| Message ID | 0c88607c-1b63-e8b5-8a84-14b63e55e8e2@paneda.se |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [net-next,v3] macvlan: Support for high multicast packet rate | expand |
On Mon, 30 Nov 2020 15:00:43 +0100 Thomas Karlsson wrote: > Background: > Broadcast and multicast packages are enqueued for later processing. > This queue was previously hardcoded to 1000. > > This proved insufficient for handling very high packet rates. > This resulted in packet drops for multicast. > While at the same time unicast worked fine. > > The change: > This patch make the queue length adjustable to accommodate > for environments with very high multicast packet rate. > But still keeps the default value of 1000 unless specified. > > The queue length is specified as a request per macvlan > using the IFLA_MACVLAN_BC_QUEUE_LEN parameter. > > The actual used queue length will then be the maximum of > any macvlan connected to the same port. The actual used > queue length for the port can be retrieved (read only) > by the IFLA_MACVLAN_BC_QUEUE_LEN_USED parameter for verification. > > This will be followed up by a patch to iproute2 > in order to adjust the parameter from userspace. > > Signed-off-by: Thomas Karlsson <thomas.karlsson@paneda.se> Looks good! Minor nits below: > @@ -1218,6 +1220,7 @@ static int macvlan_port_create(struct net_device *dev) > for (i = 0; i < MACVLAN_HASH_SIZE; i++) > INIT_HLIST_HEAD(&port->vlan_source_hash[i]); > > + port->bc_queue_len_used = MACVLAN_DEFAULT_BC_QUEUE_LEN; Should this be inited to 0? Otherwise if the first link asks for lower queue len than the default it will not get set, right? > skb_queue_head_init(&port->bc_queue); > INIT_WORK(&port->bc_work, macvlan_process_broadcast); > > @@ -1486,6 +1489,12 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev, > goto destroy_macvlan_port; > } > > + vlan->bc_queue_len_requested = MACVLAN_DEFAULT_BC_QUEUE_LEN; > + if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN]) > + vlan->bc_queue_len_requested = nla_get_u32(data[IFLA_MACVLAN_BC_QUEUE_LEN]); > + if (vlan->bc_queue_len_requested > port->bc_queue_len_used) > + port->bc_queue_len_used = vlan->bc_queue_len_requested; Or perhaps we should just call update_port_bc_queue_len() here? > err = register_netdevice(dev); > if (err < 0) > goto destroy_macvlan_port; > @@ -1658,6 +1684,8 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = { > [IFLA_MACVLAN_MACADDR] = { .type = NLA_BINARY, .len = MAX_ADDR_LEN }, > [IFLA_MACVLAN_MACADDR_DATA] = { .type = NLA_NESTED }, > [IFLA_MACVLAN_MACADDR_COUNT] = { .type = NLA_U32 }, > + [IFLA_MACVLAN_BC_QUEUE_LEN] = { .type = NLA_U32 }, > + [IFLA_MACVLAN_BC_QUEUE_LEN_USED] = { .type = NLA_U32 }, This is an input policy, so you can set type to NLA_REJECT and you won't have to check if it's set on input. > }; > > int macvlan_link_register(struct rtnl_link_ops *ops) > @@ -1688,6 +1716,18 @@ static struct rtnl_link_ops macvlan_link_ops = { > .priv_size = sizeof(struct macvlan_dev), > }; > > +static void update_port_bc_queue_len(struct macvlan_port *port) > +{ > + struct macvlan_dev *vlan; > + u32 max_bc_queue_len_requested = 0; Please reorder so that the vars are longest line to shortest. > + list_for_each_entry_rcu(vlan, &port->vlans, list) { I don't think you need the _rcu() flavor here, this is always called from the configuration paths holding RTNL lock, right? > + if (vlan->bc_queue_len_requested > max_bc_queue_len_requested) > + max_bc_queue_len_requested = vlan->bc_queue_len_requested; > + } > + port->bc_queue_len_used = max_bc_queue_len_requested; > +} > + > static int macvlan_device_event(struct notifier_block *unused, > unsigned long event, void *ptr) > {
On 2020-12-01 20:11, Jakub Kicinski wrote: > On Mon, 30 Nov 2020 15:00:43 +0100 Thomas Karlsson wrote: >> Background: >> Broadcast and multicast packages are enqueued for later processing. >> This queue was previously hardcoded to 1000. >> >> This proved insufficient for handling very high packet rates. >> This resulted in packet drops for multicast. >> While at the same time unicast worked fine. >> >> The change: >> This patch make the queue length adjustable to accommodate >> for environments with very high multicast packet rate. >> But still keeps the default value of 1000 unless specified. >> >> The queue length is specified as a request per macvlan >> using the IFLA_MACVLAN_BC_QUEUE_LEN parameter. >> >> The actual used queue length will then be the maximum of >> any macvlan connected to the same port. The actual used >> queue length for the port can be retrieved (read only) >> by the IFLA_MACVLAN_BC_QUEUE_LEN_USED parameter for verification. >> >> This will be followed up by a patch to iproute2 >> in order to adjust the parameter from userspace. >> >> Signed-off-by: Thomas Karlsson <thomas.karlsson@paneda.se> > > Looks good! Minor nits below: :) > >> @@ -1218,6 +1220,7 @@ static int macvlan_port_create(struct net_device *dev) >> for (i = 0; i < MACVLAN_HASH_SIZE; i++) >> INIT_HLIST_HEAD(&port->vlan_source_hash[i]); >> >> + port->bc_queue_len_used = MACVLAN_DEFAULT_BC_QUEUE_LEN; > > Should this be inited to 0? Otherwise if the first link asks for lower > queue len than the default it will not get set, right? Indeed, looks you are right, see also below >> skb_queue_head_init(&port->bc_queue); >> INIT_WORK(&port->bc_work, macvlan_process_broadcast); >> >> @@ -1486,6 +1489,12 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev, >> goto destroy_macvlan_port; >> } >> >> + vlan->bc_queue_len_requested = MACVLAN_DEFAULT_BC_QUEUE_LEN; >> + if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN]) >> + vlan->bc_queue_len_requested = nla_get_u32(data[IFLA_MACVLAN_BC_QUEUE_LEN]); >> + if (vlan->bc_queue_len_requested > port->bc_queue_len_used) >> + port->bc_queue_len_used = vlan->bc_queue_len_requested; > > Or perhaps we should just call update_port_bc_queue_len() here? That would also have prevented the above bug... So yes, I think that is better to keep the logic only in one place. I'll change to that. >> err = register_netdevice(dev); >> if (err < 0) >> goto destroy_macvlan_port; > >> @@ -1658,6 +1684,8 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = { >> [IFLA_MACVLAN_MACADDR] = { .type = NLA_BINARY, .len = MAX_ADDR_LEN }, >> [IFLA_MACVLAN_MACADDR_DATA] = { .type = NLA_NESTED }, >> [IFLA_MACVLAN_MACADDR_COUNT] = { .type = NLA_U32 }, >> + [IFLA_MACVLAN_BC_QUEUE_LEN] = { .type = NLA_U32 }, >> + [IFLA_MACVLAN_BC_QUEUE_LEN_USED] = { .type = NLA_U32 }, > > This is an input policy, so you can set type to NLA_REJECT and you > won't have to check if it's set on input. > Great! >> }; >> >> int macvlan_link_register(struct rtnl_link_ops *ops) >> @@ -1688,6 +1716,18 @@ static struct rtnl_link_ops macvlan_link_ops = { >> .priv_size = sizeof(struct macvlan_dev), >> }; >> >> +static void update_port_bc_queue_len(struct macvlan_port *port) >> +{ >> + struct macvlan_dev *vlan; >> + u32 max_bc_queue_len_requested = 0; > > Please reorder so that the vars are longest line to shortest. > got it >> + list_for_each_entry_rcu(vlan, &port->vlans, list) { > > I don't think you need the _rcu() flavor here, this is always called > from the configuration paths holding RTNL lock, right? > To be honest, what to use/not to use when traversing the list was what caused me the most doubt/trouble of the patch :) I sort of assumed that there must be some outer synchronisation that prevented two or more concurrent calls to new/delte/change link. but wasn't sure how and where that synchonisation took place. Now that I have googled RTLN lock I understand that part much better. The main reason I went with _rcu was because the existing code is using list_del_rcu and list_add_tail_rcu when modifying the list as well as _rcu when accessing/traversing (in some places). So I figured if they needed the _rcu variants I too would need that. But from a closer inspection I think in that situation it is only needed because the list is accessed from for example macvlan_handle_frame (obviously not protected by the RTLN lock) using _rcu version and under the rcu_read_lock as protection. So then it must also be updated with _rcu in all places of course. Even if all the updates are done under the RTNL lock. This was a long ramble :) But thanks, I think I understand the synchronisation mechanism in the kernel a bit better now! As I'm only calling my function from the netlink configuration functions under RTLN lock It should be safe to drop the _rcu version as you say, because the list is only modified in those functions too. Great! >> + if (vlan->bc_queue_len_requested > max_bc_queue_len_requested) >> + max_bc_queue_len_requested = vlan->bc_queue_len_requested; >> + } >> + port->bc_queue_len_used = max_bc_queue_len_requested; >> +} >> + >> static int macvlan_device_event(struct notifier_block *unused, >> unsigned long event, void *ptr) >> { I also noticed I got a few line length warnings in patchworks but none when I ran the ./scrips/checkpatch.pl locally. So is the net tree using strict 80 chars? I would prefer not to introduce extra line breaks on those lines as I think it will hurt readability but of course I will if needed. I will publish a v4 later today. /Thomas
On Wed, 2 Dec 2020 12:28:47 +0100 Thomas Karlsson wrote: > >> + if (vlan->bc_queue_len_requested > max_bc_queue_len_requested) > >> + max_bc_queue_len_requested = vlan->bc_queue_len_requested; > >> + } > >> + port->bc_queue_len_used = max_bc_queue_len_requested; > >> +} > >> + > >> static int macvlan_device_event(struct notifier_block *unused, > >> unsigned long event, void *ptr) > >> { > > I also noticed I got a few line length warnings in patchworks but none when I ran the ./scrips/checkpatch.pl > locally. So is the net tree using strict 80 chars? I would prefer not to introduce extra line breaks > on those lines as I think it will hurt readability but of course I will if needed. I run checkpatch with --max-line-length=80, I think the 80 char limitation is quite reasonable and leads to more readable code. In your case I'd do s/bc_queue_len_requested/bc_queue_len_req/. But it's up to you.
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index d9b6c44a5911..b8197761248e 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -35,7 +35,7 @@ #define MACVLAN_HASH_BITS 8 #define MACVLAN_HASH_SIZE (1<<MACVLAN_HASH_BITS) -#define MACVLAN_BC_QUEUE_LEN 1000 +#define MACVLAN_DEFAULT_BC_QUEUE_LEN 1000 #define MACVLAN_F_PASSTHRU 1 #define MACVLAN_F_ADDRCHANGE 2 @@ -46,6 +46,7 @@ struct macvlan_port { struct list_head vlans; struct sk_buff_head bc_queue; struct work_struct bc_work; + u32 bc_queue_len_used; u32 flags; int count; struct hlist_head vlan_source_hash[MACVLAN_HASH_SIZE]; @@ -67,6 +68,7 @@ struct macvlan_skb_cb { #define MACVLAN_SKB_CB(__skb) ((struct macvlan_skb_cb *)&((__skb)->cb[0])) static void macvlan_port_destroy(struct net_device *dev); +static void update_port_bc_queue_len(struct macvlan_port *port); static inline bool macvlan_passthru(const struct macvlan_port *port) { @@ -354,7 +356,7 @@ static void macvlan_broadcast_enqueue(struct macvlan_port *port, MACVLAN_SKB_CB(nskb)->src = src; spin_lock(&port->bc_queue.lock); - if (skb_queue_len(&port->bc_queue) < MACVLAN_BC_QUEUE_LEN) { + if (skb_queue_len(&port->bc_queue) < port->bc_queue_len_used) { if (src) dev_hold(src->dev); __skb_queue_tail(&port->bc_queue, nskb); @@ -1218,6 +1220,7 @@ static int macvlan_port_create(struct net_device *dev) for (i = 0; i < MACVLAN_HASH_SIZE; i++) INIT_HLIST_HEAD(&port->vlan_source_hash[i]); + port->bc_queue_len_used = MACVLAN_DEFAULT_BC_QUEUE_LEN; skb_queue_head_init(&port->bc_queue); INIT_WORK(&port->bc_work, macvlan_process_broadcast); @@ -1486,6 +1489,12 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev, goto destroy_macvlan_port; } + vlan->bc_queue_len_requested = MACVLAN_DEFAULT_BC_QUEUE_LEN; + if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN]) + vlan->bc_queue_len_requested = nla_get_u32(data[IFLA_MACVLAN_BC_QUEUE_LEN]); + if (vlan->bc_queue_len_requested > port->bc_queue_len_used) + port->bc_queue_len_used = vlan->bc_queue_len_requested; + err = register_netdevice(dev); if (err < 0) goto destroy_macvlan_port; @@ -1529,6 +1538,7 @@ void macvlan_dellink(struct net_device *dev, struct list_head *head) if (vlan->mode == MACVLAN_MODE_SOURCE) macvlan_flush_sources(vlan->port, vlan); list_del_rcu(&vlan->list); + update_port_bc_queue_len(vlan->port); unregister_netdevice_queue(dev, head); netdev_upper_dev_unlink(vlan->lowerdev, dev); } @@ -1572,6 +1582,15 @@ static int macvlan_changelink(struct net_device *dev, } vlan->flags = flags; } + + if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN_USED]) + return -EINVAL; /* Trying to set a read only attribute */ + + if (data && data[IFLA_MACVLAN_BC_QUEUE_LEN]) { + vlan->bc_queue_len_requested = nla_get_u32(data[IFLA_MACVLAN_BC_QUEUE_LEN]); + update_port_bc_queue_len(vlan->port); + } + if (set_mode) vlan->mode = mode; if (data && data[IFLA_MACVLAN_MACADDR_MODE]) { @@ -1602,6 +1621,8 @@ static size_t macvlan_get_size(const struct net_device *dev) + nla_total_size(2) /* IFLA_MACVLAN_FLAGS */ + nla_total_size(4) /* IFLA_MACVLAN_MACADDR_COUNT */ + macvlan_get_size_mac(vlan) /* IFLA_MACVLAN_MACADDR */ + + nla_total_size(4) /* IFLA_MACVLAN_BC_QUEUE_LEN */ + + nla_total_size(4) /* IFLA_MACVLAN_BC_QUEUE_LEN_USED */ ); } @@ -1625,6 +1646,7 @@ static int macvlan_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct macvlan_dev *vlan = netdev_priv(dev); + struct macvlan_port *port = vlan->port; int i; struct nlattr *nest; @@ -1645,6 +1667,10 @@ static int macvlan_fill_info(struct sk_buff *skb, } nla_nest_end(skb, nest); } + if (nla_put_u32(skb, IFLA_MACVLAN_BC_QUEUE_LEN, vlan->bc_queue_len_requested)) + goto nla_put_failure; + if (nla_put_u32(skb, IFLA_MACVLAN_BC_QUEUE_LEN_USED, port->bc_queue_len_used)) + goto nla_put_failure; return 0; nla_put_failure: @@ -1658,6 +1684,8 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = { [IFLA_MACVLAN_MACADDR] = { .type = NLA_BINARY, .len = MAX_ADDR_LEN }, [IFLA_MACVLAN_MACADDR_DATA] = { .type = NLA_NESTED }, [IFLA_MACVLAN_MACADDR_COUNT] = { .type = NLA_U32 }, + [IFLA_MACVLAN_BC_QUEUE_LEN] = { .type = NLA_U32 }, + [IFLA_MACVLAN_BC_QUEUE_LEN_USED] = { .type = NLA_U32 }, }; int macvlan_link_register(struct rtnl_link_ops *ops) @@ -1688,6 +1716,18 @@ static struct rtnl_link_ops macvlan_link_ops = { .priv_size = sizeof(struct macvlan_dev), }; +static void update_port_bc_queue_len(struct macvlan_port *port) +{ + struct macvlan_dev *vlan; + u32 max_bc_queue_len_requested = 0; + + list_for_each_entry_rcu(vlan, &port->vlans, list) { + if (vlan->bc_queue_len_requested > max_bc_queue_len_requested) + max_bc_queue_len_requested = vlan->bc_queue_len_requested; + } + port->bc_queue_len_used = max_bc_queue_len_requested; +} + static int macvlan_device_event(struct notifier_block *unused, unsigned long event, void *ptr) { diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index a367ead4bf4b..c3923fdbe1f0 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -30,6 +30,7 @@ struct macvlan_dev { enum macvlan_mode mode; u16 flags; unsigned int macaddr_count; + u32 bc_queue_len_requested; #ifdef CONFIG_NET_POLL_CONTROLLER struct netpoll *netpoll; #endif diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index c4b23f06f69e..874cc12a34d9 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -588,6 +588,8 @@ enum { IFLA_MACVLAN_MACADDR, IFLA_MACVLAN_MACADDR_DATA, IFLA_MACVLAN_MACADDR_COUNT, + IFLA_MACVLAN_BC_QUEUE_LEN, + IFLA_MACVLAN_BC_QUEUE_LEN_USED, __IFLA_MACVLAN_MAX, }; diff --git a/tools/include/uapi/linux/if_link.h b/tools/include/uapi/linux/if_link.h index 781e482dc499..d208b2af697f 100644 --- a/tools/include/uapi/linux/if_link.h +++ b/tools/include/uapi/linux/if_link.h @@ -409,6 +409,8 @@ enum { IFLA_MACVLAN_MACADDR, IFLA_MACVLAN_MACADDR_DATA, IFLA_MACVLAN_MACADDR_COUNT, + IFLA_MACVLAN_BC_QUEUE_LEN, + IFLA_MACVLAN_BC_QUEUE_LEN_USED, __IFLA_MACVLAN_MAX, };
Background: Broadcast and multicast packages are enqueued for later processing. This queue was previously hardcoded to 1000. This proved insufficient for handling very high packet rates. This resulted in packet drops for multicast. While at the same time unicast worked fine. The change: This patch make the queue length adjustable to accommodate for environments with very high multicast packet rate. But still keeps the default value of 1000 unless specified. The queue length is specified as a request per macvlan using the IFLA_MACVLAN_BC_QUEUE_LEN parameter. The actual used queue length will then be the maximum of any macvlan connected to the same port. The actual used queue length for the port can be retrieved (read only) by the IFLA_MACVLAN_BC_QUEUE_LEN_USED parameter for verification. This will be followed up by a patch to iproute2 in order to adjust the parameter from userspace. Signed-off-by: Thomas Karlsson <thomas.karlsson@paneda.se> --- v3 switched to using netlink attributes v1/2 used module_param drivers/net/macvlan.c | 44 ++++++++++++++++++++++++++++-- include/linux/if_macvlan.h | 1 + include/uapi/linux/if_link.h | 2 ++ tools/include/uapi/linux/if_link.h | 2 ++ 4 files changed, 47 insertions(+), 2 deletions(-)