From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-04 22:00:59
For a switch to offload insertion of HSR/PRP tags, frames must not be
sent to the CPU facing switch port with a tag. Generate supervision frames
(eth type ETH_P_PRP) without HSR v1 (ETH_P_HSR)/PRP tag and rely on
create_tagged_frame which inserts it later. This will allow skipping the
tag insertion for all outgoing frames in the future which is required for
HSR v1/PRP tag insertions to be offloaded.
HSR v0 supervision frames always contain tag information so insertion of
the tag can't be offloaded. IEC 62439-3 Ed.2.0 (HSR v1) specifically
notes that this was changed since v0 to allow offloading.
Signed-off-by: George McCollister <george.mccollister@gmail.com>
---
net/hsr/hsr_device.c | 32 ++++----------------------------
net/hsr/hsr_forward.c | 6 +++++-
2 files changed, 9 insertions(+), 29 deletions(-)
@@ -454,7 +454,11 @@ static void handle_std_frame(struct sk_buff *skb,voidhsr_fill_frame_info(__be16proto,structsk_buff*skb,structhsr_frame_info*frame){-if(proto==htons(ETH_P_PRP)||+structhsr_port*port=frame->port_rcv;+structhsr_priv*hsr=port->hsr;++/* HSRv0 supervisory frames double as a tag so treat them as tagged. */+if((!hsr->prot_version&&proto==htons(ETH_P_PRP))||proto==htons(ETH_P_HSR)){/* HSR tagged frame :- Data or Supervision */frame->skb_std=NULL;
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-04 22:00:59
Add support for offloading of HSR/PRP (IEC 62439-3) tag insertion
tag removal, duplicate generation and forwarding.
For HSR, insertion involves the switch adding a 6 byte HSR header after
the 14 byte Ethernet header. For PRP it adds a 6 byte trailer.
Tag removal involves automatically stripping the HSR/PRP header/trailer
in the switch. This is possible when the switch also performs auto
deduplication using the HSR/PRP header/trailer (making it no longer
required).
Forwarding involves automatically forwarding between redundant ports in
an HSR. This is crucial because delay is accumulated as a frame passes
through each node in the ring.
Duplication involves the switch automatically sending a single frame
from the CPU port to both redundant ports. This is required because the
inserted HSR/PRP header/trailer must contain the same sequence number
on the frames sent out both redundant ports.
Export is_hsr_master so DSA can tell them apart from other devices in
dsa_slave_changeupper.
Signed-off-by: George McCollister <george.mccollister@gmail.com>
---
Documentation/networking/netdev-features.rst | 21 +++++++++++++++++++++
include/linux/if_hsr.h | 27 +++++++++++++++++++++++++++
include/linux/netdev_features.h | 9 +++++++++
net/ethtool/common.c | 4 ++++
net/hsr/hsr_device.c | 14 +++-----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 27 ++++++++++++++++++++++++---
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 ++
net/hsr/hsr_main.c | 11 +++++++++++
net/hsr/hsr_main.h | 8 +-------
net/hsr/hsr_slave.c | 10 ++++++----
12 files changed, 109 insertions(+), 26 deletions(-)
create mode 100644 include/linux/if_hsr.h
@@ -182,3 +182,24 @@ stricter than Hardware LRO. A packet stream merged by Hardware GRO must be re-segmentable by GSO or TSO back to the exact original packet stream. Hardware GRO is dependent on RXCSUM since every packet successfully merged by hardware must also have the checksum verified by hardware.++* hsr-tag-ins-offload++This should be set for devices which insert an HSR (High-availability Seamless+Redundancy) or PRP (Parallel Redundancy Protocol) tag automatically.++* hsr-tag-rm-offload++This should be set for devices which remove HSR (High-availability Seamless+Redundancy) or PRP (Parallel Redundancy Protocol) tags automatically.++* hsr-fwd-offload++This should be set for devices which forward HSR (High-availability Seamless+Redundancy) frames from one port to another in hardware.++* hsr-dup-offload++This should be set for devices which duplicate outgoing HSR (High-availability+Seamless Redundancy) or PRP (Parallel Redundancy Protocol) tags automatically+frames in hardware.
@@ -159,6 +164,10 @@ enum {#define NETIF_F_GSO_FRAGLIST __NETIF_F(GSO_FRAGLIST)#define NETIF_F_HW_MACSEC __NETIF_F(HW_MACSEC)#define NETIF_F_GRO_UDP_FWD __NETIF_F(GRO_UDP_FWD)+#define NETIF_F_HW_HSR_TAG_INS __NETIF_F(HW_HSR_TAG_INS)+#define NETIF_F_HW_HSR_TAG_RM __NETIF_F(HW_HSR_TAG_RM)+#define NETIF_F_HW_HSR_FWD __NETIF_F(HW_HSR_FWD)+#define NETIF_F_HW_HSR_DUP __NETIF_F(HW_HSR_DUP)/* Finds the next feature with the highest number of the range of start till 0.*/
@@ -465,10 +466,11 @@ void hsr_dev_setup(struct net_device *dev)/* Return true if dev is a HSR master; return false otherwise.*/-inlineboolis_hsr_master(structnet_device*dev)+boolis_hsr_master(structnet_device*dev){return(dev->netdev_ops->ndo_start_xmit==hsr_dev_xmit);}+EXPORT_SYMBOL(is_hsr_master);/* Default multicast address for HSR Supervision frames */staticconstunsignedchardef_multicast_addr[ETH_ALEN]__aligned(2)={
@@ -521,16 +523,6 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2],hsr->prot_version=protocol_version;-/* FIXME: should I modify the value of these?-*-*-hsr_dev->flags-i.e.-*IFF_MASTER/SLAVE?-*-hsr_dev->priv_flags-i.e.-*IFF_EBRIDGE?-*IFF_TX_SKB_SHARING?-*IFF_HSR_MASTER/SLAVE?-*/-/* Make sure the 1st call to netif_carrier_on() gets through */netif_carrier_off(hsr_dev);
@@ -247,6 +247,8 @@ struct sk_buff *hsr_create_tagged_frame(struct hsr_frame_info *frame,/* set the lane id properly */hsr_set_path_id(hsr_ethhdr,port);returnskb_clone(frame->skb_hsr,GFP_ATOMIC);+}elseif(port->dev->features&NETIF_F_HW_HSR_TAG_INS){+returnskb_clone(frame->skb_std,GFP_ATOMIC);}/* Create the new skb with enough headroom to fit the HSR tag */
@@ -372,6 +385,12 @@ static void hsr_forward_do(struct hsr_frame_info *frame)if(port->type!=HSR_PT_MASTER&&frame->is_local_exclusive)continue;+/* If hardware duplicate generation is enabled, only send out+*oneport.+*/+if((port->dev->features&NETIF_F_HW_HSR_DUP)&&sent)+continue;+/* Don't send frame over port where it has been sent before.*AlsofroSAN,thisshouldn'tbedone.*/
@@ -13,6 +13,7 @@#include<linux/netdevice.h>#include<linux/list.h>#include<linux/if_vlan.h>+#include<linux/if_hsr.h>/* Time constants as specified in the HSR specification (IEC-62439-3 2010)*Table8.
@@ -171,13 +172,6 @@ struct hsr_port {enumhsr_port_typetype;};-/* used by driver internally to differentiate various protocols */-enumhsr_version{-HSR_V0=0,-HSR_V1,-PRP_V1,-};-structhsr_frame_info;structhsr_node;
@@ -48,12 +48,14 @@ static rx_handler_result_t hsr_handle_frame(struct sk_buff **pskb)gotofinish_consume;}-/* For HSR, only tagged frames are expected, but for PRP-*therecouldbenontaggedframesaswellfromSingle-*attachednodes(SANs).+/* For HSR, only tagged frames are expected (unless the device offloads+*HSRtagremoval),butforPRPtherecouldbenontaggedframesas+*wellfromSingleattachednodes(SANs).*/protocol=eth_hdr(skb)->h_proto;-if(hsr->proto_ops->invalid_dan_ingress_frame&&++if(!(port->dev->features&NETIF_F_HW_HSR_TAG_RM)&&+hsr->proto_ops->invalid_dan_ingress_frame&&hsr->proto_ops->invalid_dan_ingress_frame(protocol))gotofinish_pass;
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-04 22:01:07
Add support for offloading of HSR/PRP (IEC 62439-3) tag insertion
tag removal, duplicate generation and forwarding on DSA switches.
Add DSA_NOTIFIER_HSR_JOIN and DSA_NOTIFIER_HSR_LEAVE which trigger calls
to .port_hsr_join and .port_hsr_leave in the DSA driver for the switch.
The DSA switch driver should then set netdev feature flags for the
HSR/PRP operation that it offloads.
NETIF_F_HW_HSR_TAG_INS
NETIF_F_HW_HSR_TAG_RM
NETIF_F_HW_HSR_FWD
NETIF_F_HW_HSR_DUP
Signed-off-by: George McCollister <george.mccollister@gmail.com>
---
include/net/dsa.h | 13 +++++++++++++
net/dsa/dsa_priv.h | 11 +++++++++++
net/dsa/port.c | 34 ++++++++++++++++++++++++++++++++++
net/dsa/slave.c | 14 ++++++++++++++
net/dsa/switch.c | 24 ++++++++++++++++++++++++
5 files changed, 96 insertions(+)
@@ -100,6 +102,13 @@ struct dsa_switchdev_event_work {u16vid;};+/* DSA_NOTIFIER_HSR_* */+structdsa_notifier_hsr_info{+structnet_device*hsr;+intsw_index;+intport;+};+structdsa_slave_priv{/* Copy of CPU port xmit for faster access in slave transmit hot path */structsk_buff*(*xmit)(structsk_buff*skb,
@@ -183,6 +192,8 @@ int dsa_port_vlan_del(struct dsa_port *dp,conststructswitchdev_obj_port_vlan*vlan);intdsa_port_link_register_of(structdsa_port*dp);voiddsa_port_link_unregister_of(structdsa_port*dp);+intdsa_port_hsr_join(structdsa_port*dp,structnet_device*hsr);+voiddsa_port_hsr_leave(structdsa_port*dp,structnet_device*hsr);externconststructphylink_mac_opsdsa_port_phylink_mac_ops;staticinlinebooldsa_port_offloads_netdev(structdsa_port*dp,
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-04 22:01:09
Add offloading for HSR/PRP (IEC 62439-3) tag insertion, tag removal
forwarding and duplication supported by the xrs7000 series switches.
Only HSR v1 and PRP v1 are supported by the xrs7000 series switches (HSR
v0 is not).
Signed-off-by: George McCollister <george.mccollister@gmail.com>
---
drivers/net/dsa/xrs700x/xrs700x.c | 106 ++++++++++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
net/dsa/tag_xrs700x.c | 7 ++-
3 files changed, 117 insertions(+), 1 deletion(-)
@@ -496,6 +498,108 @@ static void xrs700x_bridge_leave(struct dsa_switch *ds, int port,xrs700x_bridge_common(ds,port,bridge,false);}+staticintxrs700x_hsr_join(structdsa_switch*ds,intport,+structnet_device*hsr)+{+unsignedintval=XRS_HSR_CFG_HSR_PRP;+structdsa_port*partner=NULL,*dp;+structxrs700x*priv=ds->priv;+structnet_device*slave;+enumhsr_versionver;+intret;++ret=hsr_get_version(hsr,&ver);+if(ret)+returnret;++if(ver==HSR_V1)+val|=XRS_HSR_CFG_HSR;+elseif(ver==PRP_V1)+val|=XRS_HSR_CFG_PRP;+else+return-EOPNOTSUPP;++dsa_hsr_foreach_port(dp,ds,hsr){+partner=dp;+}++/* We can't enable redundancy on the switch until both+*redundantportshavesignedup.+*/+if(!partner)+return0;++regmap_fields_write(priv->ps_forward,partner->index,+XRS_PORT_DISABLED);+regmap_fields_write(priv->ps_forward,port,XRS_PORT_DISABLED);++regmap_write(priv->regmap,XRS_HSR_CFG(partner->index),+val|XRS_HSR_CFG_LANID_A);+regmap_write(priv->regmap,XRS_HSR_CFG(port),+val|XRS_HSR_CFG_LANID_B);++/* Clear bits for both redundant ports (HSR only) and the CPU port to+*enableforwarding.+*/+val=GENMASK(ds->num_ports-1,0);+if(ver==HSR_V1){+val&=~BIT(partner->index);+val&=~BIT(port);+}+val&=~BIT(dsa_upstream_port(ds,port));+regmap_write(priv->regmap,XRS_PORT_FWD_MASK(partner->index),val);+regmap_write(priv->regmap,XRS_PORT_FWD_MASK(port),val);++regmap_fields_write(priv->ps_forward,partner->index,+XRS_PORT_FORWARDING);+regmap_fields_write(priv->ps_forward,port,XRS_PORT_FORWARDING);++slave=dsa_to_port(ds,port)->slave;++slave->features|=NETIF_F_HW_HSR_TAG_INS|NETIF_F_HW_HSR_TAG_RM|+NETIF_F_HW_HSR_FWD|NETIF_F_HW_HSR_DUP;++return0;+}++staticvoidxrs700x_hsr_leave(structdsa_switch*ds,intport,+structnet_device*hsr)+{+structdsa_port*partner=NULL,*dp;+structxrs700x*priv=ds->priv;+structnet_device*slave;+unsignedintval;++dsa_hsr_foreach_port(dp,ds,hsr){+partner=dp;+}++if(!partner)+return;++regmap_fields_write(priv->ps_forward,partner->index,+XRS_PORT_DISABLED);+regmap_fields_write(priv->ps_forward,port,XRS_PORT_DISABLED);++regmap_write(priv->regmap,XRS_HSR_CFG(partner->index),0);+regmap_write(priv->regmap,XRS_HSR_CFG(port),0);++/* Clear bit for the CPU port to enable forwarding. */+val=GENMASK(ds->num_ports-1,0);+val&=~BIT(dsa_upstream_port(ds,port));+regmap_write(priv->regmap,XRS_PORT_FWD_MASK(partner->index),val);+regmap_write(priv->regmap,XRS_PORT_FWD_MASK(port),val);++regmap_fields_write(priv->ps_forward,partner->index,+XRS_PORT_FORWARDING);+regmap_fields_write(priv->ps_forward,port,XRS_PORT_FORWARDING);++slave=dsa_to_port(ds,port)->slave;++slave->features&=~(NETIF_F_HW_HSR_TAG_INS|NETIF_F_HW_HSR_TAG_RM|+NETIF_F_HW_HSR_FWD|NETIF_F_HW_HSR_DUP);+}+staticconststructdsa_switch_opsxrs700x_ops={.get_tag_protocol=xrs700x_get_tag_protocol,.setup=xrs700x_setup,
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-06 23:54:53
On Thu, Feb 04, 2021 at 03:59:26PM -0600, George McCollister wrote:
+static int xrs700x_hsr_join(struct dsa_switch *ds, int port,
+ struct net_device *hsr)
+{
+ unsigned int val = XRS_HSR_CFG_HSR_PRP;
+ struct dsa_port *partner = NULL, *dp;
+ struct xrs700x *priv = ds->priv;
+ struct net_device *slave;
+ enum hsr_version ver;
+ int ret;
+
+ ret = hsr_get_version(hsr, &ver);
+ if (ret)
+ return ret;
+
+ if (ver == HSR_V1)
+ val |= XRS_HSR_CFG_HSR;
+ else if (ver == PRP_V1)
+ val |= XRS_HSR_CFG_PRP;
+ else
+ return -EOPNOTSUPP;
+
+ dsa_hsr_foreach_port(dp, ds, hsr) {
+ partner = dp;
+ }
+
+ /* We can't enable redundancy on the switch until both
+ * redundant ports have signed up.
+ */
+ if (!partner)
+ return 0;
+
+ regmap_fields_write(priv->ps_forward, partner->index,
+ XRS_PORT_DISABLED);
+ regmap_fields_write(priv->ps_forward, port, XRS_PORT_DISABLED);
+
+ regmap_write(priv->regmap, XRS_HSR_CFG(partner->index),
+ val | XRS_HSR_CFG_LANID_A);
+ regmap_write(priv->regmap, XRS_HSR_CFG(port),
+ val | XRS_HSR_CFG_LANID_B);
+
+ /* Clear bits for both redundant ports (HSR only) and the CPU port to
+ * enable forwarding.
+ */
+ val = GENMASK(ds->num_ports - 1, 0);
+ if (ver == HSR_V1) {
+ val &= ~BIT(partner->index);
+ val &= ~BIT(port);
+ }
+ val &= ~BIT(dsa_upstream_port(ds, port));
+ regmap_write(priv->regmap, XRS_PORT_FWD_MASK(partner->index), val);
+ regmap_write(priv->regmap, XRS_PORT_FWD_MASK(port), val);
+
+ regmap_fields_write(priv->ps_forward, partner->index,
+ XRS_PORT_FORWARDING);
+ regmap_fields_write(priv->ps_forward, port, XRS_PORT_FORWARDING);
+
+ slave = dsa_to_port(ds, port)->slave;
+
+ slave->features |= NETIF_F_HW_HSR_TAG_INS | NETIF_F_HW_HSR_TAG_RM |
+ NETIF_F_HW_HSR_FWD | NETIF_F_HW_HSR_DUP;
+
+ return 0;
+}
Is it deliberate that only one slave HSR/PRP port will have the offload
ethtool features set? If yes, then I find that a bit odd from a user
point of view.
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-07 01:27:55
On Thu, Feb 04, 2021 at 03:59:23PM -0600, George McCollister wrote:
quoted hunk
@@ -289,23 +286,12 @@ static void send_hsr_supervision_frame(struct hsr_port *master, hsr->announce_count++; }- if (!hsr->prot_version)- proto = ETH_P_PRP;- else- proto = ETH_P_HSR;-- skb = hsr_init_skb(master, proto);+ skb = hsr_init_skb(master, ETH_P_PRP); if (!skb) { WARN_ONCE(1, "HSR: Could not send supervision frame\n"); return; }
I wonder why you aren't even more aggressive, just remove the proto
argument from hsr_init_skb and delay setting skb->protocol to
hsr_fill_tag or the whereabouts. This is probably also more correct
since as far as I can see, nobody is updating the skb->proto of
supervision frames from HSR v0 to v1 after your change.
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-08 14:48:50
On Sat, Feb 6, 2021 at 5:53 PM Vladimir Oltean [off-list ref] wrote:
On Thu, Feb 04, 2021 at 03:59:26PM -0600, George McCollister wrote:
quoted
+static int xrs700x_hsr_join(struct dsa_switch *ds, int port,
+ struct net_device *hsr)
+{
+ unsigned int val = XRS_HSR_CFG_HSR_PRP;
+ struct dsa_port *partner = NULL, *dp;
+ struct xrs700x *priv = ds->priv;
+ struct net_device *slave;
+ enum hsr_version ver;
+ int ret;
+
+ ret = hsr_get_version(hsr, &ver);
+ if (ret)
+ return ret;
+
+ if (ver == HSR_V1)
+ val |= XRS_HSR_CFG_HSR;
+ else if (ver == PRP_V1)
+ val |= XRS_HSR_CFG_PRP;
+ else
+ return -EOPNOTSUPP;
+
+ dsa_hsr_foreach_port(dp, ds, hsr) {
+ partner = dp;
+ }
+
+ /* We can't enable redundancy on the switch until both
+ * redundant ports have signed up.
+ */
+ if (!partner)
+ return 0;
+
+ regmap_fields_write(priv->ps_forward, partner->index,
+ XRS_PORT_DISABLED);
+ regmap_fields_write(priv->ps_forward, port, XRS_PORT_DISABLED);
+
+ regmap_write(priv->regmap, XRS_HSR_CFG(partner->index),
+ val | XRS_HSR_CFG_LANID_A);
+ regmap_write(priv->regmap, XRS_HSR_CFG(port),
+ val | XRS_HSR_CFG_LANID_B);
+
+ /* Clear bits for both redundant ports (HSR only) and the CPU port to
+ * enable forwarding.
+ */
+ val = GENMASK(ds->num_ports - 1, 0);
+ if (ver == HSR_V1) {
+ val &= ~BIT(partner->index);
+ val &= ~BIT(port);
+ }
+ val &= ~BIT(dsa_upstream_port(ds, port));
+ regmap_write(priv->regmap, XRS_PORT_FWD_MASK(partner->index), val);
+ regmap_write(priv->regmap, XRS_PORT_FWD_MASK(port), val);
+
+ regmap_fields_write(priv->ps_forward, partner->index,
+ XRS_PORT_FORWARDING);
+ regmap_fields_write(priv->ps_forward, port, XRS_PORT_FORWARDING);
+
+ slave = dsa_to_port(ds, port)->slave;
+
+ slave->features |= NETIF_F_HW_HSR_TAG_INS | NETIF_F_HW_HSR_TAG_RM |
+ NETIF_F_HW_HSR_FWD | NETIF_F_HW_HSR_DUP;
+
+ return 0;
+}
Is it deliberate that only one slave HSR/PRP port will have the offload
ethtool features set? If yes, then I find that a bit odd from a user
point of view.
No. Good catch. This is a mistake I introduced when I added the code
for finding the partner. Originally for testing I had hacks that hard
coded the ports used and reconfigured HSR for each join.
If you return zero, the software fallback is never going to kick in.
For join and leave? How is this not a problem for the bridge and lag
functions? They work the same way don't they? I figured it would be
safe to follow what they were doing.
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-08 17:34:44
On Sat, Feb 6, 2021 at 7:26 PM Vladimir Oltean [off-list ref] wrote:
On Thu, Feb 04, 2021 at 03:59:23PM -0600, George McCollister wrote:
quoted
@@ -289,23 +286,12 @@ static void send_hsr_supervision_frame(struct hsr_port *master, hsr->announce_count++; }- if (!hsr->prot_version)- proto = ETH_P_PRP;- else- proto = ETH_P_HSR;-- skb = hsr_init_skb(master, proto);+ skb = hsr_init_skb(master, ETH_P_PRP); if (!skb) { WARN_ONCE(1, "HSR: Could not send supervision frame\n"); return; }
I wonder why you aren't even more aggressive, just remove the proto
argument from hsr_init_skb and delay setting skb->protocol to
hsr_fill_tag or the whereabouts. This is probably also more correct
since as far as I can see, nobody is updating the skb->proto of
supervision frames from HSR v0 to v1 after your change.
On Thu, Feb 04, 2021 at 15:59, George McCollister [off-list ref] wrote:
Add support for offloading HSR/PRP (IEC 62439-3) tag insertion, tag
removal, forwarding and duplication on DSA switches.
This series adds offloading to the xrs700x DSA driver.
Changes since RFC:
* Split hsr and dsa patches. (Florian Fainelli)
Changes since v1:
* Fixed some typos/wording. (Vladimir Oltean)
* eliminate IFF_HSR and use is_hsr_master instead. (Vladimir Oltean)
* Make hsr_handle_sup_frame handle skb_std as well (required when offloading)
* Don't add hsr tag for HSR v0 supervisory frames.
* Fixed tag insertion offloading for PRP.
George McCollister (4):
net: hsr: generate supervision frame without HSR/PRP tag
net: hsr: add offloading support
net: dsa: add support for offloading HSR
net: dsa: xrs700x: add HSR offloading support
Documentation/networking/netdev-features.rst | 21 ++++++
drivers/net/dsa/xrs700x/xrs700x.c | 106 +++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
include/linux/if_hsr.h | 27 +++++++
include/linux/netdev_features.h | 9 +++
include/net/dsa.h | 13 ++++
net/dsa/dsa_priv.h | 11 +++
net/dsa/port.c | 34 +++++++++
net/dsa/slave.c | 14 ++++
net/dsa/switch.c | 24 ++++++
net/dsa/tag_xrs700x.c | 7 +-
net/ethtool/common.c | 4 +
net/hsr/hsr_device.c | 46 ++----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 33 ++++++++-
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 +
net/hsr/hsr_main.c | 11 +++
net/hsr/hsr_main.h | 8 +-
net/hsr/hsr_slave.c | 10 ++-
20 files changed, 331 insertions(+), 56 deletions(-)
create mode 100644 include/linux/if_hsr.h
--
2.11.0
Hi George,
I will hopefully have some more time to look into this during the coming
weeks. What follows are some random thoughts so far, I hope you can
accept the windy road :)
Broadly speaking, I gather there are two common topologies that will be
used with the XRS chip: "End-device" and "RedBox".
End-device: RedBox:
.-----. .-----.
| CPU | | CPU |
'--+--' '--+--'
| |
.---0---. .---0---.
| XRS | | XRS 3--- Non-redundant network
'-1---2-' '-1---2-'
| | | |
HSR Ring HSR Ring
From the looks of it, this series only deals with the end-device
use-case. Is that right?
I will be targeting a RedBox setup, and I believe that means that the
remaining port has to be configured as an "interlink". (HSR/PRP is still
pretty new to me). Is that equivalent to a Linux config like this:
br0
/ \
hsr0 \
/ \ \
swp1 swp2 swp3
Or are there some additional semantics involved in forwarding between
the redundant ports and the interlink?
The chip is very rigid in the sense that most roles are statically
allocated to specific ports. I think we need to add checks for this.
Looking at the packets being generated on the redundant ports, both
regular traffic and supervision frames seem to be HSR-tagged. Are
supervision frames not supposed to be sent with an outer ethertype of
0x88fb? The manual talks about the possibility of setting up a policy
entry to bypass HSR-tagging (section 6.1.5), is this what that is for?
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Thanks for working on this!
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-08 21:11:07
On Mon, Feb 8, 2021 at 2:16 PM Tobias Waldekranz [off-list ref] wrote:
On Thu, Feb 04, 2021 at 15:59, George McCollister [off-list ref] wrote:
quoted
Add support for offloading HSR/PRP (IEC 62439-3) tag insertion, tag
removal, forwarding and duplication on DSA switches.
This series adds offloading to the xrs700x DSA driver.
Changes since RFC:
* Split hsr and dsa patches. (Florian Fainelli)
Changes since v1:
* Fixed some typos/wording. (Vladimir Oltean)
* eliminate IFF_HSR and use is_hsr_master instead. (Vladimir Oltean)
* Make hsr_handle_sup_frame handle skb_std as well (required when offloading)
* Don't add hsr tag for HSR v0 supervisory frames.
* Fixed tag insertion offloading for PRP.
George McCollister (4):
net: hsr: generate supervision frame without HSR/PRP tag
net: hsr: add offloading support
net: dsa: add support for offloading HSR
net: dsa: xrs700x: add HSR offloading support
Documentation/networking/netdev-features.rst | 21 ++++++
drivers/net/dsa/xrs700x/xrs700x.c | 106 +++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
include/linux/if_hsr.h | 27 +++++++
include/linux/netdev_features.h | 9 +++
include/net/dsa.h | 13 ++++
net/dsa/dsa_priv.h | 11 +++
net/dsa/port.c | 34 +++++++++
net/dsa/slave.c | 14 ++++
net/dsa/switch.c | 24 ++++++
net/dsa/tag_xrs700x.c | 7 +-
net/ethtool/common.c | 4 +
net/hsr/hsr_device.c | 46 ++----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 33 ++++++++-
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 +
net/hsr/hsr_main.c | 11 +++
net/hsr/hsr_main.h | 8 +-
net/hsr/hsr_slave.c | 10 ++-
20 files changed, 331 insertions(+), 56 deletions(-)
create mode 100644 include/linux/if_hsr.h
--
2.11.0
Hi George,
I will hopefully have some more time to look into this during the coming
weeks. What follows are some random thoughts so far, I hope you can
accept the windy road :)
Broadly speaking, I gather there are two common topologies that will be
used with the XRS chip: "End-device" and "RedBox".
End-device: RedBox:
.-----. .-----.
| CPU | | CPU |
'--+--' '--+--'
| |
.---0---. .---0---.
| XRS | | XRS 3--- Non-redundant network
'-1---2-' '-1---2-'
| | | |
HSR Ring HSR Ring
There is also the HSR-HSR use case and HSR-PRP use case.
From the looks of it, this series only deals with the end-device
use-case. Is that right?
Correct. net/hsr doesn't support this use case right now. It will
stomp the outgoing source MAC with that of the interface for instance.
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
I will be targeting a RedBox setup, and I believe that means that the
remaining port has to be configured as an "interlink". (HSR/PRP is still
pretty new to me). Is that equivalent to a Linux config like this:
Depends what you mean by configured as an interlink. I believe bit 9
of HSR_CFG in the switch is only supposed to be used for the HSR-HSR
and HSR-PRP use case, not HSR-SAN.
br0
/ \
hsr0 \
/ \ \
swp1 swp2 swp3
Or are there some additional semantics involved in forwarding between
the redundant ports and the interlink?
That sounds right.
The chip is very rigid in the sense that most roles are statically
allocated to specific ports. I think we need to add checks for this.
Okay. I'll look into this. Though a lot of the restrictions have to do
with using the third gigabit port for an HSR/PRP interlink (not
HSR-SAN) which I'm not currently supporting anyway.
Looking at the packets being generated on the redundant ports, both
regular traffic and supervision frames seem to be HSR-tagged. Are
supervision frames not supposed to be sent with an outer ethertype of
0x88fb? The manual talks about the possibility of setting up a policy
entry to bypass HSR-tagging (section 6.1.5), is this what that is for?
This was changed between 62439-3:2010 and 62439-3:2012.
"Prefixing the supervision frames on HSR by an HSR tag to simplify the hardware
implementation and introduce a unique EtherType for HSR to simplify processing."
The Linux HSR driver calls the former HSR v0 and the later HSR v1. I'm
not sure what their intention was with this feature. The inbound
policies are pretty flexible so maybe they didn't have anything so
specific in mind.
I don't think the xrs7000 series could offload HSR v0 anyway because
the tag ether type is different.
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Describe what you mean a bit more. Do you mean join and leave should
each only be called once with both hsr ports being passed in?
On Mon, Feb 08, 2021 at 15:09, George McCollister [off-list ref] wrote:
On Mon, Feb 8, 2021 at 2:16 PM Tobias Waldekranz [off-list ref] wrote:
quoted
On Thu, Feb 04, 2021 at 15:59, George McCollister [off-list ref] wrote:
quoted
Add support for offloading HSR/PRP (IEC 62439-3) tag insertion, tag
removal, forwarding and duplication on DSA switches.
This series adds offloading to the xrs700x DSA driver.
Changes since RFC:
* Split hsr and dsa patches. (Florian Fainelli)
Changes since v1:
* Fixed some typos/wording. (Vladimir Oltean)
* eliminate IFF_HSR and use is_hsr_master instead. (Vladimir Oltean)
* Make hsr_handle_sup_frame handle skb_std as well (required when offloading)
* Don't add hsr tag for HSR v0 supervisory frames.
* Fixed tag insertion offloading for PRP.
George McCollister (4):
net: hsr: generate supervision frame without HSR/PRP tag
net: hsr: add offloading support
net: dsa: add support for offloading HSR
net: dsa: xrs700x: add HSR offloading support
Documentation/networking/netdev-features.rst | 21 ++++++
drivers/net/dsa/xrs700x/xrs700x.c | 106 +++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
include/linux/if_hsr.h | 27 +++++++
include/linux/netdev_features.h | 9 +++
include/net/dsa.h | 13 ++++
net/dsa/dsa_priv.h | 11 +++
net/dsa/port.c | 34 +++++++++
net/dsa/slave.c | 14 ++++
net/dsa/switch.c | 24 ++++++
net/dsa/tag_xrs700x.c | 7 +-
net/ethtool/common.c | 4 +
net/hsr/hsr_device.c | 46 ++----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 33 ++++++++-
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 +
net/hsr/hsr_main.c | 11 +++
net/hsr/hsr_main.h | 8 +-
net/hsr/hsr_slave.c | 10 ++-
20 files changed, 331 insertions(+), 56 deletions(-)
create mode 100644 include/linux/if_hsr.h
--
2.11.0
Hi George,
I will hopefully have some more time to look into this during the coming
weeks. What follows are some random thoughts so far, I hope you can
accept the windy road :)
Broadly speaking, I gather there are two common topologies that will be
used with the XRS chip: "End-device" and "RedBox".
End-device: RedBox:
.-----. .-----.
| CPU | | CPU |
'--+--' '--+--'
| |
.---0---. .---0---.
| XRS | | XRS 3--- Non-redundant network
'-1---2-' '-1---2-'
| | | |
HSR Ring HSR Ring
There is also the HSR-HSR use case and HSR-PRP use case.
HSR-HSR is also known as a "QuadBox", yes? HSR-PRP is the same thing,
but having two PRP networks on one side and an HSR ring on the other?
quoted
From the looks of it, this series only deals with the end-device
use-case. Is that right?
Correct. net/hsr doesn't support this use case right now. It will
stomp the outgoing source MAC with that of the interface for instance.
Good to know! When would that behavior be required? Presumably it is not
overriding the SA just for fun?
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
So what is missing is basically to expand the current facility for
generating sequence numbers to maintain a table of such associations,
keyed by the SA?
Is the lack of that table the reason for enforcing that the SA match the
HSR netdev?
quoted
I will be targeting a RedBox setup, and I believe that means that the
remaining port has to be configured as an "interlink". (HSR/PRP is still
pretty new to me). Is that equivalent to a Linux config like this:
Depends what you mean by configured as an interlink. I believe bit 9
of HSR_CFG in the switch is only supposed to be used for the HSR-HSR
and HSR-PRP use case, not HSR-SAN.
Interesting, section 6.4.1 of the XRS manual states: "The interlink port
can be either in HSR, PRP or normal (non-HSR, non-PRP) mode." Maybe the
term is overloaded?
quoted
br0
/ \
hsr0 \
/ \ \
swp1 swp2 swp3
Or are there some additional semantics involved in forwarding between
the redundant ports and the interlink?
That sounds right.
quoted
The chip is very rigid in the sense that most roles are statically
allocated to specific ports. I think we need to add checks for this.
Okay. I'll look into this. Though a lot of the restrictions have to do
with using the third gigabit port for an HSR/PRP interlink (not
HSR-SAN) which I'm not currently supporting anyway.
But nothing is stopping me from trying to setup an HSR ring between port
(2,3) or (1,3), right? And that is not supported by the chip as I
understand it from looking at table 25.
quoted
Looking at the packets being generated on the redundant ports, both
regular traffic and supervision frames seem to be HSR-tagged. Are
supervision frames not supposed to be sent with an outer ethertype of
0x88fb? The manual talks about the possibility of setting up a policy
entry to bypass HSR-tagging (section 6.1.5), is this what that is for?
This was changed between 62439-3:2010 and 62439-3:2012.
"Prefixing the supervision frames on HSR by an HSR tag to simplify the hardware
implementation and introduce a unique EtherType for HSR to simplify
processing."
Thank you, that would have taken me a long time to figure out :)
The Linux HSR driver calls the former HSR v0 and the later HSR v1. I'm
not sure what their intention was with this feature. The inbound
policies are pretty flexible so maybe they didn't have anything so
specific in mind.
Now that I think of it, maybe you want things like LLDP to still operate
hop-by-hop over the ring?
I don't think the xrs7000 series could offload HSR v0 anyway because
the tag ether type is different.
quoted
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Describe what you mean a bit more. Do you mean join and leave should
each only be called once with both hsr ports being passed in?
Exactly. Maybe we could use `netdev_for_each_lower_dev` to figure out if
the other port has already been switched over to the new upper or
something. I find it hard to believe that there is any hardware out
there that can do something useful with a single HSR/PRP port anyway.
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-09 17:05:40
On Tue, Feb 9, 2021 at 8:38 AM Tobias Waldekranz [off-list ref] wrote:
On Mon, Feb 08, 2021 at 15:09, George McCollister [off-list ref] wrote:
quoted
On Mon, Feb 8, 2021 at 2:16 PM Tobias Waldekranz [off-list ref] wrote:
quoted
On Thu, Feb 04, 2021 at 15:59, George McCollister [off-list ref] wrote:
quoted
Add support for offloading HSR/PRP (IEC 62439-3) tag insertion, tag
removal, forwarding and duplication on DSA switches.
This series adds offloading to the xrs700x DSA driver.
Changes since RFC:
* Split hsr and dsa patches. (Florian Fainelli)
Changes since v1:
* Fixed some typos/wording. (Vladimir Oltean)
* eliminate IFF_HSR and use is_hsr_master instead. (Vladimir Oltean)
* Make hsr_handle_sup_frame handle skb_std as well (required when offloading)
* Don't add hsr tag for HSR v0 supervisory frames.
* Fixed tag insertion offloading for PRP.
George McCollister (4):
net: hsr: generate supervision frame without HSR/PRP tag
net: hsr: add offloading support
net: dsa: add support for offloading HSR
net: dsa: xrs700x: add HSR offloading support
Documentation/networking/netdev-features.rst | 21 ++++++
drivers/net/dsa/xrs700x/xrs700x.c | 106 +++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
include/linux/if_hsr.h | 27 +++++++
include/linux/netdev_features.h | 9 +++
include/net/dsa.h | 13 ++++
net/dsa/dsa_priv.h | 11 +++
net/dsa/port.c | 34 +++++++++
net/dsa/slave.c | 14 ++++
net/dsa/switch.c | 24 ++++++
net/dsa/tag_xrs700x.c | 7 +-
net/ethtool/common.c | 4 +
net/hsr/hsr_device.c | 46 ++----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 33 ++++++++-
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 +
net/hsr/hsr_main.c | 11 +++
net/hsr/hsr_main.h | 8 +-
net/hsr/hsr_slave.c | 10 ++-
20 files changed, 331 insertions(+), 56 deletions(-)
create mode 100644 include/linux/if_hsr.h
--
2.11.0
Hi George,
I will hopefully have some more time to look into this during the coming
weeks. What follows are some random thoughts so far, I hope you can
accept the windy road :)
Broadly speaking, I gather there are two common topologies that will be
used with the XRS chip: "End-device" and "RedBox".
End-device: RedBox:
.-----. .-----.
| CPU | | CPU |
'--+--' '--+--'
| |
.---0---. .---0---.
| XRS | | XRS 3--- Non-redundant network
'-1---2-' '-1---2-'
| | | |
HSR Ring HSR Ring
There is also the HSR-HSR use case and HSR-PRP use case.
HSR-HSR is also known as a "QuadBox", yes? HSR-PRP is the same thing,
but having two PRP networks on one side and an HSR ring on the other?
Yes. I believe you are correct.
From the spec:
"QuadBox
Quadruple port device connecting two peer HSR rings, which behaves as
an HSR node in
each ring and is able to filter the traffic and forward it from ring to ring."
quoted
quoted
From the looks of it, this series only deals with the end-device
use-case. Is that right?
Correct. net/hsr doesn't support this use case right now. It will
stomp the outgoing source MAC with that of the interface for instance.
Good to know! When would that behavior be required? Presumably it is not
overriding the SA just for fun?
Over the last few weeks I've looked over that code for way longer than
I'd like to admit and I'm still not sure. As far as I can tell, the
original authors have disappeared. My guess is it has something to do
with a configuration in which they had each redundant interface set to
a different MAC address and wanted the frames to go out with the
associated MAC address. As far as I can tell this is a violation of
the spec.
quoted
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
So what is missing is basically to expand the current facility for
generating sequence numbers to maintain a table of such associations,
keyed by the SA?
For the software implementation it would also need to use the
ProxyNodeTable to prevent forwarding matching frames on the ring and
delivering them to the hsr master port. It's also supposed to drop
frames coming in on a redundant port if the source address is in the
ProxyNodeTable.
Is the lack of that table the reason for enforcing that the SA match the
HSR netdev?
Could be.
quoted
quoted
I will be targeting a RedBox setup, and I believe that means that the
remaining port has to be configured as an "interlink". (HSR/PRP is still
pretty new to me). Is that equivalent to a Linux config like this:
Depends what you mean by configured as an interlink. I believe bit 9
of HSR_CFG in the switch is only supposed to be used for the HSR-HSR
and HSR-PRP use case, not HSR-SAN.
Interesting, section 6.4.1 of the XRS manual states: "The interlink port
can be either in HSR, PRP or normal (non-HSR, non-PRP) mode." Maybe the
term is overloaded?
Yeah I guess since it's the only port that can be used for QuadBox
they call it the interlink port even if it doesn't have the HSR/PRP
mode enabled with the interlink bit set.
quoted
quoted
br0
/ \
hsr0 \
/ \ \
swp1 swp2 swp3
Or are there some additional semantics involved in forwarding between
the redundant ports and the interlink?
That sounds right.
quoted
The chip is very rigid in the sense that most roles are statically
allocated to specific ports. I think we need to add checks for this.
Okay. I'll look into this. Though a lot of the restrictions have to do
with using the third gigabit port for an HSR/PRP interlink (not
HSR-SAN) which I'm not currently supporting anyway.
But nothing is stopping me from trying to setup an HSR ring between port
(2,3) or (1,3), right? And that is not supported by the chip as I
understand it from looking at table 25.
Yeah. That's why I said I'd look into it :). Wasn't an issue for my
board since port 0 isn't connected and port 3 is used as the CPU
facing port.
quoted
quoted
Looking at the packets being generated on the redundant ports, both
regular traffic and supervision frames seem to be HSR-tagged. Are
supervision frames not supposed to be sent with an outer ethertype of
0x88fb? The manual talks about the possibility of setting up a policy
entry to bypass HSR-tagging (section 6.1.5), is this what that is for?
This was changed between 62439-3:2010 and 62439-3:2012.
"Prefixing the supervision frames on HSR by an HSR tag to simplify the hardware
implementation and introduce a unique EtherType for HSR to simplify
processing."
Thank you, that would have taken me a long time to figure out :)
quoted
The Linux HSR driver calls the former HSR v0 and the later HSR v1. I'm
not sure what their intention was with this feature. The inbound
policies are pretty flexible so maybe they didn't have anything so
specific in mind.
Now that I think of it, maybe you want things like LLDP to still operate
hop-by-hop over the ring?
Not sure. Would need to look into it.
quoted
I don't think the xrs7000 series could offload HSR v0 anyway because
the tag ether type is different.
quoted
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Describe what you mean a bit more. Do you mean join and leave should
each only be called once with both hsr ports being passed in?
Exactly. Maybe we could use `netdev_for_each_lower_dev` to figure out if
the other port has already been switched over to the new upper or
something. I find it hard to believe that there is any hardware out
there that can do something useful with a single HSR/PRP port anyway.
If one port failed maybe it would still be useful to join one port if
the switch supported it? Maybe this couldn't ever happen anyway due
the way hsr is designed.
How were you thinking this would work? Would it just not use
dsa_port_notify() and call a switch op directly after the second
port's dsa_slave_changeupper() call? Or would we instead keep port
notifiers and calls to dsa_switch_hsr_join for each port and just make
dsa_switch_hsr_join() not call the switch op to create the HSR until
the second port called it? I'm not all that familiar with how these
dsa notifiers work and would prefer to stick with using a similar
mechanism to the bridge and lag support. It would be nice to get some
feedback from the DSA maintainers on how they would prefer it to work
if they indeed had a preference at all.
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-09 17:15:32
On Tue, Feb 09, 2021 at 11:04:08AM -0600, George McCollister wrote:
quoted
quoted
quoted
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Describe what you mean a bit more. Do you mean join and leave should
each only be called once with both hsr ports being passed in?
Exactly. Maybe we could use `netdev_for_each_lower_dev` to figure out if
the other port has already been switched over to the new upper or
something. I find it hard to believe that there is any hardware out
there that can do something useful with a single HSR/PRP port anyway.
If one port failed maybe it would still be useful to join one port if
the switch supported it? Maybe this couldn't ever happen anyway due
the way hsr is designed.
How were you thinking this would work? Would it just not use
dsa_port_notify() and call a switch op directly after the second
port's dsa_slave_changeupper() call? Or would we instead keep port
notifiers and calls to dsa_switch_hsr_join for each port and just make
dsa_switch_hsr_join() not call the switch op to create the HSR until
the second port called it? I'm not all that familiar with how these
dsa notifiers work and would prefer to stick with using a similar
mechanism to the bridge and lag support. It would be nice to get some
feedback from the DSA maintainers on how they would prefer it to work
if they indeed had a preference at all.
Even though I understand where this is coming from, I have grown to
dislike overengineered solutions. DSA is there to standardize how an
Ethernet-connected switch interacts with the network stack, not to be
the middle man that tries to offer a lending hand everywhere.
If there is a 50/50 chance that this extra logic in the DSA mid layer
will not be needed for the second switch that offloads HSR/PRP, then I'd
go with "don't do it". Just emit a hsr_join for both ports and let the
driver deal with it.
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-09 17:21:11
On Mon, Feb 08, 2021 at 11:21:26AM -0600, George McCollister wrote:
quoted
If you return zero, the software fallback is never going to kick in.
For join and leave? How is this not a problem for the bridge and lag
functions? They work the same way don't they? I figured it would be
safe to follow what they were doing.
I didn't say that the bridge and LAG offloading logic does the right
thing, but it is on its way there...
Those "XXX not offloaded" messages were tested with cases where the
.port_lag_join callback _is_ present, but fails (due to things like
incompatible xmit hashing policy). They were not tested with the case
where the driver does not implement .port_lag_join at all.
Doesn't mean you shouldn't do the right thing. I'll send some patches
soon, hopefully, fixing that for LAG and the bridge, you can concentrate
on HSR. For the non-offload scenario where the port is basically
standalone, we also need to disable the other bridge functions such as
address learning, otherwise it won't work properly, and that's where
I've been focusing my attention lately. You can't offload the bridge in
software, or a LAG, if you have address learning enabled. For HSR it's
even more interesting, you need to have address learning disabled even
when you offload the DANH/DANP.
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-09 18:56:25
On Tue, Feb 9, 2021 at 11:20 AM Vladimir Oltean [off-list ref] wrote:
On Mon, Feb 08, 2021 at 11:21:26AM -0600, George McCollister wrote:
quoted
quoted
If you return zero, the software fallback is never going to kick in.
For join and leave? How is this not a problem for the bridge and lag
functions? They work the same way don't they? I figured it would be
safe to follow what they were doing.
I didn't say that the bridge and LAG offloading logic does the right
thing, but it is on its way there...
Those "XXX not offloaded" messages were tested with cases where the
.port_lag_join callback _is_ present, but fails (due to things like
incompatible xmit hashing policy). They were not tested with the case
where the driver does not implement .port_lag_join at all.
Doesn't mean you shouldn't do the right thing. I'll send some patches
soon, hopefully, fixing that for LAG and the bridge, you can concentrate
on HSR. For the non-offload scenario where the port is basically
standalone, we also need to disable the other bridge functions such as
address learning, otherwise it won't work properly, and that's where
I've been focusing my attention lately. You can't offload the bridge in
software, or a LAG, if you have address learning enabled. For HSR it's
even more interesting, you need to have address learning disabled even
when you offload the DANH/DANP.
Do I just return -EOPNOTSUPP instead of 0 in dsa_switch_hsr_join and
dsa_switch_hsr_leave?
I'm not sure exactly what you're saying needs to be done wrt to
address learning with HSR. The switch does address learning
internally. Are you saying the DSA address learning needs to be
disabled? If that's something I need for this patch some tips on what
to do would be appreciated because I'm a bit lost.
Thanks
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-09 19:06:07
On Tue, Feb 09, 2021 at 12:37:38PM -0600, George McCollister wrote:
On Tue, Feb 9, 2021 at 11:20 AM Vladimir Oltean [off-list ref] wrote:
quoted
On Mon, Feb 08, 2021 at 11:21:26AM -0600, George McCollister wrote:
quoted
quoted
If you return zero, the software fallback is never going to kick in.
For join and leave? How is this not a problem for the bridge and lag
functions? They work the same way don't they? I figured it would be
safe to follow what they were doing.
I didn't say that the bridge and LAG offloading logic does the right
thing, but it is on its way there...
Those "XXX not offloaded" messages were tested with cases where the
.port_lag_join callback _is_ present, but fails (due to things like
incompatible xmit hashing policy). They were not tested with the case
where the driver does not implement .port_lag_join at all.
Doesn't mean you shouldn't do the right thing. I'll send some patches
soon, hopefully, fixing that for LAG and the bridge, you can concentrate
on HSR. For the non-offload scenario where the port is basically
standalone, we also need to disable the other bridge functions such as
address learning, otherwise it won't work properly, and that's where
I've been focusing my attention lately. You can't offload the bridge in
software, or a LAG, if you have address learning enabled. For HSR it's
even more interesting, you need to have address learning disabled even
when you offload the DANH/DANP.
Do I just return -EOPNOTSUPP instead of 0 in dsa_switch_hsr_join and
dsa_switch_hsr_leave?
Yes, return -EOPNOTSUPP if the callbacks are not implemented please.
I'm not sure exactly what you're saying needs to be done wrt to
address learning with HSR. The switch does address learning
internally. Are you saying the DSA address learning needs to be
disabled?
I'm saying that when you're doing any sort of redundancy protocol, the
switch will get confused by address learning if it performs it at
physical port level, because it will see the same source MAC address
coming in from 2 (or more) different physical ports. And when it sees a
packet coming in through a port that it had already learned it should be
the destination for the MAC address because an earlier packet came
having that MAC address as source, it will think it should do
hairpinning which it's configured not to => it'll drop the packet.
Now, your switch might have some sort of concept of address learning at
logical port level, where the "logical port" would roughly correspond to
the hsr0 upper (I haven't opened the XRS700x manual to know if it does
this, sorry). Basically if you support RedBox I expect that the switch
is able to learn at the level of "this MAC address came from the HSR
ring, aka from one or both of the individual ring ports". But for
configuring that in Linux, you'd need something like:
ip link set hsr0 master br0
ip link set hsr0 type bridge_slave learning on
and then catch from DSA the switchdev notification emitted for hsr0, and
use that to configure learning on the logical port of your switch
corresponding to hsr0, instead of learning on the physical ports that
offload it.
There are similar issues related to address learning for everything
except a bridge port, basically.
If that's something I need for this patch some tips on what
to do would be appreciated because I'm a bit lost.
I didn't say you need to change something related to learning for this
series, because you can't anyway - DSA doesn't give you the knobs to
configure address learning yet. The series where I try to add those are
here:
https://patchwork.kernel.org/project/netdevbpf/cover/20210209151936.97382-1-olteanv@gmail.com/
The take-away is that with those changes, a DSA driver should start its
ports in standalone mode with learning disabled and flooding of all
kinds enabled, and then treat the .port_bridge_flags callback which
should do the right thing and enable/disable address learning only when
necessary.
All I said is that address learning remaining enabled has been an issue
that prevented the non-offload scenarios from really working, but you
shouldn't be too hung up on that, and still return -EOPNOTSUPP, thereby
allowing the software fallback to kick in, even if it doesn't work.
From: George McCollister <george.mccollister@gmail.com> Date: 2021-02-09 22:55:53
On Tue, Feb 9, 2021 at 12:51 PM Vladimir Oltean [off-list ref] wrote:
On Tue, Feb 09, 2021 at 12:37:38PM -0600, George McCollister wrote:
quoted
On Tue, Feb 9, 2021 at 11:20 AM Vladimir Oltean [off-list ref] wrote:
quoted
On Mon, Feb 08, 2021 at 11:21:26AM -0600, George McCollister wrote:
quoted
quoted
If you return zero, the software fallback is never going to kick in.
For join and leave? How is this not a problem for the bridge and lag
functions? They work the same way don't they? I figured it would be
safe to follow what they were doing.
I didn't say that the bridge and LAG offloading logic does the right
thing, but it is on its way there...
Those "XXX not offloaded" messages were tested with cases where the
.port_lag_join callback _is_ present, but fails (due to things like
incompatible xmit hashing policy). They were not tested with the case
where the driver does not implement .port_lag_join at all.
Doesn't mean you shouldn't do the right thing. I'll send some patches
soon, hopefully, fixing that for LAG and the bridge, you can concentrate
on HSR. For the non-offload scenario where the port is basically
standalone, we also need to disable the other bridge functions such as
address learning, otherwise it won't work properly, and that's where
I've been focusing my attention lately. You can't offload the bridge in
software, or a LAG, if you have address learning enabled. For HSR it's
even more interesting, you need to have address learning disabled even
when you offload the DANH/DANP.
Do I just return -EOPNOTSUPP instead of 0 in dsa_switch_hsr_join and
dsa_switch_hsr_leave?
Yes, return -EOPNOTSUPP if the callbacks are not implemented please.
Will do.
quoted
I'm not sure exactly what you're saying needs to be done wrt to
address learning with HSR. The switch does address learning
internally. Are you saying the DSA address learning needs to be
disabled?
I'm saying that when you're doing any sort of redundancy protocol, the
switch will get confused by address learning if it performs it at
physical port level, because it will see the same source MAC address
coming in from 2 (or more) different physical ports. And when it sees a
packet coming in through a port that it had already learned it should be
the destination for the MAC address because an earlier packet came
having that MAC address as source, it will think it should do
hairpinning which it's configured not to => it'll drop the packet.
Now, your switch might have some sort of concept of address learning at
logical port level, where the "logical port" would roughly correspond to
the hsr0 upper (I haven't opened the XRS700x manual to know if it does
this, sorry). Basically if you support RedBox I expect that the switch
is able to learn at the level of "this MAC address came from the HSR
ring, aka from one or both of the individual ring ports". But for
configuring that in Linux, you'd need something like:
Yup, the switch has provisions to deal with this.
ip link set hsr0 master br0
ip link set hsr0 type bridge_slave learning on
and then catch from DSA the switchdev notification emitted for hsr0, and
use that to configure learning on the logical port of your switch
corresponding to hsr0, instead of learning on the physical ports that
offload it.
There are similar issues related to address learning for everything
except a bridge port, basically.
quoted
If that's something I need for this patch some tips on what
to do would be appreciated because I'm a bit lost.
I didn't say you need to change something related to learning for this
series, because you can't anyway - DSA doesn't give you the knobs to
configure address learning yet. The series where I try to add those are
here:
https://patchwork.kernel.org/project/netdevbpf/cover/20210209151936.97382-1-olteanv@gmail.com/
The take-away is that with those changes, a DSA driver should start its
ports in standalone mode with learning disabled and flooding of all
kinds enabled, and then treat the .port_bridge_flags callback which
should do the right thing and enable/disable address learning only when
necessary.
All I said is that address learning remaining enabled has been an issue
that prevented the non-offload scenarios from really working, but you
shouldn't be too hung up on that, and still return -EOPNOTSUPP, thereby
allowing the software fallback to kick in, even if it doesn't work.
Thanks for clearing that up and explaining how this will work in the future.
On Tue, Feb 09, 2021 at 11:04, George McCollister [off-list ref] wrote:
On Tue, Feb 9, 2021 at 8:38 AM Tobias Waldekranz [off-list ref] wrote:
quoted
On Mon, Feb 08, 2021 at 15:09, George McCollister [off-list ref] wrote:
quoted
On Mon, Feb 8, 2021 at 2:16 PM Tobias Waldekranz [off-list ref] wrote:
quoted
On Thu, Feb 04, 2021 at 15:59, George McCollister [off-list ref] wrote:
quoted
Add support for offloading HSR/PRP (IEC 62439-3) tag insertion, tag
removal, forwarding and duplication on DSA switches.
This series adds offloading to the xrs700x DSA driver.
Changes since RFC:
* Split hsr and dsa patches. (Florian Fainelli)
Changes since v1:
* Fixed some typos/wording. (Vladimir Oltean)
* eliminate IFF_HSR and use is_hsr_master instead. (Vladimir Oltean)
* Make hsr_handle_sup_frame handle skb_std as well (required when offloading)
* Don't add hsr tag for HSR v0 supervisory frames.
* Fixed tag insertion offloading for PRP.
George McCollister (4):
net: hsr: generate supervision frame without HSR/PRP tag
net: hsr: add offloading support
net: dsa: add support for offloading HSR
net: dsa: xrs700x: add HSR offloading support
Documentation/networking/netdev-features.rst | 21 ++++++
drivers/net/dsa/xrs700x/xrs700x.c | 106 +++++++++++++++++++++++++++
drivers/net/dsa/xrs700x/xrs700x_reg.h | 5 ++
include/linux/if_hsr.h | 27 +++++++
include/linux/netdev_features.h | 9 +++
include/net/dsa.h | 13 ++++
net/dsa/dsa_priv.h | 11 +++
net/dsa/port.c | 34 +++++++++
net/dsa/slave.c | 14 ++++
net/dsa/switch.c | 24 ++++++
net/dsa/tag_xrs700x.c | 7 +-
net/ethtool/common.c | 4 +
net/hsr/hsr_device.c | 46 ++----------
net/hsr/hsr_device.h | 1 -
net/hsr/hsr_forward.c | 33 ++++++++-
net/hsr/hsr_forward.h | 1 +
net/hsr/hsr_framereg.c | 2 +
net/hsr/hsr_main.c | 11 +++
net/hsr/hsr_main.h | 8 +-
net/hsr/hsr_slave.c | 10 ++-
20 files changed, 331 insertions(+), 56 deletions(-)
create mode 100644 include/linux/if_hsr.h
--
2.11.0
Hi George,
I will hopefully have some more time to look into this during the coming
weeks. What follows are some random thoughts so far, I hope you can
accept the windy road :)
Broadly speaking, I gather there are two common topologies that will be
used with the XRS chip: "End-device" and "RedBox".
End-device: RedBox:
.-----. .-----.
| CPU | | CPU |
'--+--' '--+--'
| |
.---0---. .---0---.
| XRS | | XRS 3--- Non-redundant network
'-1---2-' '-1---2-'
| | | |
HSR Ring HSR Ring
There is also the HSR-HSR use case and HSR-PRP use case.
HSR-HSR is also known as a "QuadBox", yes? HSR-PRP is the same thing,
but having two PRP networks on one side and an HSR ring on the other?
Yes. I believe you are correct.
From the spec:
"QuadBox
Quadruple port device connecting two peer HSR rings, which behaves as
an HSR node in
each ring and is able to filter the traffic and forward it from ring to ring."
quoted
quoted
quoted
From the looks of it, this series only deals with the end-device
use-case. Is that right?
Correct. net/hsr doesn't support this use case right now. It will
stomp the outgoing source MAC with that of the interface for instance.
Good to know! When would that behavior be required? Presumably it is not
overriding the SA just for fun?
Over the last few weeks I've looked over that code for way longer than
I'd like to admit and I'm still not sure. As far as I can tell, the
original authors have disappeared. My guess is it has something to do
with a configuration in which they had each redundant interface set to
a different MAC address and wanted the frames to go out with the
associated MAC address. As far as I can tell this is a violation of
the spec.
Alright, so maybe a "Works for Me (TM)" solution initially, where any
devince stacking was out of scope.
quoted
quoted
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
So what is missing is basically to expand the current facility for
generating sequence numbers to maintain a table of such associations,
keyed by the SA?
For the software implementation it would also need to use the
ProxyNodeTable to prevent forwarding matching frames on the ring and
delivering them to the hsr master port. It's also supposed to drop
frames coming in on a redundant port if the source address is in the
ProxyNodeTable.
This whole thing sounds an awful lot like an FDB. I suppose an option
would be to implement the RedBox/QuadBox roles in the bridge, perhaps by
building on the work done for MRP? Feel free to tell me I'm crazy :)
quoted
Is the lack of that table the reason for enforcing that the SA match the
HSR netdev?
Could be.
quoted
quoted
quoted
I will be targeting a RedBox setup, and I believe that means that the
remaining port has to be configured as an "interlink". (HSR/PRP is still
pretty new to me). Is that equivalent to a Linux config like this:
Depends what you mean by configured as an interlink. I believe bit 9
of HSR_CFG in the switch is only supposed to be used for the HSR-HSR
and HSR-PRP use case, not HSR-SAN.
Interesting, section 6.4.1 of the XRS manual states: "The interlink port
can be either in HSR, PRP or normal (non-HSR, non-PRP) mode." Maybe the
term is overloaded?
Yeah I guess since it's the only port that can be used for QuadBox
they call it the interlink port even if it doesn't have the HSR/PRP
mode enabled with the interlink bit set.
Right, that makes sense.
quoted
quoted
quoted
br0
/ \
hsr0 \
/ \ \
swp1 swp2 swp3
Or are there some additional semantics involved in forwarding between
the redundant ports and the interlink?
That sounds right.
quoted
The chip is very rigid in the sense that most roles are statically
allocated to specific ports. I think we need to add checks for this.
Okay. I'll look into this. Though a lot of the restrictions have to do
with using the third gigabit port for an HSR/PRP interlink (not
HSR-SAN) which I'm not currently supporting anyway.
But nothing is stopping me from trying to setup an HSR ring between port
(2,3) or (1,3), right? And that is not supported by the chip as I
understand it from looking at table 25.
Yeah. That's why I said I'd look into it :). Wasn't an issue for my
board since port 0 isn't connected and port 3 is used as the CPU
facing port.
Fair enough :)
quoted
quoted
quoted
Looking at the packets being generated on the redundant ports, both
regular traffic and supervision frames seem to be HSR-tagged. Are
supervision frames not supposed to be sent with an outer ethertype of
0x88fb? The manual talks about the possibility of setting up a policy
entry to bypass HSR-tagging (section 6.1.5), is this what that is for?
This was changed between 62439-3:2010 and 62439-3:2012.
"Prefixing the supervision frames on HSR by an HSR tag to simplify the hardware
implementation and introduce a unique EtherType for HSR to simplify
processing."
Thank you, that would have taken me a long time to figure out :)
quoted
The Linux HSR driver calls the former HSR v0 and the later HSR v1. I'm
not sure what their intention was with this feature. The inbound
policies are pretty flexible so maybe they didn't have anything so
specific in mind.
Now that I think of it, maybe you want things like LLDP to still operate
hop-by-hop over the ring?
Not sure. Would need to look into it.
quoted
quoted
I don't think the xrs7000 series could offload HSR v0 anyway because
the tag ether type is different.
quoted
In the DSA layer (dsa_slave_changeupper), could we merge the two HSR
join/leave calls somehow? My guess is all drivers are going to end up
having to do the same dance of deferring configuration until both ports
are known.
Describe what you mean a bit more. Do you mean join and leave should
each only be called once with both hsr ports being passed in?
Exactly. Maybe we could use `netdev_for_each_lower_dev` to figure out if
the other port has already been switched over to the new upper or
something. I find it hard to believe that there is any hardware out
there that can do something useful with a single HSR/PRP port anyway.
If one port failed maybe it would still be useful to join one port if
the switch supported it? Maybe this couldn't ever happen anyway due
the way hsr is designed.
How were you thinking this would work? Would it just not use
dsa_port_notify() and call a switch op directly after the second
port's dsa_slave_changeupper() call? Or would we instead keep port
Yeah this is what I was thinking. But I understand where Vladimir is
coming from. Fundamentally, I am also on the library side of the
"library vs. framework" spectrum.
notifiers and calls to dsa_switch_hsr_join for each port and just make
dsa_switch_hsr_join() not call the switch op to create the HSR until
the second port called it? I'm not all that familiar with how these
dsa notifiers work and would prefer to stick with using a similar
mechanism to the bridge and lag support. It would be nice to get some
feedback from the DSA maintainers on how they would prefer it to work
if they indeed had a preference at all.
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-10 21:56:12
On Wed, Feb 10, 2021 at 10:10:14PM +0100, Tobias Waldekranz wrote:
On Tue, Feb 09, 2021 at 11:04, George McCollister [off-list ref] wrote:
quoted
quoted
quoted
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
So what is missing is basically to expand the current facility for
generating sequence numbers to maintain a table of such associations,
keyed by the SA?
For the software implementation it would also need to use the
ProxyNodeTable to prevent forwarding matching frames on the ring and
delivering them to the hsr master port. It's also supposed to drop
frames coming in on a redundant port if the source address is in the
ProxyNodeTable.
This whole thing sounds an awful lot like an FDB. I suppose an option
would be to implement the RedBox/QuadBox roles in the bridge, perhaps by
building on the work done for MRP? Feel free to tell me I'm crazy :)
As far as I understand, the VDAN needs to generate supervision frames on
behalf of all nodes that it proxies. Therefore, implementing the
RedBox/QuadBox in the bridge is probably not practical. What I was
discussing with George though is that maybe we can make hsr a consumer
of SWITCHDEV_FDB_ADD_TO_DEVICE events, similar to DSA with its
assisted_learning_on_cpu_port functionality, and that would be how it
populates its proxy node table. A RedBox becomes a bridge with one HSR
interface and one or more standalone interfaces, and a QuadBox becomes a
bridge with two HSR interfaces. How does that sound?
On Wed, Feb 10, 2021 at 23:55, Vladimir Oltean [off-list ref] wrote:
On Wed, Feb 10, 2021 at 10:10:14PM +0100, Tobias Waldekranz wrote:
quoted
On Tue, Feb 09, 2021 at 11:04, George McCollister [off-list ref] wrote:
quoted
quoted
quoted
It also doesn't implement a ProxyNodeTable (though that actually
wouldn't matter if you were offloading to the xrs700x I think). Try
commenting out the ether_addr_copy() line in hsr_xmit and see if it
makes your use case work.
So what is missing is basically to expand the current facility for
generating sequence numbers to maintain a table of such associations,
keyed by the SA?
For the software implementation it would also need to use the
ProxyNodeTable to prevent forwarding matching frames on the ring and
delivering them to the hsr master port. It's also supposed to drop
frames coming in on a redundant port if the source address is in the
ProxyNodeTable.
This whole thing sounds an awful lot like an FDB. I suppose an option
would be to implement the RedBox/QuadBox roles in the bridge, perhaps by
building on the work done for MRP? Feel free to tell me I'm crazy :)
As far as I understand, the VDAN needs to generate supervision frames on
behalf of all nodes that it proxies. Therefore, implementing the
RedBox/QuadBox in the bridge is probably not practical. What I was
discussing with George though is that maybe we can make hsr a consumer
of SWITCHDEV_FDB_ADD_TO_DEVICE events, similar to DSA with its
assisted_learning_on_cpu_port functionality, and that would be how it
populates its proxy node table.
Is it not easier to just implement learning in the HSR layer? Seeing as
you need to look up the table for each packet anyway, you might as well
add a new entry on a miss. Otherwise you run the risk of filling up your
proxy table with entries that never egress the HSR device. Perhaps not
likely on this particular device, but on a 48-port switch with HSR
offloading it might be.
This should also work for more exotic configs with multiple macvlans for
example:
macvlan0 macvlan1
\ /
hsr0
/ \
swp1 swp2
A RedBox becomes a bridge with one HSR
interface and one or more standalone interfaces, and a QuadBox becomes a
bridge with two HSR interfaces. How does that sound?
Yeah that is the straight forward solution, and what I tried to describe
earlier in the thread with this illustration:
>> >> br0
>> >> / \
>> >> hsr0 \
>> >> / \ \
>> >> swp1 swp2 swp3
I just wanted to double check that we had not overlooked a better
solution outside of the existing HSR code.
From: Vladimir Oltean <olteanv@gmail.com> Date: 2021-02-13 00:44:39
On Sat, Feb 13, 2021 at 12:52:36AM +0100, Tobias Waldekranz wrote:
On Wed, Feb 10, 2021 at 23:55, Vladimir Oltean [off-list ref] wrote:
quoted
On Wed, Feb 10, 2021 at 10:10:14PM +0100, Tobias Waldekranz wrote:
quoted
This whole thing sounds an awful lot like an FDB. I suppose an option
would be to implement the RedBox/QuadBox roles in the bridge, perhaps by
building on the work done for MRP? Feel free to tell me I'm crazy :)
As far as I understand, the VDAN needs to generate supervision frames on
behalf of all nodes that it proxies. Therefore, implementing the
RedBox/QuadBox in the bridge is probably not practical. What I was
discussing with George though is that maybe we can make hsr a consumer
of SWITCHDEV_FDB_ADD_TO_DEVICE events, similar to DSA with its
assisted_learning_on_cpu_port functionality, and that would be how it
populates its proxy node table.
Is it not easier to just implement learning in the HSR layer? Seeing as
you need to look up the table for each packet anyway, you might as well
add a new entry on a miss. Otherwise you run the risk of filling up your
proxy table with entries that never egress the HSR device. Perhaps not
likely on this particular device, but on a 48-port switch with HSR
offloading it might be.
In the HSR layer, sure, I didn't mean to suggest otherwise, I thought
you wanted to, when you said "I suppose an option would be to implement
the RedBox/QuadBox roles in the bridge".
So then the SWITCHDEV_FDB_ADD_TO_DEVICE events might be too much.
Learning / populating the proxy node table can be probably done from the
xmit function, with the only potential issue that the first packets will
probably be lost, since no supervision frames have yet been transmitted
for those proxied nodes.
This should also work for more exotic configs with multiple macvlans for
example:
macvlan0 macvlan1
\ /
hsr0
/ \
swp1 swp2
Yes, I don't think macvlan uses switchdev.
quoted
A RedBox becomes a bridge with one HSR
interface and one or more standalone interfaces, and a QuadBox becomes a
bridge with two HSR interfaces. How does that sound?
Yeah that is the straight forward solution, and what I tried to describe
earlier in the thread with this illustration:
>> >> br0
>> >> / \
>> >> hsr0 \
>> >> / \ \
>> >> swp1 swp2 swp3
I just wanted to double check that we had not overlooked a better
solution outside of the existing HSR code.
I'm not aware of a better solution, but I'm also interested if there is one.