSE,
- };
+ struct netdev_phys_item_id ppid = { };
+ int err = -EOPNOTSUPP;
- if (ops->ndo_get_port_parent_id)
- err = dev_get_port_parent_id(dev, &attr.u.ppid, false);
- else
- err = switchdev_port_attr_get(dev, &attr);
+ err = dev_get_port_parent_id(dev, &ppid, false);
if (err) {
if (err == -EOPNOTSUPP)
return 0;
return err;
}
- if (nla_put(skb, IFLA_PHYS_SWITCH_ID, attr.u.ppid.id_len,
- attr.u.ppid.id))
+ if (nla_put(skb, IFLA_PHYS_SWITCH_ID, ppid.id_len, ppid.id))
return -EMSGSIZE;
return 0;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index c71bcc42d66d..83bf5760066f 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -67,7 +67,6 @@
#include <net/fib_rules.h>
#include <linux/netconf.h>
#include <net/nexthop.h>
-#include <net/switchdev.h>
#include <linux/nospec.h>
@@ -837,11 +836,9 @@ static void ipmr_update_thresholds(struct mr_table *mrt, struct mr_mfc *cache,
static int vif_add(struct net *net, struct mr_table *mrt,
struct vifctl *vifc, int mrtsock)
{
+ struct netdev_phys_item_id ppid = { };
const struct net_device_ops *ops;
int vifi = vifc->vifc_vifi;
- struct switchdev_attr attr = {
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- };
struct vif_device *v = &mrt->vif_table[vifi];
struct net_device *dev;
struct in_device *in_dev;
@@ -920,15 +917,10 @@ static int vif_add(struct net *net, struct mr_table *mrt,
vifc->vifc_flags | (!mrtsock ? VIFF_STATIC : 0),
(VIFF_TUNNEL | VIFF_REGISTER));
- attr.orig_dev = dev;
ops = dev->netdev_ops;
- if (ops->ndo_get_port_parent_id &&
- !dev_get_port_parent_id(dev, &attr.u.ppid, true)) {
- memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
- v->dev_parent_id.id_len = attr.u.ppid.id_len;
- } else if (!switchdev_port_attr_get(dev, &attr)) {
- memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
- v->dev_parent_id.id_len = attr.u.ppid.id_len;
+ if (!dev_get_port_parent_id(dev, &ppid, true)) {
+ memcpy(v->dev_parent_id.id, ppid.id, ppid.id_len);
+ v->dev_parent_id.id_len = ppid.id_len;
} else {
v->dev_parent_id.id_len = 0;
}
diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c
index cd78253de31d..7e1357db33d7 100644
--- a/net/switchdev/switchdev.c
+++ b/net/switchdev/switchdev.c
@@ -592,26 +592,6 @@ int call_switchdev_blocking_notifiers(unsigned long val, struct net_device *dev,
}
EXPORT_SYMBOL_GPL(call_switchdev_blocking_notifiers);
-bool switchdev_port_same_parent_id(struct net_device *a,
- struct net_device *b)
-{
- struct switchdev_attr a_attr = {
- .orig_dev = a,
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- };
- struct switchdev_attr b_attr = {
- .orig_dev = b,
- .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
- };
-
- if (switchdev_port_attr_get(a, &a_attr) ||
- switchdev_port_attr_get(b, &b_attr))
- return false;
-
- return netdev_phys_item_id_same(&a_attr.u.ppid, &b_attr.u.ppid);
-}
-EXPORT_SYMBOL_GPL(switchdev_port_same_parent_id);
-
static int __switchdev_handle_port_obj_add(struct net_device *dev,
struct switchdev_notifier_port_obj_info *port_obj_info,
bool (*check_cb)(const struct net_device *dev),
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 10/12] staging: fsl-dpaa2: ethsw: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
ethsw implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/drivers/staging/fsl-dpaa2/ethsw/ethsw.c b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
index daabaceeea52..e559f4c25cf7 100644
--- a/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
+++ b/drivers/staging/fsl-dpaa2/ethsw/ethsw.c
@@ -505,6 +505,17 @@ static netdev_tx_t port_dropframe(struct sk_buff *skb,
return NETDEV_TX_OK;
}
+static int swdev_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
+{
+ struct ethsw_port_priv *port_priv = netdev_priv(dev);
+
+ ppid->id_len = 1;
+ ppid->id[0] = port_priv->ethsw_data->dev_id;
+
+ return 0;
+}
+
static const struct net_device_ops ethsw_port_ops = {
.ndo_open = port_open,
.ndo_stop = port_stop,
@@ -515,6 +526,7 @@ static const struct net_device_ops ethsw_port_ops = {
.ndo_get_offload_stats = port_get_offload_stats,
.ndo_start_xmit = port_dropframe,
+ .ndo_get_port_parent_id = swdev_get_port_parent_id,
};
static void ethsw_links_state_update(struct ethsw_core *ethsw)
@@ -634,10 +646,6 @@ static int swdev_port_attr_get(struct net_device *netdev,
struct ethsw_port_priv *port_priv = netdev_priv(netdev);
switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = 1;
- attr->u.ppid.id[0] = port_priv->ethsw_data->dev_id;
- break;
case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
attr->u.brport_flags =
(port_priv->ethsw_data->learning ? BR_LEARNING : 0) |
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 11/12] net: dsa: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
DSA implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
net/dsa/slave.c | 18 ++++++++++++------
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 91de3a663226..70395a0ae52e 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -362,18 +362,23 @@ static int dsa_slave_port_obj_del(struct net_device *dev,
return err;
}
-static int dsa_slave_port_attr_get(struct net_device *dev,
- struct switchdev_attr *attr)
+static int dsa_slave_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_switch *ds = dp->ds;
struct dsa_switch_tree *dst = ds->dst;
+ ppid->id_len = sizeof(dst->index);
+ memcpy(&ppid->id, &dst->index, ppid->id_len);
+
+ return 0;
+}
+
+static int dsa_slave_port_attr_get(struct net_device *dev,
+ struct switchdev_attr *attr)
+{
switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(dst->index);
- memcpy(&attr->u.ppid.id, &dst->index, attr->u.ppid.id_len);
- break;
case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS_SUPPORT:
attr->u.brport_flags_support = 0;
break;
@@ -1046,6 +1051,7 @@ static const struct net_device_ops dsa_slave_netdev_ops = {
.ndo_get_phys_port_name = dsa_slave_get_phys_port_name,
.ndo_setup_tc = dsa_slave_setup_tc,
.ndo_get_stats64 = dsa_slave_get_stats64,
+ .ndo_get_port_parent_id = dsa_slave_get_port_parent_id,
};
static const struct switchdev_ops dsa_slave_switchdev_ops = {
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 09/12] netdevsim: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
netdevsim only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/netdevsim/netdev.c | 23 ++++++-----------------
1 file changed, 6 insertions(+), 17 deletions(-)
diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c
index 8d8e2b3f263e..75a50b59cb8f 100644
--- a/drivers/net/netdevsim/netdev.c
+++ b/drivers/net/netdevsim/netdev.c
@@ -22,7 +22,6 @@
#include <net/netlink.h>
#include <net/pkt_cls.h>
#include <net/rtnetlink.h>
-#include <net/switchdev.h>
#include "netdevsim.h"
@@ -148,26 +147,16 @@ static struct device_type nsim_dev_type = {
.release = nsim_dev_release,
};
-static int
-nsim_port_attr_get(struct net_device *dev, struct switchdev_attr *attr)
+static int nsim_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct netdevsim *ns = netdev_priv(dev);
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(ns->sdev->switch_id);
- memcpy(&attr->u.ppid.id, &ns->sdev->switch_id,
- attr->u.ppid.id_len);
- return 0;
- default:
- return -EOPNOTSUPP;
- }
+ ppid->id_len = sizeof(ns->sdev->switch_id);
+ memcpy(&ppid->id, &ns->sdev->switch_id, ppid->id_len);
+ return 0;
}
-static const struct switchdev_ops nsim_switchdev_ops = {
- .switchdev_port_attr_get = nsim_port_attr_get,
-};
-
static int nsim_init(struct net_device *dev)
{
char sdev_ddir_name[10], sdev_link_name[32];
@@ -214,7 +203,6 @@ static int nsim_init(struct net_device *dev)
goto err_bpf_uninit;
SET_NETDEV_DEV(dev, &ns->dev);
- SWITCHDEV_SET_OPS(dev, &nsim_switchdev_ops);
err = nsim_devlink_setup(ns);
if (err)
@@ -493,6 +481,7 @@ static const struct net_device_ops nsim_netdev_ops = {
.ndo_setup_tc = nsim_setup_tc,
.ndo_set_features = nsim_set_features,
.ndo_bpf = nsim_bpf,
+ .ndo_get_port_parent_id = nsim_get_port_parent_id,
};
static void nsim_setup(struct net_device *dev)
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 07/12] nfp: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
NFP only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get().
Since NFP uses switchdev_port_same_parent_id() convert it to use
netdev_port_same_parent_id().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
.../ethernet/netronome/nfp/flower/action.c | 3 +--
.../ethernet/netronome/nfp/nfp_net_common.c | 4 +---
.../net/ethernet/netronome/nfp/nfp_net_repr.c | 4 +---
drivers/net/ethernet/netronome/nfp/nfp_port.c | 23 +++++--------------
drivers/net/ethernet/netronome/nfp/nfp_port.h | 4 +++-
5 files changed, 12 insertions(+), 26 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c
index 8d54b36afee8..7b217d7285c7 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/action.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/action.c
@@ -3,7 +3,6 @@
#include <linux/bitfield.h>
#include <net/pkt_cls.h>
-#include <net/switchdev.h>
#include <net/tc_act/tc_csum.h>
#include <net/tc_act/tc_gact.h>
#include <net/tc_act/tc_mirred.h>
@@ -137,7 +136,7 @@ nfp_fl_output(struct nfp_app *app, struct nfp_fl_output *output,
if (nfp_netdev_is_nfp_repr(in_dev)) {
/* Confirm ingress and egress are on same device. */
- if (!switchdev_port_same_parent_id(in_dev, out_dev))
+ if (!netdev_port_same_parent_id(in_dev, out_dev))
return -EOPNOTSUPP;
}
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 7d2d4241498f..776f6c07701b 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -36,7 +36,6 @@
#include <linux/vmalloc.h>
#include <linux/ktime.h>
-#include <net/switchdev.h>
#include <net/vxlan.h>
#include "nfpcore/nfp_nsp.h"
@@ -3531,6 +3530,7 @@ const struct net_device_ops nfp_net_netdev_ops = {
.ndo_udp_tunnel_add = nfp_net_add_vxlan_port,
.ndo_udp_tunnel_del = nfp_net_del_vxlan_port,
.ndo_bpf = nfp_net_xdp,
+ .ndo_get_port_parent_id = nfp_port_get_port_parent_id,
};
/**
@@ -3815,8 +3815,6 @@ static void nfp_net_netdev_init(struct nfp_net *nn)
netdev->netdev_ops = &nfp_net_netdev_ops;
netdev->watchdog_timeo = msecs_to_jiffies(5 * 1000);
- SWITCHDEV_SET_OPS(netdev, &nfp_port_switchdev_ops);
-
/* MTU range: 68 - hw-specific max */
netdev->min_mtu = ETH_MIN_MTU;
netdev->max_mtu = nn->max_mtu;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
index 69d7aebda09b..62839807e21e 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
@@ -5,7 +5,6 @@
#include <linux/io-64-nonatomic-hi-lo.h>
#include <linux/lockdep.h>
#include <net/dst_metadata.h>
-#include <net/switchdev.h>
#include "nfpcore/nfp_cpp.h"
#include "nfpcore/nfp_nsp.h"
@@ -273,6 +272,7 @@ const struct net_device_ops nfp_repr_netdev_ops = {
.ndo_fix_features = nfp_repr_fix_features,
.ndo_set_features = nfp_port_set_features,
.ndo_set_mac_address = eth_mac_addr,
+ .ndo_get_port_parent_id = nfp_port_get_port_parent_id,
};
void
@@ -336,8 +336,6 @@ int nfp_repr_init(struct nfp_app *app, struct net_device *netdev,
netdev->max_mtu = pf_netdev->max_mtu;
- SWITCHDEV_SET_OPS(netdev, &nfp_port_switchdev_ops);
-
/* Set features the lower device can support with representors */
if (repr_cap & NFP_NET_CFG_CTRL_LIVE_ADDR)
netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_port.c b/drivers/net/ethernet/netronome/nfp/nfp_port.c
index 86bc149ca231..b9aa7eb99067 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_port.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_port.c
@@ -31,34 +31,23 @@ struct nfp_port *nfp_port_from_netdev(struct net_device *netdev)
return NULL;
}
-static int
-nfp_port_attr_get(struct net_device *netdev, struct switchdev_attr *attr)
+int nfp_port_get_port_parent_id(struct net_device *netdev,
+ struct netdev_phys_item_id *ppid)
{
struct nfp_port *port;
+ const u8 *serial;
port = nfp_port_from_netdev(netdev);
if (!port)
return -EOPNOTSUPP;
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID: {
- const u8 *serial;
- /* N.B: attr->u.ppid.id is binary data */
- attr->u.ppid.id_len = nfp_cpp_serial(port->app->cpp, &serial);
- memcpy(&attr->u.ppid.id, serial, attr->u.ppid.id_len);
- break;
- }
- default:
- return -EOPNOTSUPP;
- }
+ /* N.B: attr->u.ppid.id is binary data */
+ ppid->id_len = nfp_cpp_serial(port->app->cpp, &serial);
+ memcpy(&ppid->id, serial, ppid->id_len);
return 0;
}
-const struct switchdev_ops nfp_port_switchdev_ops = {
- .switchdev_port_attr_get = nfp_port_attr_get,
-};
-
int nfp_port_setup_tc(struct net_device *netdev, enum tc_setup_type type,
void *type_data)
{
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_port.h b/drivers/net/ethernet/netronome/nfp/nfp_port.h
index b2479a2a49e5..90ae053f5c07 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_port.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_port.h
@@ -7,6 +7,7 @@
#include <net/devlink.h>
struct net_device;
+struct netdev_phys_item_id;
struct nfp_app;
struct nfp_pf;
struct nfp_port;
@@ -90,7 +91,6 @@ struct nfp_port {
};
extern const struct ethtool_ops nfp_port_ethtool_ops;
-extern const struct switchdev_ops nfp_port_switchdev_ops;
__printf(2, 3) u8 *nfp_pr_et(u8 *data, const char *fmt, ...);
@@ -106,6 +106,8 @@ int
nfp_port_set_features(struct net_device *netdev, netdev_features_t features);
struct nfp_port *nfp_port_from_netdev(struct net_device *netdev);
+int nfp_port_get_port_parent_id(struct net_device *netdev,
+ struct netdev_phys_item_id *ppid);
struct nfp_port *
nfp_port_from_id(struct nfp_pf *pf, enum nfp_port_type type, unsigned int id);
struct nfp_eth_table_port *__nfp_port_get_eth_port(struct nfp_port *port);
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 08/12] rocker: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
mlxsw implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/rocker/rocker_main.c | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 62a205eba9f7..596f44dbfae1 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -2026,6 +2026,18 @@ static void rocker_port_neigh_destroy(struct net_device *dev,
err);
}
+static int rocker_port_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
+{
+ const struct rocker_port *rocker_port = netdev_priv(dev);
+ const struct rocker *rocker = rocker_port->rocker;
+
+ ppid->id_len = sizeof(rocker->hw.id);
+ memcpy(&ppid->id, &rocker->hw.id, ppid->id_len);
+
+ return 0;
+}
+
static const struct net_device_ops rocker_port_netdev_ops = {
.ndo_open = rocker_port_open,
.ndo_stop = rocker_port_stop,
@@ -2035,6 +2047,7 @@ static const struct net_device_ops rocker_port_netdev_ops = {
.ndo_get_phys_port_name = rocker_port_get_phys_port_name,
.ndo_change_proto_down = rocker_port_change_proto_down,
.ndo_neigh_destroy = rocker_port_neigh_destroy,
+ .ndo_get_port_parent_id = rocker_port_get_port_parent_id,
};
/********************
@@ -2049,10 +2062,6 @@ static int rocker_port_attr_get(struct net_device *dev,
int err = 0;
switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(rocker->hw.id);
- memcpy(&attr->u.ppid.id, &rocker->hw.id, attr->u.ppid.id_len);
- break;
case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
err = rocker_world_port_attr_bridge_flags_get(rocker_port,
&attr->u.brport_flags);
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 04/12] net/mlx5e: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
mlx5e only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get().
Since mlx5e makes use of switchdev_port_parent_id() convert it to use
netdev_port_same_parent_id().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
.../ethernet/mellanox/mlx5/core/en/tc_tun.c | 2 +-
.../net/ethernet/mellanox/mlx5/core/en_rep.c | 31 +++++++------------
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 5 ++-
3 files changed, 14 insertions(+), 24 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
index 046948ead152..19dc4a963b90 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
@@ -25,7 +25,7 @@ static int get_route_and_out_devs(struct mlx5e_priv *priv,
/* if the egress device isn't on the same HW e-switch or
* it's a LAG device, use the uplink
*/
- if (!switchdev_port_same_parent_id(priv->netdev, dev) ||
+ if (!netdev_port_same_parent_id(priv->netdev, dev) ||
dst_is_lag_dev) {
*route_dev = uplink_dev;
*out_dev = *route_dev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 5d2e0c2f6624..0b1988b330f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -381,7 +381,8 @@ static const struct ethtool_ops mlx5e_uplink_rep_ethtool_ops = {
.set_pauseparam = mlx5e_uplink_rep_set_pauseparam,
};
-static int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr)
+static int mlx5e_rep_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct mlx5e_priv *priv = netdev_priv(dev);
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
@@ -398,20 +399,14 @@ static int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr)
uplink_priv = netdev_priv(uplink_dev);
}
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = ETH_ALEN;
- if (uplink_upper && mlx5_lag_is_sriov(uplink_priv->mdev)) {
- ether_addr_copy(attr->u.ppid.id, uplink_upper->dev_addr);
- } else {
- struct mlx5e_rep_priv *rpriv = priv->ppriv;
- struct mlx5_eswitch_rep *rep = rpriv->rep;
+ ppid->id_len = ETH_ALEN;
+ if (uplink_upper && mlx5_lag_is_sriov(uplink_priv->mdev)) {
+ ether_addr_copy(ppid->id, uplink_upper->dev_addr);
+ } else {
+ struct mlx5e_rep_priv *rpriv = priv->ppriv;
+ struct mlx5_eswitch_rep *rep = rpriv->rep;
- ether_addr_copy(attr->u.ppid.id, rep->hw_id);
- }
- break;
- default:
- return -EOPNOTSUPP;
+ ether_addr_copy(ppid->id, rep->hw_id);
}
return 0;
@@ -1284,10 +1279,6 @@ static int mlx5e_uplink_rep_set_vf_vlan(struct net_device *dev, int vf, u16 vlan
return 0;
}
-static const struct switchdev_ops mlx5e_rep_switchdev_ops = {
- .switchdev_port_attr_get = mlx5e_attr_get,
-};
-
static const struct net_device_ops mlx5e_netdev_ops_vf_rep = {
.ndo_open = mlx5e_vf_rep_open,
.ndo_stop = mlx5e_vf_rep_close,
@@ -1298,6 +1289,7 @@ static const struct net_device_ops mlx5e_netdev_ops_vf_rep = {
.ndo_has_offload_stats = mlx5e_rep_has_offload_stats,
.ndo_get_offload_stats = mlx5e_rep_get_offload_stats,
.ndo_change_mtu = mlx5e_vf_rep_change_mtu,
+ .ndo_get_port_parent_id = mlx5e_rep_get_port_parent_id,
};
static const struct net_device_ops mlx5e_netdev_ops_uplink_rep = {
@@ -1319,6 +1311,7 @@ static const struct net_device_ops mlx5e_netdev_ops_uplink_rep = {
.ndo_get_vf_config = mlx5e_get_vf_config,
.ndo_get_vf_stats = mlx5e_get_vf_stats,
.ndo_set_vf_vlan = mlx5e_uplink_rep_set_vf_vlan,
+ .ndo_get_port_parent_id = mlx5e_rep_get_port_parent_id,
};
bool mlx5e_eswitch_rep(struct net_device *netdev)
@@ -1393,8 +1386,6 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev)
netdev->watchdog_timeo = 15 * HZ;
- netdev->switchdev_ops = &mlx5e_rep_switchdev_ops;
-
netdev->features |= NETIF_F_HW_TC | NETIF_F_NETNS_LOCAL;
netdev->hw_features |= NETIF_F_HW_TC;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 74159d39dd66..098b28feba0c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -38,7 +38,6 @@
#include <linux/mlx5/fs.h>
#include <linux/mlx5/device.h>
#include <linux/rhashtable.h>
-#include <net/switchdev.h>
#include <net/tc_act/tc_mirred.h>
#include <net/tc_act/tc_vlan.h>
#include <net/tc_act/tc_tunnel_key.h>
@@ -2569,8 +2568,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
MLX5_FLOW_CONTEXT_ACTION_COUNT;
- if (switchdev_port_same_parent_id(priv->netdev,
- out_dev) ||
+ if (netdev_port_same_parent_id(priv->netdev,
+ out_dev) ||
is_merged_eswitch_dev(priv, out_dev)) {
struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
struct net_device *uplink_dev = mlx5_eswitch_uplink_get_proto_dev(esw, REP_ETH);
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 05/12] mlxsw: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
mlxsw implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
.../net/ethernet/mellanox/mlxsw/spectrum.c | 13 +++++++
.../mellanox/mlxsw/spectrum_switchdev.c | 5 ---
.../net/ethernet/mellanox/mlxsw/switchx2.c | 36 +++++++------------
3 files changed, 26 insertions(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index a88169738b4a..8dd808b7f931 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -1700,6 +1700,18 @@ static int mlxsw_sp_set_features(struct net_device *dev,
mlxsw_sp_feature_hw_tc);
}
+static int mlxsw_sp_port_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
+{
+ struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+ struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+
+ ppid->id_len = sizeof(mlxsw_sp->base_mac);
+ memcpy(&ppid->id, &mlxsw_sp->base_mac, ppid->id_len);
+
+ return 0;
+}
+
static const struct net_device_ops mlxsw_sp_port_netdev_ops = {
.ndo_open = mlxsw_sp_port_open,
.ndo_stop = mlxsw_sp_port_stop,
@@ -1715,6 +1727,7 @@ static const struct net_device_ops mlxsw_sp_port_netdev_ops = {
.ndo_vlan_rx_kill_vid = mlxsw_sp_port_kill_vid,
.ndo_get_phys_port_name = mlxsw_sp_port_get_phys_port_name,
.ndo_set_features = mlxsw_sp_set_features,
+ .ndo_get_port_parent_id = mlxsw_sp_port_get_port_parent_id,
};
static void mlxsw_sp_port_get_drvinfo(struct net_device *dev,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index a4a9fe992193..95e37de3e48f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -451,11 +451,6 @@ static int mlxsw_sp_port_attr_get(struct net_device *dev,
struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(mlxsw_sp->base_mac);
- memcpy(&attr->u.ppid.id, &mlxsw_sp->base_mac,
- attr->u.ppid.id_len);
- break;
case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
mlxsw_sp_port_bridge_flags_get(mlxsw_sp->bridge, attr->orig_dev,
&attr->u.brport_flags);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
index 2d4f213e154d..3814ba8af517 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
@@ -390,6 +390,18 @@ static int mlxsw_sx_port_get_phys_port_name(struct net_device *dev, char *name,
name, len);
}
+static int mlxsw_sx_port_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
+{
+ struct mlxsw_sx_port *mlxsw_sx_port = netdev_priv(dev);
+ struct mlxsw_sx *mlxsw_sx = mlxsw_sx_port->mlxsw_sx;
+
+ ppid->id_len = sizeof(mlxsw_sx->hw_id);
+ memcpy(&ppid->id, &mlxsw_sx->hw_id, ppid->id_len);
+
+ return 0;
+}
+
static const struct net_device_ops mlxsw_sx_port_netdev_ops = {
.ndo_open = mlxsw_sx_port_open,
.ndo_stop = mlxsw_sx_port_stop,
@@ -397,6 +409,7 @@ static const struct net_device_ops mlxsw_sx_port_netdev_ops = {
.ndo_change_mtu = mlxsw_sx_port_change_mtu,
.ndo_get_stats64 = mlxsw_sx_port_get_stats64,
.ndo_get_phys_port_name = mlxsw_sx_port_get_phys_port_name,
+ .ndo_get_port_parent_id = mlxsw_sx_port_get_port_parent_id,
};
static void mlxsw_sx_port_get_drvinfo(struct net_device *dev,
@@ -901,28 +914,6 @@ static const struct ethtool_ops mlxsw_sx_port_ethtool_ops = {
.set_link_ksettings = mlxsw_sx_port_set_link_ksettings,
};
-static int mlxsw_sx_port_attr_get(struct net_device *dev,
- struct switchdev_attr *attr)
-{
- struct mlxsw_sx_port *mlxsw_sx_port = netdev_priv(dev);
- struct mlxsw_sx *mlxsw_sx = mlxsw_sx_port->mlxsw_sx;
-
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(mlxsw_sx->hw_id);
- memcpy(&attr->u.ppid.id, &mlxsw_sx->hw_id, attr->u.ppid.id_len);
- break;
- default:
- return -EOPNOTSUPP;
- }
-
- return 0;
-}
-
-static const struct switchdev_ops mlxsw_sx_port_switchdev_ops = {
- .switchdev_port_attr_get = mlxsw_sx_port_attr_get,
-};
-
static int mlxsw_sx_hw_id_get(struct mlxsw_sx *mlxsw_sx)
{
char spad_pl[MLXSW_REG_SPAD_LEN] = {0};
@@ -1034,7 +1025,6 @@ static int __mlxsw_sx_port_eth_create(struct mlxsw_sx *mlxsw_sx, u8 local_port,
dev->netdev_ops = &mlxsw_sx_port_netdev_ops;
dev->ethtool_ops = &mlxsw_sx_port_ethtool_ops;
- dev->switchdev_ops = &mlxsw_sx_port_switchdev_ops;
err = mlxsw_sx_port_dev_addr_get(mlxsw_sx_port);
if (err) {
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 02/12] bnxt: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
BNXT only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get(). The conversion is
straight forward here since the PF and VF code use the same getter.
Since bnxt makes uses of switchdev_port_same_parent_id() convert it to
use netdev_port_same_parent_id().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 28 ++++++-------------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 4 +--
drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c | 3 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c | 12 +++-----
4 files changed, 16 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 6a512871176b..1c2987c3d708 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -9981,8 +9981,11 @@ static int bnxt_get_phys_port_name(struct net_device *dev, char *buf,
return 0;
}
-int bnxt_port_attr_get(struct bnxt *bp, struct switchdev_attr *attr)
+int bnxt_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
+ struct bnxt *bp = netdev_priv(dev);
+
if (bp->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
return -EOPNOTSUPP;
@@ -9990,27 +9993,12 @@ int bnxt_port_attr_get(struct bnxt *bp, struct switchdev_attr *attr)
if (!BNXT_PF(bp))
return -EOPNOTSUPP;
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = sizeof(bp->switch_id);
- memcpy(attr->u.ppid.id, bp->switch_id, attr->u.ppid.id_len);
- break;
- default:
- return -EOPNOTSUPP;
- }
- return 0;
-}
+ ppid->id_len = sizeof(bp->switch_id);
+ memcpy(ppid->id, bp->switch_id, ppid->id_len);
-static int bnxt_swdev_port_attr_get(struct net_device *dev,
- struct switchdev_attr *attr)
-{
- return bnxt_port_attr_get(netdev_priv(dev), attr);
+ return 0;
}
-static const struct switchdev_ops bnxt_switchdev_ops = {
- .switchdev_port_attr_get = bnxt_swdev_port_attr_get
-};
-
static const struct net_device_ops bnxt_netdev_ops = {
.ndo_open = bnxt_open,
.ndo_start_xmit = bnxt_start_xmit,
@@ -10042,6 +10030,7 @@ static const struct net_device_ops bnxt_netdev_ops = {
.ndo_bpf = bnxt_xdp,
.ndo_bridge_getlink = bnxt_bridge_getlink,
.ndo_bridge_setlink = bnxt_bridge_setlink,
+ .ndo_get_port_parent_id = bnxt_get_port_parent_id,
.ndo_get_phys_port_name = bnxt_get_phys_port_name
};
@@ -10400,7 +10389,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
dev->netdev_ops = &bnxt_netdev_ops;
dev->watchdog_timeo = BNXT_TX_TIMEOUT;
dev->ethtool_ops = &bnxt_ethtool_ops;
- SWITCHDEV_SET_OPS(dev, &bnxt_switchdev_ops);
pci_set_drvdata(pdev, dev);
rc = bnxt_alloc_hwrm_resources(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 5c886a700bcc..17554d4be651 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -22,7 +22,6 @@
#include <linux/rhashtable.h>
#include <net/devlink.h>
#include <net/dst_metadata.h>
-#include <net/switchdev.h>
#include <net/xdp.h>
#include <linux/net_dim.h>
@@ -1795,7 +1794,8 @@ int bnxt_check_rings(struct bnxt *bp, int tx, int rx, bool sh, int tcs,
int bnxt_setup_mq_tc(struct net_device *dev, u8 tc);
int bnxt_get_max_rings(struct bnxt *, int *, int *, bool);
int bnxt_restore_pf_fw_resources(struct bnxt *bp);
-int bnxt_port_attr_get(struct bnxt *bp, struct switchdev_attr *attr);
+int bnxt_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid);
void bnxt_dim_work(struct work_struct *work);
int bnxt_hwrm_set_ring_coal(struct bnxt *bp, struct bnxt_napi *bnapi);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
index c683b5e96b1d..170d2fbbb91b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
@@ -12,6 +12,7 @@
#include <linux/if_vlan.h>
#include <net/flow_dissector.h>
#include <net/pkt_cls.h>
+#include <net/switchdev.h>
#include <net/tc_act/tc_gact.h>
#include <net/tc_act/tc_skbedit.h>
#include <net/tc_act/tc_mirred.h>
@@ -45,7 +46,7 @@ static u16 bnxt_flow_get_dst_fid(struct bnxt *pf_bp, struct net_device *dev)
struct bnxt *bp;
/* check if dev belongs to the same switch */
- if (!switchdev_port_same_parent_id(pf_bp->dev, dev)) {
+ if (!netdev_port_same_parent_id(pf_bp->dev, dev)) {
netdev_info(pf_bp->dev, "dev(ifindex=%d) not on same switch",
dev->ifindex);
return BNXT_FID_INVALID;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
index 9a25c05aa571..2bdd2da9aac7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
@@ -237,21 +237,17 @@ static void bnxt_vf_rep_get_drvinfo(struct net_device *dev,
strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
}
-static int bnxt_vf_rep_port_attr_get(struct net_device *dev,
- struct switchdev_attr *attr)
+static int bnxt_vf_rep_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct bnxt_vf_rep *vf_rep = netdev_priv(dev);
/* as only PORT_PARENT_ID is supported currently use common code
* between PF and VF-rep for now.
*/
- return bnxt_port_attr_get(vf_rep->bp, attr);
+ return bnxt_get_port_parent_id(vf_rep->bp->dev, ppid);
}
-static const struct switchdev_ops bnxt_vf_rep_switchdev_ops = {
- .switchdev_port_attr_get = bnxt_vf_rep_port_attr_get
-};
-
static const struct ethtool_ops bnxt_vf_rep_ethtool_ops = {
.get_drvinfo = bnxt_vf_rep_get_drvinfo
};
@@ -262,6 +258,7 @@ static const struct net_device_ops bnxt_vf_rep_netdev_ops = {
.ndo_start_xmit = bnxt_vf_rep_xmit,
.ndo_get_stats64 = bnxt_vf_rep_get_stats64,
.ndo_setup_tc = bnxt_vf_rep_setup_tc,
+ .ndo_get_port_parent_id = bnxt_vf_rep_get_port_parent_id,
.ndo_get_phys_port_name = bnxt_vf_rep_get_phys_port_name
};
@@ -392,7 +389,6 @@ static void bnxt_vf_rep_netdev_init(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
dev->netdev_ops = &bnxt_vf_rep_netdev_ops;
dev->ethtool_ops = &bnxt_vf_rep_ethtool_ops;
- SWITCHDEV_SET_OPS(dev, &bnxt_vf_rep_switchdev_ops);
/* Just inherit all the featues of the parent PF as the VF-R
* uses the RX/TX rings of the parent PF
*/
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 03/12] liquidio: Implement ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
Liquidio only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it
a great candidate to be converted to use the ndo_get_port_parent_id()
NDO instead of implementing switchdev_port_attr_get().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
.../net/ethernet/cavium/liquidio/lio_main.c | 22 ++++------------
.../net/ethernet/cavium/liquidio/lio_vf_rep.c | 25 ++++++-------------
2 files changed, 12 insertions(+), 35 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 3d24133e5e49..e97e6754ee09 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -21,7 +21,6 @@
#include <linux/firmware.h>
#include <net/vxlan.h>
#include <linux/kthread.h>
-#include <net/switchdev.h>
#include "liquidio_common.h"
#include "octeon_droq.h"
#include "octeon_iq.h"
@@ -3184,7 +3183,8 @@ static const struct devlink_ops liquidio_devlink_ops = {
};
static int
-lio_pf_switchdev_attr_get(struct net_device *dev, struct switchdev_attr *attr)
+liquidio_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct lio *lio = GET_LIO(dev);
struct octeon_device *oct = lio->oct_dev;
@@ -3192,24 +3192,12 @@ lio_pf_switchdev_attr_get(struct net_device *dev, struct switchdev_attr *attr)
if (oct->eswitch_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV)
return -EOPNOTSUPP;
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = ETH_ALEN;
- ether_addr_copy(attr->u.ppid.id,
- (void *)&lio->linfo.hw_addr + 2);
- break;
-
- default:
- return -EOPNOTSUPP;
- }
+ ppid->id_len = ETH_ALEN;
+ ether_addr_copy(ppid->id, (void *)&lio->linfo.hw_addr + 2);
return 0;
}
-static const struct switchdev_ops lio_pf_switchdev_ops = {
- .switchdev_port_attr_get = lio_pf_switchdev_attr_get,
-};
-
static int liquidio_get_vf_stats(struct net_device *netdev, int vfidx,
struct ifla_vf_stats *vf_stats)
{
@@ -3259,6 +3247,7 @@ static const struct net_device_ops lionetdevops = {
.ndo_set_vf_trust = liquidio_set_vf_trust,
.ndo_set_vf_link_state = liquidio_set_vf_link_state,
.ndo_get_vf_stats = liquidio_get_vf_stats,
+ .ndo_get_port_parent_id = liquidio_get_port_parent_id,
};
/** \brief Entry point for the liquidio module
@@ -3534,7 +3523,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
* netdev tasks.
*/
netdev->netdev_ops = &lionetdevops;
- SWITCHDEV_SET_OPS(netdev, &lio_pf_switchdev_ops);
retval = netif_set_real_num_rx_queues(netdev, num_oqueues);
if (retval) {
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
index de61060721c4..f3f2e71431ac 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c
@@ -25,7 +25,6 @@
#include "octeon_nic.h"
#include "octeon_main.h"
#include "octeon_network.h"
-#include <net/switchdev.h>
#include "lio_vf_rep.h"
static int lio_vf_rep_open(struct net_device *ndev);
@@ -38,6 +37,8 @@ static int lio_vf_rep_phys_port_name(struct net_device *dev,
static void lio_vf_rep_get_stats64(struct net_device *dev,
struct rtnl_link_stats64 *stats64);
static int lio_vf_rep_change_mtu(struct net_device *ndev, int new_mtu);
+static int lio_vf_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid);
static const struct net_device_ops lio_vf_rep_ndev_ops = {
.ndo_open = lio_vf_rep_open,
@@ -47,6 +48,7 @@ static const struct net_device_ops lio_vf_rep_ndev_ops = {
.ndo_get_phys_port_name = lio_vf_rep_phys_port_name,
.ndo_get_stats64 = lio_vf_rep_get_stats64,
.ndo_change_mtu = lio_vf_rep_change_mtu,
+ .ndo_get_port_parent_id = lio_vf_get_port_parent_id,
};
static int
@@ -443,31 +445,19 @@ lio_vf_rep_pkt_xmit(struct sk_buff *skb, struct net_device *ndev)
return NETDEV_TX_OK;
}
-static int
-lio_vf_rep_attr_get(struct net_device *dev, struct switchdev_attr *attr)
+static int lio_vf_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid)
{
struct lio_vf_rep_desc *vf_rep = netdev_priv(dev);
struct net_device *parent_ndev = vf_rep->parent_ndev;
struct lio *lio = GET_LIO(parent_ndev);
- switch (attr->id) {
- case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- attr->u.ppid.id_len = ETH_ALEN;
- ether_addr_copy(attr->u.ppid.id,
- (void *)&lio->linfo.hw_addr + 2);
- break;
-
- default:
- return -EOPNOTSUPP;
- }
+ ppid->id_len = ETH_ALEN;
+ ether_addr_copy(ppid->id, (void *)&lio->linfo.hw_addr + 2);
return 0;
}
-static const struct switchdev_ops lio_vf_rep_switchdev_ops = {
- .switchdev_port_attr_get = lio_vf_rep_attr_get,
-};
-
static void
lio_vf_rep_fetch_stats(struct work_struct *work)
{
@@ -524,7 +514,6 @@ lio_vf_rep_create(struct octeon_device *oct)
ndev->min_mtu = LIO_MIN_MTU_SIZE;
ndev->max_mtu = LIO_MAX_MTU_SIZE;
ndev->netdev_ops = &lio_vf_rep_ndev_ops;
- SWITCHDEV_SET_OPS(ndev, &lio_vf_rep_switchdev_ops);
vf_rep = netdev_priv(ndev);
memset(vf_rep, 0, sizeof(*vf_rep));
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 01/12] net: Introduce ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
In-Reply-To: <20190205235326.14600-1-f.fainelli@gmail.com>
In preparation for getting rid of switchdev_ops, create a dedicated NDO
operation for getting the port's parent identifier. There are
essentially two classes of drivers that need to implement getting the
port's parent ID which are VF/PF drivers with a built-in switch, and
pure switchdev drivers such as mlxsw, ocelot, dsa etc.
We introduce a helper function: dev_get_port_parent_id() which supports
recursing into the lower devices to obtain the first port's parent ID.
Convert the bridge, core and ipv4 multicast routing code to check for
such ndo_get_port_parent_id() and call the helper functino when valid
before falling back to switchdev_port_attr_get(). This will allow us to
convert all relevant drivers in one go instead of having to implement
both switchdev_port_attr_get() and ndo_get_port_parent_id() operations,
then get rid of switchdev_port_attr_get().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
include/linux/netdevice.h | 9 +++++++
net/bridge/br_switchdev.c | 9 +++++--
net/core/dev.c | 57 +++++++++++++++++++++++++++++++++++++++
net/core/net-sysfs.c | 7 ++++-
net/core/rtnetlink.c | 6 ++++-
net/ipv4/ipmr.c | 8 +++++-
6 files changed, 91 insertions(+), 5 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ba57d0ba425e..1d95e634f3fe 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1188,6 +1188,10 @@ struct dev_ifalias {
* not implement this, it is assumed that the hw is not able to have
* multiple net devices on single physical port.
*
+ * int (*ndo_get_port_parent_id)(struct net_device *dev,
+ * struct netdev_phys_item_id *ppid)
+ * Called to get the parent ID of the physical port of this device.
+ *
* void (*ndo_udp_tunnel_add)(struct net_device *dev,
* struct udp_tunnel_info *ti);
* Called by UDP tunnel to notify a driver about the UDP port and socket
@@ -1412,6 +1416,8 @@ struct net_device_ops {
bool new_carrier);
int (*ndo_get_phys_port_id)(struct net_device *dev,
struct netdev_phys_item_id *ppid);
+ int (*ndo_get_port_parent_id)(struct net_device *dev,
+ struct netdev_phys_item_id *ppid);
int (*ndo_get_phys_port_name)(struct net_device *dev,
char *name, size_t len);
void (*ndo_udp_tunnel_add)(struct net_device *dev,
@@ -3651,6 +3657,9 @@ int dev_get_phys_port_id(struct net_device *dev,
struct netdev_phys_item_id *ppid);
int dev_get_phys_port_name(struct net_device *dev,
char *name, size_t len);
+int dev_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid, bool recurse);
+bool netdev_port_same_parent_id(struct net_device *a, struct net_device *b);
int dev_change_proto_down(struct net_device *dev, bool proto_down);
struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev, bool *again);
struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c
index 4d2b9eb7604a..06b0ae44585f 100644
--- a/net/bridge/br_switchdev.c
+++ b/net/bridge/br_switchdev.c
@@ -14,7 +14,8 @@ static int br_switchdev_mark_get(struct net_bridge *br, struct net_device *dev)
/* dev is yet to be added to the port list. */
list_for_each_entry(p, &br->port_list, list) {
- if (switchdev_port_same_parent_id(dev, p->dev))
+ if (netdev_port_same_parent_id(dev, p->dev) ||
+ switchdev_port_same_parent_id(dev, p->dev))
return p->offload_fwd_mark;
}
@@ -23,6 +24,7 @@ static int br_switchdev_mark_get(struct net_bridge *br, struct net_device *dev)
int nbp_switchdev_mark_set(struct net_bridge_port *p)
{
+ const struct net_device_ops *ops = p->dev->netdev_ops;
struct switchdev_attr attr = {
.orig_dev = p->dev,
.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
@@ -31,7 +33,10 @@ int nbp_switchdev_mark_set(struct net_bridge_port *p)
ASSERT_RTNL();
- err = switchdev_port_attr_get(p->dev, &attr);
+ if (ops->ndo_get_port_parent_id)
+ err = dev_get_port_parent_id(p->dev, &attr.u.ppid, true);
+ else
+ err = switchdev_port_attr_get(p->dev, &attr);
if (err) {
if (err == -EOPNOTSUPP)
return 0;
diff --git a/net/core/dev.c b/net/core/dev.c
index bfa4be42afff..8c6d5cf8a308 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7877,6 +7877,63 @@ int dev_get_phys_port_name(struct net_device *dev,
}
EXPORT_SYMBOL(dev_get_phys_port_name);
+/**
+ * dev_get_port_parent_id - Get the device's port parent identifier
+ * @dev: network device
+ * @ppid: pointer to a storage for the port's parent identifier
+ * @recurse: allow/disallow recursion to lower devices
+ *
+ * Get the devices's port parent identifier
+ */
+int dev_get_port_parent_id(struct net_device *dev,
+ struct netdev_phys_item_id *ppid,
+ bool recurse)
+{
+ const struct net_device_ops *ops = dev->netdev_ops;
+ struct netdev_phys_item_id first = { };
+ struct net_device *lower_dev;
+ struct list_head *iter;
+ int err = -EOPNOTSUPP;
+
+ if (ops->ndo_get_port_parent_id)
+ return ops->ndo_get_port_parent_id(dev, ppid);
+
+ if (!recurse)
+ return err;
+
+ netdev_for_each_lower_dev(dev, lower_dev, iter) {
+ err = dev_get_port_parent_id(lower_dev, ppid, recurse);
+ if (err)
+ break;
+ if (!first.id_len)
+ first = *ppid;
+ else if (memcmp(&first, ppid, sizeof(*ppid)))
+ return -ENODATA;
+ }
+
+ return err;
+}
+EXPORT_SYMBOL(dev_get_port_parent_id);
+
+/**
+ * netdev_port_same_parent_id - Indicate if two network devices have
+ * the same port parent identifier
+ * @a: first network device
+ * @b: second network device
+ */
+bool netdev_port_same_parent_id(struct net_device *a, struct net_device *b)
+{
+ struct netdev_phys_item_id a_id = { };
+ struct netdev_phys_item_id b_id = { };
+
+ if (dev_get_port_parent_id(a, &a_id, true) ||
+ dev_get_port_parent_id(b, &b_id, true))
+ return false;
+
+ return netdev_phys_item_id_same(&a_id, &b_id);
+}
+EXPORT_SYMBOL(netdev_port_same_parent_id);
+
/**
* dev_change_proto_down - update protocol port state information
* @dev: device
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index ff9fd2bb4ce4..4eace9f1dcf9 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -495,6 +495,7 @@ static ssize_t phys_switch_id_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct net_device *netdev = to_net_dev(dev);
+ const struct net_device_ops *ops = netdev->netdev_ops;
ssize_t ret = -EINVAL;
if (!rtnl_trylock())
@@ -507,7 +508,11 @@ static ssize_t phys_switch_id_show(struct device *dev,
.flags = SWITCHDEV_F_NO_RECURSE,
};
- ret = switchdev_port_attr_get(netdev, &attr);
+ if (ops->ndo_get_port_parent_id)
+ ret = dev_get_port_parent_id(netdev, &attr.u.ppid,
+ false);
+ else
+ ret = switchdev_port_attr_get(netdev, &attr);
if (!ret)
ret = sprintf(buf, "%*phN\n", attr.u.ppid.id_len,
attr.u.ppid.id);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index f5a98082ac7a..90dd02c1f561 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1146,6 +1146,7 @@ static int rtnl_phys_port_name_fill(struct sk_buff *skb, struct net_device *dev)
static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
{
+ const struct net_device_ops *ops = dev->netdev_ops;
int err;
struct switchdev_attr attr = {
.orig_dev = dev,
@@ -1153,7 +1154,10 @@ static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev)
.flags = SWITCHDEV_F_NO_RECURSE,
};
- err = switchdev_port_attr_get(dev, &attr);
+ if (ops->ndo_get_port_parent_id)
+ err = dev_get_port_parent_id(dev, &attr.u.ppid, false);
+ else
+ err = switchdev_port_attr_get(dev, &attr);
if (err) {
if (err == -EOPNOTSUPP)
return 0;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index fb99002c3d4e..c71bcc42d66d 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -837,6 +837,7 @@ static void ipmr_update_thresholds(struct mr_table *mrt, struct mr_mfc *cache,
static int vif_add(struct net *net, struct mr_table *mrt,
struct vifctl *vifc, int mrtsock)
{
+ const struct net_device_ops *ops;
int vifi = vifc->vifc_vifi;
struct switchdev_attr attr = {
.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
@@ -920,7 +921,12 @@ static int vif_add(struct net *net, struct mr_table *mrt,
(VIFF_TUNNEL | VIFF_REGISTER));
attr.orig_dev = dev;
- if (!switchdev_port_attr_get(dev, &attr)) {
+ ops = dev->netdev_ops;
+ if (ops->ndo_get_port_parent_id &&
+ !dev_get_port_parent_id(dev, &attr.u.ppid, true)) {
+ memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
+ v->dev_parent_id.id_len = attr.u.ppid.id_len;
+ } else if (!switchdev_port_attr_get(dev, &attr)) {
memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
v->dev_parent_id.id_len = attr.u.ppid.id_len;
} else {
--
2.17.1
^ permalink raw reply related
* [PATCH net-next v3 00/12] net: Introduce ndo_get_port_parent_id()
From: Florian Fainelli @ 2019-02-05 23:53 UTC (permalink / raw)
To: netdev
Cc: Florian Fainelli, David S. Miller, Ido Schimmel, open list,
open list:MELLANOX MLX5 core VPI driver,
open list:NETRONOME ETHERNET DRIVERS, open list:STAGING SUBSYSTEM,
moderated list:ETHERNET BRIDGE
Hi all,
Based on discussion with Ido and feedback from Jakub there are clearly
two classes of users that implement SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
- PF/VF drivers which typically only implement return the port's parent
ID, yet have to implement switchdev_port_attr_get() just for that
- Ethernet switch drivers: mlxsw, ocelot, DSA, etc. which implement more
attributes which we want to be able to eventually veto in the context
of the caller, thus making them candidates for using a blocking notifier
chain
Changes in v3:
- keep ethsw's switchdev_ops assignment
- remove inclusion of net/switchdev.h in netdevsim which is no longer
necesary
Changes in v2:
- resolved build failures spotted by kbuild test robot
- added helpers functions into the core network device layer:
dev_get_port_parent_id() and netdev_port_same_parent_id();
- added support for recursion to lower devices
Changes from RFC:
- introduce a ndo_get_port_parent_id() and convert all relevant drivers
to use it
- get rid of SWITCHDEV_ATTR_ID_PORT_PARENT_ID
A subsequent set of patches will convert switchdev_port_attr_set() to
use a blocking notifier call, and still get rid of
switchdev_port_attr_get() altogether.
Florian Fainelli (12):
net: Introduce ndo_get_port_parent_id()
bnxt: Implement ndo_get_port_parent_id()
liquidio: Implement ndo_get_port_parent_id()
net/mlx5e: Implement ndo_get_port_parent_id()
mlxsw: Implement ndo_get_port_parent_id()
mscc: ocelot: Implement ndo_get_port_parent_id()
nfp: Implement ndo_get_port_parent_id()
rocker: Implement ndo_get_port_parent_id()
netdevsim: Implement ndo_get_port_parent_id()
staging: fsl-dpaa2: ethsw: Implement ndo_get_port_parent_id()
net: dsa: Implement ndo_get_port_parent_id()
net: Get rid of SWITCHDEV_ATTR_ID_PORT_PARENT_ID
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 28 +++------
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 4 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c | 3 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c | 12 ++--
.../net/ethernet/cavium/liquidio/lio_main.c | 22 ++-----
.../net/ethernet/cavium/liquidio/lio_vf_rep.c | 25 +++-----
.../ethernet/mellanox/mlx5/core/en/tc_tun.c | 2 +-
.../net/ethernet/mellanox/mlx5/core/en_rep.c | 31 ++++------
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 5 +-
.../net/ethernet/mellanox/mlxsw/spectrum.c | 13 +++++
.../mellanox/mlxsw/spectrum_switchdev.c | 5 --
.../net/ethernet/mellanox/mlxsw/switchx2.c | 36 +++++-------
drivers/net/ethernet/mscc/ocelot.c | 33 +++++------
.../ethernet/netronome/nfp/flower/action.c | 3 +-
.../ethernet/netronome/nfp/nfp_net_common.c | 4 +-
.../net/ethernet/netronome/nfp/nfp_net_repr.c | 4 +-
drivers/net/ethernet/netronome/nfp/nfp_port.c | 23 ++------
drivers/net/ethernet/netronome/nfp/nfp_port.h | 4 +-
drivers/net/ethernet/rocker/rocker_main.c | 17 ++++--
drivers/net/netdevsim/netdev.c | 23 ++------
drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 16 ++++--
include/linux/netdevice.h | 9 +++
include/net/switchdev.h | 11 ----
net/bridge/br_switchdev.c | 11 ++--
net/core/dev.c | 57 +++++++++++++++++++
net/core/net-sysfs.c | 12 +---
net/core/rtnetlink.c | 14 ++---
net/dsa/slave.c | 18 ++++--
net/ipv4/ipmr.c | 14 ++---
net/switchdev/switchdev.c | 20 -------
30 files changed, 219 insertions(+), 260 deletions(-)
--
2.17.1
^ permalink raw reply
* [PATCH v2 1/6] ethtool: move option parsing related code into function
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
Move option parsing code into find_option function.
No behavior changes.
Based on patch by Kan Liang <kan.liang@intel.com>
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.c | 49 +++++++++++++++++++++++++++++++------------------
1 file changed, 31 insertions(+), 18 deletions(-)
diff --git a/ethtool.c b/ethtool.c
index 2f7e96b..8b7c224 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -5265,6 +5265,29 @@ static int show_usage(struct cmd_context *ctx)
return 0;
}
+static int find_option(int argc, char **argp)
+{
+ const char *opt;
+ size_t len;
+ int k;
+
+ for (k = 0; args[k].opts; k++) {
+ opt = args[k].opts;
+ for (;;) {
+ len = strcspn(opt, "|");
+ if (strncmp(*argp, opt, len) == 0 &&
+ (*argp)[len] == 0)
+ return k;
+
+ if (opt[len] == 0)
+ break;
+ opt += len + 1;
+ }
+ }
+
+ return -1;
+}
+
int main(int argc, char **argp)
{
int (*func)(struct cmd_context *);
@@ -5284,24 +5307,14 @@ int main(int argc, char **argp)
*/
if (argc == 0)
exit_bad_args();
- for (k = 0; args[k].opts; k++) {
- const char *opt;
- size_t len;
- opt = args[k].opts;
- for (;;) {
- len = strcspn(opt, "|");
- if (strncmp(*argp, opt, len) == 0 &&
- (*argp)[len] == 0) {
- argp++;
- argc--;
- func = args[k].func;
- want_device = args[k].want_device;
- goto opt_found;
- }
- if (opt[len] == 0)
- break;
- opt += len + 1;
- }
+
+ k = find_option(argc, argp);
+ if (k >= 0) {
+ argp++;
+ argc--;
+ func = args[k].func;
+ want_device = args[k].want_device;
+ goto opt_found;
}
if ((*argp)[0] == '-')
exit_bad_args();
--
2.20.1
^ permalink raw reply related
* [PATCH v2 4/6] ethtool: support per-queue sub command --show-coalesce
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
In-Reply-To: <20190206000106.24364-1-jeffrey.t.kirsher@intel.com>
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
Get all masked queues' coalesce settings from kernel and dump them one by
one.
Example:
$ sudo ./ethtool --set-perqueue-command eth5 queue_mask 0x11
--show-coalesce
Queue: 0
Adaptive RX: off TX: off
stats-block-usecs: 0
sample-interval: 0
pkt-rate-low: 0
pkt-rate-high: 0
rx-usecs: 222
rx-frames: 0
rx-usecs-irq: 0
rx-frames-irq: 256
tx-usecs: 222
tx-frames: 0
tx-usecs-irq: 0
tx-frames-irq: 256
rx-usecs-low: 0
rx-frame-low: 0
tx-usecs-low: 0
tx-frame-low: 0
rx-usecs-high: 0
rx-frame-high: 0
tx-usecs-high: 0
tx-frame-high: 0
Queue: 4
Adaptive RX: off TX: off
stats-block-usecs: 0
sample-interval: 0
pkt-rate-low: 0
pkt-rate-high: 0
rx-usecs: 222
rx-frames: 0
rx-usecs-irq: 0
rx-frames-irq: 256
tx-usecs: 222
tx-frames: 0
tx-usecs-irq: 0
tx-frames-irq: 256
rx-usecs-low: 0
rx-frame-low: 0
tx-usecs-low: 0
tx-frame-low: 0
rx-usecs-high: 0
rx-frame-high: 0
tx-usecs-high: 0
tx-frame-high: 0
Based on patch by Kan Liang <kan.liang@intel.com>
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.8.in | 2 +-
ethtool.c | 64 ++++++++++++++++++++++++++++++++++++++++++++++++++--
2 files changed, 63 insertions(+), 3 deletions(-)
diff --git a/ethtool.8.in b/ethtool.8.in
index 0aaca2c..71bb962 100644
--- a/ethtool.8.in
+++ b/ethtool.8.in
@@ -1153,7 +1153,7 @@ Sets the specific queues which the sub command is applied to.
If queue_mask is not set, the sub command will be applied to all queues.
.TP
.B sub_command
-Sets the sub command.
+Sets the sub command. The supported sub commands include --show-coalesce.
.RE
.SH BUGS
Not supported (in part or whole) on all network drivers.
diff --git a/ethtool.c b/ethtool.c
index 4dc725c..9a1b83b 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -1409,6 +1409,29 @@ static int dump_coalesce(const struct ethtool_coalesce *ecoal)
return 0;
}
+void dump_per_queue_coalesce(struct ethtool_per_queue_op *per_queue_opt,
+ __u32 *queue_mask)
+{
+ char *addr;
+ int i;
+
+ addr = (char *)per_queue_opt + sizeof(*per_queue_opt);
+ for (i = 0; i < __KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32); i++) {
+ int queue = i * 32;
+ __u32 mask = queue_mask[i];
+
+ while (mask > 0) {
+ if (mask & 0x1) {
+ fprintf(stdout, "Queue: %d\n", queue);
+ dump_coalesce((struct ethtool_coalesce *)addr);
+ addr += sizeof(struct ethtool_coalesce);
+ }
+ mask = mask >> 1;
+ queue++;
+ }
+ }
+}
+
struct feature_state {
u32 off_flags;
struct ethtool_gfeatures features;
@@ -5245,7 +5268,8 @@ static const struct option {
{ "--show-fec", 1, do_gfec, "Show FEC settings"},
{ "--set-fec", 1, do_sfec, "Set FEC settings",
" [ encoding auto|off|rs|baser [...]]\n"},
- { "--set-perqueue-command", 1, do_perqueue, "Set per queue command",
+ { "--set-perqueue-command", 1, do_perqueue, "Set per queue command. "
+ "The supported sub commands include --show-coalesce",
" [queue_mask %x] SUB_COMMAND\n"},
{ "-h|--help", 0, show_usage, "Show this help" },
{ "--version", 0, do_version, "Show version number" },
@@ -5350,8 +5374,32 @@ static int find_max_num_queues(struct cmd_context *ctx)
echannels.combined_count);
}
+static struct ethtool_per_queue_op *
+get_per_queue_coalesce(struct cmd_context *ctx, __u32 *queue_mask, int n_queues)
+{
+ struct ethtool_per_queue_op *per_queue_opt;
+
+ per_queue_opt = malloc(sizeof(*per_queue_opt) + n_queues *
+ sizeof(struct ethtool_coalesce));
+ if (!per_queue_opt)
+ return NULL;
+
+ memcpy(per_queue_opt->queue_mask, queue_mask,
+ __KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32) * sizeof(__u32));
+ per_queue_opt->cmd = ETHTOOL_PERQUEUE;
+ per_queue_opt->sub_command = ETHTOOL_GCOALESCE;
+ if (send_ioctl(ctx, per_queue_opt)) {
+ free(per_queue_opt);
+ perror("Cannot get device per queue parameters");
+ return NULL;
+ }
+
+ return per_queue_opt;
+}
+
static int do_perqueue(struct cmd_context *ctx)
{
+ struct ethtool_per_queue_op *per_queue_opt;
__u32 queue_mask[__KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32)] = {0};
int i, n_queues = 0;
@@ -5390,7 +5438,19 @@ static int do_perqueue(struct cmd_context *ctx)
if (i < 0)
exit_bad_args();
- /* no sub_command support yet */
+ if (strstr(args[i].opts, "--show-coalesce") != NULL) {
+ per_queue_opt = get_per_queue_coalesce(ctx, queue_mask,
+ n_queues);
+ if (per_queue_opt == NULL) {
+ perror("Cannot get device per queue parameters");
+ return -EFAULT;
+ }
+ dump_per_queue_coalesce(per_queue_opt, queue_mask);
+ free(per_queue_opt);
+ } else {
+ perror("The subcommand is not supported yet");
+ return -EOPNOTSUPP;
+ }
return 0;
}
--
2.20.1
^ permalink raw reply related
* [PATCH v2 2/6] ethtool: move cmdline_coalesce out of do_scoalesce
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
In-Reply-To: <20190206000106.24364-1-jeffrey.t.kirsher@intel.com>
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
Move the definition of cmdline_coalesce out of do_scoalesce and into a
macro so it can be reused across functions.
No behavior change.
Based on patch by Kan Liang <kan.liang@intel.com>
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.c | 142 ++++++++++++++++++++++++++++--------------------------
1 file changed, 74 insertions(+), 68 deletions(-)
diff --git a/ethtool.c b/ethtool.c
index 8b7c224..af266c5 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -2098,78 +2098,84 @@ static int do_gcoalesce(struct cmd_context *ctx)
return 0;
}
+#define DECLARE_COALESCE_OPTION_VARS() \
+ s32 coal_stats_wanted = -1; \
+ int coal_adaptive_rx_wanted = -1; \
+ int coal_adaptive_tx_wanted = -1; \
+ s32 coal_sample_rate_wanted = -1; \
+ s32 coal_pkt_rate_low_wanted = -1; \
+ s32 coal_pkt_rate_high_wanted = -1; \
+ s32 coal_rx_usec_wanted = -1; \
+ s32 coal_rx_frames_wanted = -1; \
+ s32 coal_rx_usec_irq_wanted = -1; \
+ s32 coal_rx_frames_irq_wanted = -1; \
+ s32 coal_tx_usec_wanted = -1; \
+ s32 coal_tx_frames_wanted = -1; \
+ s32 coal_tx_usec_irq_wanted = -1; \
+ s32 coal_tx_frames_irq_wanted = -1; \
+ s32 coal_rx_usec_low_wanted = -1; \
+ s32 coal_rx_frames_low_wanted = -1; \
+ s32 coal_tx_usec_low_wanted = -1; \
+ s32 coal_tx_frames_low_wanted = -1; \
+ s32 coal_rx_usec_high_wanted = -1; \
+ s32 coal_rx_frames_high_wanted = -1; \
+ s32 coal_tx_usec_high_wanted = -1; \
+ s32 coal_tx_frames_high_wanted = -1
+
+#define COALESCE_CMDLINE_INFO(__ecoal) \
+{ \
+ { "adaptive-rx", CMDL_BOOL, &coal_adaptive_rx_wanted, \
+ &__ecoal.use_adaptive_rx_coalesce }, \
+ { "adaptive-tx", CMDL_BOOL, &coal_adaptive_tx_wanted, \
+ &__ecoal.use_adaptive_tx_coalesce }, \
+ { "sample-interval", CMDL_S32, &coal_sample_rate_wanted, \
+ &__ecoal.rate_sample_interval }, \
+ { "stats-block-usecs", CMDL_S32, &coal_stats_wanted, \
+ &__ecoal.stats_block_coalesce_usecs }, \
+ { "pkt-rate-low", CMDL_S32, &coal_pkt_rate_low_wanted, \
+ &__ecoal.pkt_rate_low }, \
+ { "pkt-rate-high", CMDL_S32, &coal_pkt_rate_high_wanted, \
+ &__ecoal.pkt_rate_high }, \
+ { "rx-usecs", CMDL_S32, &coal_rx_usec_wanted, \
+ &__ecoal.rx_coalesce_usecs }, \
+ { "rx-frames", CMDL_S32, &coal_rx_frames_wanted, \
+ &__ecoal.rx_max_coalesced_frames }, \
+ { "rx-usecs-irq", CMDL_S32, &coal_rx_usec_irq_wanted, \
+ &__ecoal.rx_coalesce_usecs_irq }, \
+ { "rx-frames-irq", CMDL_S32, &coal_rx_frames_irq_wanted, \
+ &__ecoal.rx_max_coalesced_frames_irq }, \
+ { "tx-usecs", CMDL_S32, &coal_tx_usec_wanted, \
+ &__ecoal.tx_coalesce_usecs }, \
+ { "tx-frames", CMDL_S32, &coal_tx_frames_wanted, \
+ &__ecoal.tx_max_coalesced_frames }, \
+ { "tx-usecs-irq", CMDL_S32, &coal_tx_usec_irq_wanted, \
+ &__ecoal.tx_coalesce_usecs_irq }, \
+ { "tx-frames-irq", CMDL_S32, &coal_tx_frames_irq_wanted, \
+ &__ecoal.tx_max_coalesced_frames_irq }, \
+ { "rx-usecs-low", CMDL_S32, &coal_rx_usec_low_wanted, \
+ &__ecoal.rx_coalesce_usecs_low }, \
+ { "rx-frames-low", CMDL_S32, &coal_rx_frames_low_wanted, \
+ &__ecoal.rx_max_coalesced_frames_low }, \
+ { "tx-usecs-low", CMDL_S32, &coal_tx_usec_low_wanted, \
+ &__ecoal.tx_coalesce_usecs_low }, \
+ { "tx-frames-low", CMDL_S32, &coal_tx_frames_low_wanted, \
+ &__ecoal.tx_max_coalesced_frames_low }, \
+ { "rx-usecs-high", CMDL_S32, &coal_rx_usec_high_wanted, \
+ &__ecoal.rx_coalesce_usecs_high }, \
+ { "rx-frames-high", CMDL_S32, &coal_rx_frames_high_wanted, \
+ &__ecoal.rx_max_coalesced_frames_high }, \
+ { "tx-usecs-high", CMDL_S32, &coal_tx_usec_high_wanted, \
+ &__ecoal.tx_coalesce_usecs_high }, \
+ { "tx-frames-high", CMDL_S32, &coal_tx_frames_high_wanted, \
+ &__ecoal.tx_max_coalesced_frames_high }, \
+}
+
static int do_scoalesce(struct cmd_context *ctx)
{
struct ethtool_coalesce ecoal;
int gcoalesce_changed = 0;
- s32 coal_stats_wanted = -1;
- int coal_adaptive_rx_wanted = -1;
- int coal_adaptive_tx_wanted = -1;
- s32 coal_sample_rate_wanted = -1;
- s32 coal_pkt_rate_low_wanted = -1;
- s32 coal_pkt_rate_high_wanted = -1;
- s32 coal_rx_usec_wanted = -1;
- s32 coal_rx_frames_wanted = -1;
- s32 coal_rx_usec_irq_wanted = -1;
- s32 coal_rx_frames_irq_wanted = -1;
- s32 coal_tx_usec_wanted = -1;
- s32 coal_tx_frames_wanted = -1;
- s32 coal_tx_usec_irq_wanted = -1;
- s32 coal_tx_frames_irq_wanted = -1;
- s32 coal_rx_usec_low_wanted = -1;
- s32 coal_rx_frames_low_wanted = -1;
- s32 coal_tx_usec_low_wanted = -1;
- s32 coal_tx_frames_low_wanted = -1;
- s32 coal_rx_usec_high_wanted = -1;
- s32 coal_rx_frames_high_wanted = -1;
- s32 coal_tx_usec_high_wanted = -1;
- s32 coal_tx_frames_high_wanted = -1;
- struct cmdline_info cmdline_coalesce[] = {
- { "adaptive-rx", CMDL_BOOL, &coal_adaptive_rx_wanted,
- &ecoal.use_adaptive_rx_coalesce },
- { "adaptive-tx", CMDL_BOOL, &coal_adaptive_tx_wanted,
- &ecoal.use_adaptive_tx_coalesce },
- { "sample-interval", CMDL_S32, &coal_sample_rate_wanted,
- &ecoal.rate_sample_interval },
- { "stats-block-usecs", CMDL_S32, &coal_stats_wanted,
- &ecoal.stats_block_coalesce_usecs },
- { "pkt-rate-low", CMDL_S32, &coal_pkt_rate_low_wanted,
- &ecoal.pkt_rate_low },
- { "pkt-rate-high", CMDL_S32, &coal_pkt_rate_high_wanted,
- &ecoal.pkt_rate_high },
- { "rx-usecs", CMDL_S32, &coal_rx_usec_wanted,
- &ecoal.rx_coalesce_usecs },
- { "rx-frames", CMDL_S32, &coal_rx_frames_wanted,
- &ecoal.rx_max_coalesced_frames },
- { "rx-usecs-irq", CMDL_S32, &coal_rx_usec_irq_wanted,
- &ecoal.rx_coalesce_usecs_irq },
- { "rx-frames-irq", CMDL_S32, &coal_rx_frames_irq_wanted,
- &ecoal.rx_max_coalesced_frames_irq },
- { "tx-usecs", CMDL_S32, &coal_tx_usec_wanted,
- &ecoal.tx_coalesce_usecs },
- { "tx-frames", CMDL_S32, &coal_tx_frames_wanted,
- &ecoal.tx_max_coalesced_frames },
- { "tx-usecs-irq", CMDL_S32, &coal_tx_usec_irq_wanted,
- &ecoal.tx_coalesce_usecs_irq },
- { "tx-frames-irq", CMDL_S32, &coal_tx_frames_irq_wanted,
- &ecoal.tx_max_coalesced_frames_irq },
- { "rx-usecs-low", CMDL_S32, &coal_rx_usec_low_wanted,
- &ecoal.rx_coalesce_usecs_low },
- { "rx-frames-low", CMDL_S32, &coal_rx_frames_low_wanted,
- &ecoal.rx_max_coalesced_frames_low },
- { "tx-usecs-low", CMDL_S32, &coal_tx_usec_low_wanted,
- &ecoal.tx_coalesce_usecs_low },
- { "tx-frames-low", CMDL_S32, &coal_tx_frames_low_wanted,
- &ecoal.tx_max_coalesced_frames_low },
- { "rx-usecs-high", CMDL_S32, &coal_rx_usec_high_wanted,
- &ecoal.rx_coalesce_usecs_high },
- { "rx-frames-high", CMDL_S32, &coal_rx_frames_high_wanted,
- &ecoal.rx_max_coalesced_frames_high },
- { "tx-usecs-high", CMDL_S32, &coal_tx_usec_high_wanted,
- &ecoal.tx_coalesce_usecs_high },
- { "tx-frames-high", CMDL_S32, &coal_tx_frames_high_wanted,
- &ecoal.tx_max_coalesced_frames_high },
- };
+ DECLARE_COALESCE_OPTION_VARS();
+ struct cmdline_info cmdline_coalesce[] = COALESCE_CMDLINE_INFO(ecoal);
int err, changed = 0;
parse_generic_cmdline(ctx, &gcoalesce_changed,
--
2.20.1
^ permalink raw reply related
* [PATCH v2 5/6] ethtool: support per-queue sub command --coalesce
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
In-Reply-To: <20190206000106.24364-1-jeffrey.t.kirsher@intel.com>
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
This patch adds the ability to configure the coalesce settings from
do_scoalesce on a per-queue basis.
For each masked queue the current settings are read, modified, and written
back to the kernel.
Example:
$ sudo ./ethtool --set-perqueue-command eth5 queue_mask 0x1 --coalesce
rx-usecs 10 tx-usecs 5
$ sudo ./ethtool --set-perqueue-command eth5 queue_mask 0x1
--show-coalesce
Queue: 0
Adaptive RX: on TX: on
stats-block-usecs: 0
sample-interval: 0
pkt-rate-low: 0
pkt-rate-high: 0
rx-usecs: 10
rx-frames: 0
rx-usecs-irq: 0
rx-frames-irq: 256
tx-usecs: 5
tx-frames: 0
tx-usecs-irq: 0
tx-frames-irq: 256
rx-usecs-low: 0
rx-frame-low: 0
tx-usecs-low: 0
tx-frame-low: 0
rx-usecs-high: 0
rx-frame-high: 0
tx-usecs-high: 0
tx-frame-high: 0
Based on patch by Kan Liang <kan.liang@intel.com>
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.8.in | 3 ++-
ethtool.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 61 insertions(+), 2 deletions(-)
diff --git a/ethtool.8.in b/ethtool.8.in
index 71bb962..c7202e8 100644
--- a/ethtool.8.in
+++ b/ethtool.8.in
@@ -1153,7 +1153,8 @@ Sets the specific queues which the sub command is applied to.
If queue_mask is not set, the sub command will be applied to all queues.
.TP
.B sub_command
-Sets the sub command. The supported sub commands include --show-coalesce.
+Sets the sub command. The supported sub commands include --show-coalesce and
+--coalesce.
.RE
.SH BUGS
Not supported (in part or whole) on all network drivers.
diff --git a/ethtool.c b/ethtool.c
index 9a1b83b..01bdaf1 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -5269,7 +5269,7 @@ static const struct option {
{ "--set-fec", 1, do_sfec, "Set FEC settings",
" [ encoding auto|off|rs|baser [...]]\n"},
{ "--set-perqueue-command", 1, do_perqueue, "Set per queue command. "
- "The supported sub commands include --show-coalesce",
+ "The supported sub commands include --show-coalesce, --coalesce",
" [queue_mask %x] SUB_COMMAND\n"},
{ "-h|--help", 0, show_usage, "Show this help" },
{ "--version", 0, do_version, "Show version number" },
@@ -5397,6 +5397,53 @@ get_per_queue_coalesce(struct cmd_context *ctx, __u32 *queue_mask, int n_queues)
return per_queue_opt;
}
+static void set_per_queue_coalesce(struct cmd_context *ctx,
+ struct ethtool_per_queue_op *per_queue_opt)
+{
+ struct ethtool_coalesce ecoal;
+ DECLARE_COALESCE_OPTION_VARS();
+ struct cmdline_info cmdline_coalesce[] = COALESCE_CMDLINE_INFO(ecoal);
+ __u32 *queue_mask = per_queue_opt->queue_mask;
+ char *addr = (char *)per_queue_opt + sizeof(*per_queue_opt);
+ int gcoalesce_changed = 0;
+ int i;
+
+ parse_generic_cmdline(ctx, &gcoalesce_changed,
+ cmdline_coalesce, ARRAY_SIZE(cmdline_coalesce));
+
+ for (i = 0; i < __KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32); i++) {
+ int queue = i * 32;
+ __u32 mask = queue_mask[i];
+
+ while (mask > 0) {
+ if (mask & 0x1) {
+ int changed = 0;
+
+ memcpy(&ecoal, addr,
+ sizeof(struct ethtool_coalesce));
+ do_generic_set(cmdline_coalesce,
+ ARRAY_SIZE(cmdline_coalesce),
+ &changed);
+ if (!changed)
+ fprintf(stderr,
+ "Queue %d, no coalesce parameters changed\n",
+ queue);
+ memcpy(addr, &ecoal,
+ sizeof(struct ethtool_coalesce));
+ addr += sizeof(struct ethtool_coalesce);
+ }
+ mask = mask >> 1;
+ queue++;
+ }
+ }
+
+ per_queue_opt->cmd = ETHTOOL_PERQUEUE;
+ per_queue_opt->sub_command = ETHTOOL_SCOALESCE;
+
+ if (send_ioctl(ctx, per_queue_opt))
+ perror("Cannot set device per queue parameters");
+}
+
static int do_perqueue(struct cmd_context *ctx)
{
struct ethtool_per_queue_op *per_queue_opt;
@@ -5447,6 +5494,17 @@ static int do_perqueue(struct cmd_context *ctx)
}
dump_per_queue_coalesce(per_queue_opt, queue_mask);
free(per_queue_opt);
+ } else if (strstr(args[i].opts, "--coalesce") != NULL) {
+ ctx->argc--;
+ ctx->argp++;
+ per_queue_opt = get_per_queue_coalesce(ctx, queue_mask,
+ n_queues);
+ if (per_queue_opt == NULL) {
+ perror("Cannot get device per queue parameters");
+ return -EFAULT;
+ }
+ set_per_queue_coalesce(ctx, per_queue_opt);
+ free(per_queue_opt);
} else {
perror("The subcommand is not supported yet");
return -EOPNOTSUPP;
--
2.20.1
^ permalink raw reply related
* [PATCH v2 6/6] ethtool: fix up dump_coalesce output to match actual option names
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
In-Reply-To: <20190206000106.24364-1-jeffrey.t.kirsher@intel.com>
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
When the coalesce settings are printed with --show-coalesce a few of the
option names lack the pluralization that is present in the man page and
usage info, but are otherwise identical.
This inconsistency could lead to some confusion if a user attempts to set
the coalesce settings by matching the output they see from --show-coalesce,
so fix this.
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/ethtool.c b/ethtool.c
index 01bdaf1..03e5008 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -1372,14 +1372,14 @@ static int dump_coalesce(const struct ethtool_coalesce *ecoal)
"tx-frames-irq: %u\n"
"\n"
"rx-usecs-low: %u\n"
- "rx-frame-low: %u\n"
+ "rx-frames-low: %u\n"
"tx-usecs-low: %u\n"
- "tx-frame-low: %u\n"
+ "tx-frames-low: %u\n"
"\n"
"rx-usecs-high: %u\n"
- "rx-frame-high: %u\n"
+ "rx-frames-high: %u\n"
"tx-usecs-high: %u\n"
- "tx-frame-high: %u\n"
+ "tx-frames-high: %u\n"
"\n",
ecoal->stats_block_coalesce_usecs,
ecoal->rate_sample_interval,
--
2.20.1
^ permalink raw reply related
* [PATCH v2 3/6] ethtool: introduce new ioctl for per-queue settings
From: Jeff Kirsher @ 2019-02-06 0:01 UTC (permalink / raw)
To: linville; +Cc: Nicholas Nunley, netdev, nhorman, sassmann, Jeff Kirsher
In-Reply-To: <20190206000106.24364-1-jeffrey.t.kirsher@intel.com>
From: Nicholas Nunley <nicholas.d.nunley@intel.com>
Introduce a new ioctl for setting per-queue parameters.
Users can apply commands to specific queues by setting SUB_COMMAND and
queue_mask with the following ethtool command:
ethtool --set-perqueue-command DEVNAME [queue_mask %x] SUB_COMMAND
If queue_mask is not set, the SUB_COMMAND will be applied to all queues.
SUB_COMMANDs for per-queue settings will be implemented in following
patches.
Based on patch by Kan Liang <kan.liang@intel.com>
Signed-off-by: Nicholas Nunley <nicholas.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
ethtool.8.in | 20 ++++++++++
ethtool.c | 101 +++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 121 insertions(+)
diff --git a/ethtool.8.in b/ethtool.8.in
index 5a26cff..0aaca2c 100644
--- a/ethtool.8.in
+++ b/ethtool.8.in
@@ -391,6 +391,14 @@ ethtool \- query or control network driver and hardware settings
.I devname
.B encoding
.BR auto | off | rs | baser \ [...]
+.HP
+.B ethtool \-\-set\-perqueue\-command
+.I devname
+.RB [ queue_mask
+.IR %x ]
+.I sub_command
+.RB ...
+ .
.
.\" Adjust lines (i.e. full justification) and hyphenate.
.ad
@@ -1135,6 +1143,18 @@ RS Force RS-FEC encoding
BaseR Force BaseR encoding
.TE
.RE
+.TP
+.B \-\-set\-perqueue\-command
+Sets sub command to specific queues.
+.RS 4
+.TP
+.B queue_mask %x
+Sets the specific queues which the sub command is applied to.
+If queue_mask is not set, the sub command will be applied to all queues.
+.TP
+.B sub_command
+Sets the sub command.
+.RE
.SH BUGS
Not supported (in part or whole) on all network drivers.
.SH AUTHOR
diff --git a/ethtool.c b/ethtool.c
index af266c5..4dc725c 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -5048,6 +5048,8 @@ static int do_sfec(struct cmd_context *ctx)
return 0;
}
+static int do_perqueue(struct cmd_context *ctx);
+
#ifndef TEST_ETHTOOL
int send_ioctl(struct cmd_context *ctx, void *cmd)
{
@@ -5243,6 +5245,8 @@ static const struct option {
{ "--show-fec", 1, do_gfec, "Show FEC settings"},
{ "--set-fec", 1, do_sfec, "Set FEC settings",
" [ encoding auto|off|rs|baser [...]]\n"},
+ { "--set-perqueue-command", 1, do_perqueue, "Set per queue command",
+ " [queue_mask %x] SUB_COMMAND\n"},
{ "-h|--help", 0, show_usage, "Show this help" },
{ "--version", 0, do_version, "Show version number" },
{}
@@ -5294,6 +5298,103 @@ static int find_option(int argc, char **argp)
return -1;
}
+static int set_queue_mask(u32 *queue_mask, char *str)
+{
+ int len = strlen(str);
+ int index = __KERNEL_DIV_ROUND_UP(len * 4, 32);
+ char tmp[9];
+ char *end = str + len;
+ int i, num;
+ __u32 mask;
+ int n_queues = 0;
+
+ if (len > MAX_NUM_QUEUE)
+ return -EINVAL;
+
+ for (i = 0; i < index; i++) {
+ num = end - str;
+
+ if (num >= 8) {
+ end -= 8;
+ num = 8;
+ } else {
+ end = str;
+ }
+ strncpy(tmp, end, num);
+ tmp[num] = '\0';
+
+ queue_mask[i] = strtoul(tmp, NULL, 16);
+
+ mask = queue_mask[i];
+ while (mask > 0) {
+ if (mask & 0x1)
+ n_queues++;
+ mask = mask >> 1;
+ }
+ }
+
+ return n_queues;
+}
+
+#define MAX(x, y) (x > y ? x : y)
+
+static int find_max_num_queues(struct cmd_context *ctx)
+{
+ struct ethtool_channels echannels;
+
+ echannels.cmd = ETHTOOL_GCHANNELS;
+ if (send_ioctl(ctx, &echannels))
+ return -1;
+
+ return MAX(MAX(echannels.rx_count, echannels.tx_count),
+ echannels.combined_count);
+}
+
+static int do_perqueue(struct cmd_context *ctx)
+{
+ __u32 queue_mask[__KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32)] = {0};
+ int i, n_queues = 0;
+
+ if (ctx->argc == 0)
+ exit_bad_args();
+
+ /*
+ * The sub commands will be applied to
+ * all queues if no queue_mask set
+ */
+ if (strncmp(*ctx->argp, "queue_mask", 10)) {
+ n_queues = find_max_num_queues(ctx);
+ if (n_queues < 0) {
+ perror("Cannot get number of queues");
+ return -EFAULT;
+ }
+ for (i = 0; i < n_queues / 32; i++)
+ queue_mask[i] = ~0;
+ queue_mask[i] = (1 << (n_queues - i * 32)) - 1;
+ fprintf(stdout,
+ "The sub commands will be applied to all %d queues\n",
+ n_queues);
+ } else {
+ ctx->argc--;
+ ctx->argp++;
+ n_queues = set_queue_mask(queue_mask, *ctx->argp);
+ if (n_queues < 0) {
+ perror("Invalid queue mask");
+ return n_queues;
+ }
+ ctx->argc--;
+ ctx->argp++;
+ }
+
+ i = find_option(ctx->argc, ctx->argp);
+ if (i < 0)
+ exit_bad_args();
+
+ /* no sub_command support yet */
+
+ return 0;
+}
+
int main(int argc, char **argp)
{
int (*func)(struct cmd_context *);
--
2.20.1
^ permalink raw reply related
* Re: [PATCH net] virtio_net: Account for tx bytes and packets on sending xdp_frames
From: Saeed Mahameed @ 2019-02-06 0:06 UTC (permalink / raw)
To: dsahern@gmail.com, brouer@redhat.com
Cc: thoiland@redhat.com, hawk@kernel.org,
virtualization@lists.linux-foundation.org, borkmann@iogearbox.net,
Tariq Toukan, john.fastabend@gmail.com, mst@redhat.com,
jakub.kicinski@netronome.com, netdev@vger.kernel.org,
jasowang@redhat.com, davem@davemloft.net,
makita.toshiaki@lab.ntt.co.jp
In-Reply-To: <bdcfedd6-465d-4485-e268-25c4ce6b9fcf@gmail.com>
On Mon, 2019-02-04 at 19:13 -0800, David Ahern wrote:
> On 2/4/19 3:53 AM, Jesper Dangaard Brouer wrote:
> > On Sat, 2 Feb 2019 14:27:26 -0700
> > David Ahern <dsahern@gmail.com> wrote:
> >
> > > On 1/31/19 1:15 PM, Jesper Dangaard Brouer wrote:
> > > > > David, Jesper, care to chime in where we ended up in that
> > > > > last thread
> > > > > discussion this?
> > > >
> > > > IHMO packets RX and TX on a device need to be accounted, in
> > > > standard
> > > > counters, regardless of XDP. For XDP RX the packet is counted
> > > > as RX,
> > > > regardless if XDP choose to XDP_DROP. On XDP TX which is via
> > > > XDP_REDIRECT or XDP_TX, the driver that transmit the packet
> > > > need to
> > > > account the packet in a TX counter (this if often delayed to
> > > > DMA TX
> > > > completion handling). We cannot break the expectation that RX
> > > > and TX
> > > > counter are visible to userspace stats tools. XDP should not
> > > > make these
> > > > packets invisible.
> > >
> > > Agreed. What I was pushing on that last thread was Rx, Tx and
> > > dropped
> > > are all accounted by the driver in standard stats. Basically if
> > > the
> > > driver touched it, the driver's counters should indicate that.
> >
> > Sound like we all agree (except with the dropped counter, see
> > below).
> >
> > Do notice that mlx5 driver doesn't do this. It is actually rather
> > confusing to use XDP on mlx5, as when XDP "consume" which include
> > XDP_DROP, XDP_REDIRECT or XDP_TX, then the driver standard stats
> > are
> > not incremented... the packet is invisible to "ifconfig" stat based
> > tools.
>
> mlx5 needs some work. As I recall it still has the bug/panic removing
> xdp programs - at least I don't recall seeing a patch for it.
Only when xdp_redirect to mlx5, and removing the program while redirect
is happening, this is actually due to a lack of synchronization means
between different drivers, we have some ideas to overcome this using a
standard XDP API, or just use a hack in mlx5 driver which i don't like:
https://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git/commit/?h=topic/xdp-redirect-fix&id=a3652d03cc35fd3ad62744986c8ccaca74c9f20c
I will be working on this towards the end of this week.
>
> >
> > > The push back was on dropped packets and whether that counter
> > > should be
> > > bumped on XDP_DROP.
> >
> > My opinion is the XDP_DROP action should NOT increment the drivers
> > drop
> > counter. First of all the "dropped" counter is also use for other
> > stuff, which will confuse that this counter express. Second,
> > choosing
> > XDP_DROP is a policy choice, it still means it was RX-ed at the
> > driver
> > level.
> >
>
> Understood. Hopefully in March I will get some time to come back to
> this
> and propose an idea on what I would like to see - namely, the admin
> has
> a config option at load time to enable driver counters versus custom
> map
> counters. (meaning the operator of the node chooses standard stats
> over
> strict performance.) But of course that means the drivers have the
> code
> to collect those stats.
So bottom line:
1) Driver will count rx packets as rx-ed packets regardless of XDP
decision.
2) Driver should keep track of XDP decisions statistics, report them in
ethtool and in the new API suggested by David. track even (XDP_PASS) ?
Maybe instead of having all drivers track the statistics on their own,
we should move the responsibility to upper layer.
Idea: since we already have rxq_info structure per XDP ring (no false
sharing) and available per xdp_buff we can do:
+++ b/include/linux/filter.h
@@ -651,7 +651,9 @@ static __always_inline u32 bpf_prog_run_xdp(const
struct bpf_prog *prog,
* already takes rcu_read_lock() when fetching the program, so
* it's not necessary here anymore.
*/
- return BPF_PROG_RUN(prog, xdp);
+ u32 ret = BPF_PROG_RUN(prog, xdp);
+ xdp->xdp_rxq_info.stats[ret]++
+ return ret;
}
still we need a way (API) to report the rxq_info to whoever needs to
read current XDP stats
3) Unrelated, In non XDP case, if skb allocation fails or driver fails
to pass the skb up to the stack for somereason, should the driver
increase rx packets ? IMHO the answer should be yes if we want to have
similar behavior between XDP and non XDP cases.
But this could result in netdev->stats.rx_packets + netdev-
>stats.rx_dropped to be more than the actual rx-ed packets, is this
acceptable ?
^ permalink raw reply
* Re: [RFC PATCH iproute2 2/5] act_ct: first import
From: Marcelo Ricardo Leitner @ 2019-02-06 0:09 UTC (permalink / raw)
To: Stephen Hemminger
Cc: Guy Shattah, Aaron Conole, John Hurley, Simon Horman,
Justin Pettit, Gregory Rose, Eelco Chaudron, Flavio Leitner,
Florian Westphal, Jiri Pirko, Rashid Khan, Sushil Kulkarni,
Andy Gospodarek, Roi Dayan, Yossi Kuperman, Or Gerlitz,
Rony Efraim, davem@davemloft.net, netdev
In-Reply-To: <20190205145613.49996454@hermes.lan>
On Tue, Feb 05, 2019 at 02:56:13PM -0800, Stephen Hemminger wrote:
> On Fri, 25 Jan 2019 00:33:30 -0200
> Marcelo Ricardo Leitner <mleitner@redhat.com> wrote:
>
> > +/*
> > + * m_ct.c Connection Tracking target module
> > + *
> > + * This program is free software; you can distribute it and/or
> > + * modify it under the terms of the GNU General Public License
> > + * as published by the Free Software Foundation; either version
> > + * 2 of the License, or (at your option) any later version.
>
> Please just use SPDX for license info not GPL boilerplate.
>
Will do.
Marcelo
^ permalink raw reply
* [PATCH v2 bpf-next 2/2] btf: expose API to work with raw btf data
From: Andrii Nakryiko @ 2019-02-06 0:29 UTC (permalink / raw)
To: songliubraving, yhs, ast, kafai, netdev, daniel, andrii.nakryiko
Cc: Andrii Nakryiko
In-Reply-To: <20190206002949.1915237-1-andriin@fb.com>
This patch exposes two new APIs btf__get_raw_data_size() and
btf__get_raw_data() that allows to get a copy of raw BTF data out of
struct btf. This is useful for external programs that need to manipulate
raw data, e.g., pahole using btf__dedup() to deduplicate BTF type info
and then writing it back to file.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
tools/lib/bpf/btf.c | 10 ++++++++++
tools/lib/bpf/btf.h | 2 ++
tools/lib/bpf/libbpf.map | 2 ++
3 files changed, 14 insertions(+)
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 1c2ba7182400..34bfb3641aac 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -437,6 +437,16 @@ int btf__fd(const struct btf *btf)
return btf->fd;
}
+__u32 btf__get_raw_data_size(const struct btf *btf)
+{
+ return btf->data_size;
+}
+
+void btf__get_raw_data(const struct btf *btf, char *data)
+{
+ memcpy(data, btf->data, btf->data_size);
+}
+
void btf__get_strings(const struct btf *btf, const char **strings,
__u32 *str_len)
{
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index e8410887f93a..d46f680b9416 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -66,6 +66,8 @@ LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf,
LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id);
LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id);
LIBBPF_API int btf__fd(const struct btf *btf);
+LIBBPF_API __u32 btf__get_raw_data_size(const struct btf *btf);
+LIBBPF_API void btf__get_raw_data(const struct btf *btf, char *data);
LIBBPF_API void btf__get_strings(const struct btf *btf, const char **strings,
__u32 *str_len);
LIBBPF_API const char *btf__name_by_offset(const struct btf *btf, __u32 offset);
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index f5372df143f4..0ebbee13a3cd 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -136,6 +136,8 @@ LIBBPF_0.0.2 {
btf__dedup;
btf__get_map_kv_tids;
btf__get_nr_types;
+ btf__get_raw_data;
+ btf__get_raw_data_size;
btf__get_strings;
btf__load;
btf_ext__free;
--
2.17.1
^ permalink raw reply related
* [PATCH v2 bpf-next 0/2] tools/btf: extends libbpf APIs to work with btf w/o kernel
From: Andrii Nakryiko @ 2019-02-06 0:29 UTC (permalink / raw)
To: songliubraving, yhs, ast, kafai, netdev, daniel, andrii.nakryiko
Cc: Andrii Nakryiko
This patchset changes existing btf__new() API call to only load and initialize
struct btf, while exposing new btf__load() API to attempt to load and validate
BTF in kernel. It also adds ability to copy raw BTF data out of struct btf for
further processing by external applications.
This makes utilizing libbpf's APIs that don't require kernel facilities (e.g.,
btf_dedup) simpler and more natural from external application.
v1->v2:
- btf_load() returns just error, not fd
- fix ordering in libbpf.map
Andrii Nakryiko (2):
btf: separate btf creation and loading
btf: expose API to work with raw btf data
tools/lib/bpf/btf.c | 63 +++++++++++++++++++++++++---------------
tools/lib/bpf/btf.h | 3 ++
tools/lib/bpf/libbpf.c | 2 +-
tools/lib/bpf/libbpf.map | 3 ++
4 files changed, 46 insertions(+), 25 deletions(-)
--
2.17.1
^ 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