Skip to contentSkip to conversation

net/smc: Introduce SMC-D-based OS internal communication acceleration

RFC PATCH In review Wen Gu 3 participants 17 messages Apr 23, 2023 – May 10, 2023

Subsystems (5)
lore
Share & tools
lore

Wen Gu

author reviewer
#0 lore raw

v5 cover net/smc: Introduce SMC-D-based OS internal communication acceleration

Hi, all

# Background

The background and previous discussion can be referred from [1]~[3].

We found SMC-D can be used to accelerate OS internal communication, such as loopback or between two containers within the same OS instance. So this patch set provides a kind of SMC-D dummy device (we call it the SMC-D loopback device) to emulate an ISM device, so that SMC-D can also be used on architectures other than s390. The SMC-D loopback device are designed as a system global device, visible to all containers.

# Design

This patch set basically follows the design of the previous version.

Patch #1/9 ~ #3/9 attempt to decouple ISM-related structures from the SMC-D generalized code and extract some helpers to make SMC-D protocol compatible with devices other than s390 ISM device.

Patch #4/9 introduces a kind of loopback device, which is defined as SMC-Dv2 device and designed to provide communication between SMC sockets on the same OS instance.

 +-------------------------------------------+
 |  +--------------+       +--------------+  |
 |  | SMC socket A |       | SMC socket B |  |
 |  +--------------+       +--------------+  |
 |       ^                         ^         |
 |       |    +----------------+   |         |
 |       |    |   SMC stack    |   |         |
 |       +--->| +------------+ |<--|         |
 |            | |   dummy    | |             |
 |            | |   device   | |             |
 |            +-+------------+-+             |
 |                   OS                      |
 +-------------------------------------------+

Patch #5/9 ~ #8/9 expand SMC-D protocol interface (smcd_ops) for scenarios where SMC-D is used to communicate within VM (loopback here) or between VMs on the same host (based on virtio-ism device, see [4]). What these scenarios have in common is that the local sndbuf and peer RMB can be mapped to same physical memory region, so the data copy between the local sndbuf and peer RMB can be omitted. Performance improvement brought by this extension can be found in # Benchmark Test.

 +----------+                     +----------+
 | socket A |                     | socket B |
 +----------+                     +----------+
       |                               ^
       |         +---------+           |
  regard as      |         | ----------|
  local sndbuf   |  B's    |     regard as
       |         |  RMB    |     local RMB
       |-------> |         |
                 +---------+

Patch #9/9 realizes the support of loopback device for the above-mentioned expanded SMC-D protocol interface.

# Benchmark Test

* Test environments:
- VM with Intel Xeon Platinum 8 core 2.50GHz, 16 GiB mem.
- SMC sndbuf/RMB size 1MB.

* Test object:
- TCP lo: run on TCP loopback.
- domain: run on UNIX domain.
- SMC lo: run on SMC loopback device with patch #1/9 ~ #4/9.
- SMC lo-nocpy: run on SMC loopback device with patch #1/9 ~ #9/9.

1. ipc-benchmark (see [5])

- ./<foo> -c 1000000 -s 100

TCP-lo domain SMC-lo SMC-lo-nocpy
Message
rate (msg/s) 79025 115736(+46.45%) 146760(+85.71%) 149800(+89.56%)

2. sockperf

- serv: <smc_run> taskset -c <cpu> sockperf sr --tcp
- clnt: <smc_run> taskset -c <cpu> sockperf { tp | pp } --tcp --msg-size={ 64000 for tp | 14 for pp } -i 127.0.0.1 -t 30

TCP-lo SMC-lo SMC-lo-nocpy
Bandwidth(MBps) 4822.388 4940.918(+2.56%) 8086.67(+67.69%)
Latency(us) 6.298 3.352(-46.78%) 3.35(-46.81%)

3. iperf3

- serv: <smc_run> taskset -c <cpu> iperf3 -s
- clnt: <smc_run> taskset -c <cpu> iperf3 -c 127.0.0.1 -t 15

TCP-lo SMC-lo SMC-lo-nocpy
Bitrate(Gb/s) 40.7 40.5(-0.49%) 72.4(+77.89%)

4. nginx/wrk

- serv: <smc_run> nginx
- clnt: <smc_run> wrk -t 8 -c 500 -d 30 http://127.0.0.1:80

TCP-lo SMC-lo SMC-lo-nocpy
Requests/s 155994.57 214544.79(+37.53%) 215538.55(+38.17%)

v5->v4
1. The loopback device generates SEID in the same way as the ISM devices when coexisting with ISM devices and uses a default fixed SEID in other cases.
2. Ensure each DMB token of the same loopback device is unique.
3. Fixe a crash caused by setting smcd_ops->signal_event interface to NULL.
4. Fixe a compilation warning complained by kernel test rebot.

v4->v3
1. Rebase to the latest net-next;
2. Introduce SEID helper. SMC-D loopback will return SMCD_DEFAULT_V2_SEID. And if it coexist with ISM device, the SEID of ISM device will overwrite SMCD_DEFAULT_V2_SEID as smc_ism_v2_system_eid.
3. Won't remove dmb_node from hashtable until no sndbuf attaching to it.

Something postponed in this version
1. Hierarchy perference of SMC-D devices when loopback and ISM devices coexist, which will be determinated after comparing the performance of loopback and ISM.

v3->v2
1. Adapt new generalized interface provided by [2];
2. Select loopback device through SMC-D v2 protocol;
3. Split the loopback-related implementation and generic implementation into different patches more reasonably.

v1->v2
1. Fix some build WARNINGs complained by kernel test rebot

2. Add iperf3 test data.

[1] https://lore.kernel.org/netdev/[email protected]/
[2] https://lore.kernel.org/netdev/[email protected]/
[3] https://lore.kernel.org/netdev/[email protected]/
[4] https://lore.kernel.org/all/[email protected]/
[5] https://github.com/goldsborough/ipc-bench

Wen Gu (9):
net/smc: Decouple ism_dev from SMC-D device dump net/smc: Decouple ism_dev from SMC-D DMB registration net/smc: Extract v2 check helper from SMC-D device registration net/smc: Introduce SMC-D loopback device
net/smc: Introduce an interface for getting DMB attribute net/smc: Introudce interfaces for DMB attach and detach net/smc: Avoid data copy from sndbuf to peer RMB in SMC-D net/smc: Modify cursor update logic when using mappable DMB net/smc: Add interface implementation of loopback device

 drivers/s390/net/ism_drv.c |   5 +-
 include/net/smc.h          |  18 +-
 net/smc/Makefile           |   2 +-
 net/smc/af_smc.c           |  26 ++-
 net/smc/smc_cdc.c          |  59 ++++--
 net/smc/smc_cdc.h          |   1 +
 net/smc/smc_core.c         |  70 ++++++-
 net/smc/smc_core.h         |   1 +
 net/smc/smc_ism.c          |  79 ++++++--
 net/smc/smc_ism.h          |   6 +
 net/smc/smc_loopback.c     | 491 +++++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_loopback.h     |  56 ++++++
 12 files changed, 777 insertions(+), 37 deletions(-)
 create mode 100644 net/smc/smc_loopback.c
 create mode 100644 net/smc/smc_loopback.h
Signature · 1 line
1.8.3.1
author reviewer
#1 lore raw

RFC PATCH v5 1/9 net/smc: Decouple ism_dev from SMC-D device dump

This patch helps to decouple SMC-D device and ISM device, allowing different underlying device forms, such as non-PCI devices.

---
 include/net/smc.h | 1 +
 net/smc/smc_ism.c | 6 +++---
 2 files changed, 4 insertions(+), 3 deletions(-)
include/net/smc.h +1 -0
index a002552..963ce9c 100644
@@ -77,6 +77,7 @@ struct smcd_ops {
 struct smcd_dev {
 	const struct smcd_ops *ops;
 	void *priv;
+	struct device *parent_pci_dev;
 	struct list_head list;
 	spinlock_t lock;
 	struct smc_connection **conn;
net/smc/smc_ism.c +3 -3
index fbee249..4249fb9 100644
@@ -231,11 +231,9 @@ static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
 	struct smc_pci_dev smc_pci_dev;
 	struct nlattr *port_attrs;
 	struct nlattr *attrs;
-	struct ism_dev *ism;
 	int use_cnt = 0;
 	void *nlh;
 
-	ism = smcd->priv;
 	nlh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
 			  &smc_gen_nl_family, NLM_F_MULTI,
 			  SMC_NETLINK_GET_DEV_SMCD);
@@ -250,7 +248,8 @@ static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
 	if (nla_put_u8(skb, SMC_NLA_DEV_IS_CRIT, use_cnt > 0))
 		goto errattr;
 	memset(&smc_pci_dev, 0, sizeof(smc_pci_dev));
-	smc_set_pci_values(to_pci_dev(ism->dev.parent), &smc_pci_dev);
+	if (smcd->parent_pci_dev)
+		smc_set_pci_values(to_pci_dev(smcd->parent_pci_dev), &smc_pci_dev);
 	if (nla_put_u32(skb, SMC_NLA_DEV_PCI_FID, smc_pci_dev.pci_fid))
 		goto errattr;
 	if (nla_put_u16(skb, SMC_NLA_DEV_PCI_CHID, smc_pci_dev.pci_pchid))
@@ -420,6 +419,7 @@ static void smcd_register_dev(struct ism_dev *ism)
 	if (!smcd)
 		return;
 	smcd->priv = ism;
+	smcd->parent_pci_dev = ism->dev.parent;
 	ism_set_priv(ism, &smc_ism_client, smcd);
 	if (smc_pnetid_by_dev_port(&ism->pdev->dev, 0, smcd->pnetid))
 		smc_pnetid_by_table_smcd(smcd);
Signature · 1 line
1.8.3.1
author reviewer
#2 lore raw

RFC PATCH v5 2/9 net/smc: Decouple ism_dev from SMC-D DMB registration

In reply to Wen Gu

This patch tries to decouple ISM device from SMC-D DMB registration, So that the register_dmb option is not restricted to be used by ISM device.

---
 drivers/s390/net/ism_drv.c | 5 +++--
 include/net/smc.h          | 4 ++--
 net/smc/smc_ism.c          | 7 ++-----
 3 files changed, 7 insertions(+), 9 deletions(-)
drivers/s390/net/ism_drv.c +3 -2
index 8acb9eb..5eeb54d 100644
@@ -796,9 +796,10 @@ static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid,
 }
 
 static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
-			     struct ism_client *client)
+			     void *client_priv)
 {
-	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb, client);
+	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb,
+				(struct ism_client *)client_priv);
 }
 
 static int smcd_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)
include/net/smc.h +2 -2
index 963ce9c..26206d2 100644
@@ -50,13 +50,12 @@ struct smcd_dmb {
 #define ISM_ERROR	0xFFFF
 
 struct smcd_dev;
-struct ism_client;
 
 struct smcd_ops {
 	int (*query_remote_gid)(struct smcd_dev *dev, u64 rgid, u32 vid_valid,
 				u32 vid);
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
-			    struct ism_client *client);
+			    void *client_priv);
 	int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
 	int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
 	int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
@@ -77,6 +76,7 @@ struct smcd_ops {
 struct smcd_dev {
 	const struct smcd_ops *ops;
 	void *priv;
+	void *client_priv;
 	struct device *parent_pci_dev;
 	struct list_head list;
 	spinlock_t lock;
net/smc/smc_ism.c +2 -5
index 4249fb9..d4709ca 100644
@@ -200,7 +200,6 @@ int smc_ism_unregister_dmb(struct smcd_dev *smcd, struct smc_buf_desc *dmb_desc)
 int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 			 struct smc_buf_desc *dmb_desc)
 {
-#if IS_ENABLED(CONFIG_ISM)
 	struct smcd_dmb dmb;
 	int rc;
 
@@ -209,7 +208,7 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 	dmb.sba_idx = dmb_desc->sba_idx;
 	dmb.vlan_id = lgr->vlan_id;
 	dmb.rgid = lgr->peer_gid;
-	rc = lgr->smcd->ops->register_dmb(lgr->smcd, &dmb, &smc_ism_client);
+	rc = lgr->smcd->ops->register_dmb(lgr->smcd, &dmb, lgr->smcd->client_priv);
 	if (!rc) {
 		dmb_desc->sba_idx = dmb.sba_idx;
 		dmb_desc->token = dmb.dmb_tok;
@@ -218,9 +217,6 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 		dmb_desc->len = dmb.dmb_len;
 	}
 	return rc;
-#else
-	return 0;
-#endif
 }
 
 static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
@@ -419,6 +415,7 @@ static void smcd_register_dev(struct ism_dev *ism)
 	if (!smcd)
 		return;
 	smcd->priv = ism;
+	smcd->client_priv = &smc_ism_client;
 	smcd->parent_pci_dev = ism->dev.parent;
 	ism_set_priv(ism, &smc_ism_client, smcd);
 	if (smc_pnetid_by_dev_port(&ism->pdev->dev, 0, smcd->pnetid))
Signature · 1 line
1.8.3.1
#3 lore raw

Re: [RFC PATCH net-next v5 2/9] net/smc: Decouple ism_dev from SMC-D DMB registration

Quoted 16 lines · Wen Gu

This patch tries to decouple ISM device from SMC-D DMB registration, So that the register_dmb option is not restricted to be used by ISM device.

---
 drivers/s390/net/ism_drv.c | 5 +++--
 include/net/smc.h          | 4 ++--
 net/smc/smc_ism.c          | 7 ++-----
 3 files changed, 7 insertions(+), 9 deletions(-)
drivers/s390/net/ism_drv.c +0 -0
index 8acb9eb..5eeb54d 100644

drivers/s390/net/ism_drv.c static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid, · line 802

Quoted 9 lines
drivers/s390/net/ism_drv.c +3 -2
@@ -796,9 +796,10 @@ static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid,
 }
 
 static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
-			     struct ism_client *client)
+			     void *client_priv)
 {
-	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb, client);
+	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb,
+				(struct ism_client *)client_priv);

Hi Wen Gu,

a minor nit from my side: there is no need to cast a void pointer to another type.

Quoted 3 lines
drivers/s390/net/ism_drv.c +0 -0
 }
 
 static int smcd_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)

...

author reviewer
#4 lore raw

Re: [RFC PATCH net-next v5 2/9] net/smc: Decouple ism_dev from SMC-D DMB registration

Quoted 29 lines · Simon Horman
Quoted 16 lines · Wen Gu

This patch tries to decouple ISM device from SMC-D DMB registration, So that the register_dmb option is not restricted to be used by ISM device.

---
  drivers/s390/net/ism_drv.c | 5 +++--
  include/net/smc.h          | 4 ++--
  net/smc/smc_ism.c          | 7 ++-----
  3 files changed, 7 insertions(+), 9 deletions(-)
drivers/s390/net/ism_drv.c +0 -0
index 8acb9eb..5eeb54d 100644

drivers/s390/net/ism_drv.c static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid, · line 802

Quoted 9 lines
drivers/s390/net/ism_drv.c +3 -2
@@ -796,9 +796,10 @@ static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid,
  }
 
  static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
-			     struct ism_client *client)
+			     void *client_priv)
  {
-	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb, client);
+	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb,
+				(struct ism_client *)client_priv);

Hi Wen Gu,

Quoted 8 lines

a minor nit from my side: there is no need to cast a void pointer to another type.

Quoted 3 lines
drivers/s390/net/ism_drv.c +0 -0
  }
 
  static int smcd_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)

...

Hi Simon Horman,

Thanks for your review and suggestions in this patch and others. I will take them in the next version.

Thanks!
Wen Gu

author reviewer
#5 lore raw

RFC PATCH v5 3/9 net/smc: Extract v2 check helper from SMC-D device registration

In reply to Wen Gu

This patch extracts v2-capable logic from the process of registering the ISM device as an SMC-D device, so that the registration process of other underlying devices can reuse it.

---
 net/smc/smc_ism.c | 27 +++++++++++++++++----------
 net/smc/smc_ism.h |  1 +
 2 files changed, 18 insertions(+), 10 deletions(-)
net/smc/smc_ism.c +17 -10
index d4709ca..8ad4c71 100644
@@ -69,6 +69,22 @@ bool smc_ism_is_v2_capable(void)
 	return smc_ism_v2_capable;
 }
 
+/* must be called under smcd_dev_list.mutex lock */
+void smc_ism_check_v2_capable(struct smcd_dev *smcd)
+{
+	u8 *system_eid = NULL;
+
+	if (!list_empty(&smcd_dev_list.list))
+		return;
+
+	system_eid = smcd->ops->get_system_eid();
+	if (smcd->ops->supports_v2()) {
+		smc_ism_v2_capable = true;
+		memcpy(smc_ism_v2_system_eid, system_eid,
+		       SMC_MAX_EID_LEN);
+	}
+}
+
 /* Set a connection using this DMBE. */
 void smc_ism_set_conn(struct smc_connection *conn)
 {
@@ -422,16 +438,7 @@ static void smcd_register_dev(struct ism_dev *ism)
 		smc_pnetid_by_table_smcd(smcd);
 
 	mutex_lock(&smcd_dev_list.mutex);
-	if (list_empty(&smcd_dev_list.list)) {
-		u8 *system_eid = NULL;
-
-		system_eid = smcd->ops->get_system_eid();
-		if (smcd->ops->supports_v2()) {
-			smc_ism_v2_capable = true;
-			memcpy(smc_ism_v2_system_eid, system_eid,
-			       SMC_MAX_EID_LEN);
-		}
-	}
+	smc_ism_check_v2_capable(smcd);
 	/* sort list: devices without pnetid before devices with pnetid */
 	if (smcd->pnetid[0])
 		list_add_tail(&smcd->list, &smcd_dev_list.list);
net/smc/smc_ism.h +1 -0
index 832b2f4..14d2e77 100644
@@ -42,6 +42,7 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size,
 void smc_ism_get_system_eid(u8 **eid);
 u16 smc_ism_get_chid(struct smcd_dev *dev);
 bool smc_ism_is_v2_capable(void);
+void smc_ism_check_v2_capable(struct smcd_dev *dev);
 int smc_ism_init(void);
 void smc_ism_exit(void);
 int smcd_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb);
Signature · 1 line
1.8.3.1
author reviewer
#6 lore raw

RFC PATCH v5 4/9 net/smc: Introduce SMC-D loopback device

In reply to Wen Gu

This patch introduces a kind of loopback device for SMC-D, thus enabling the SMC communication between two local sockets within one OS instance.

The loopback device supports basic capabilities defined by SMC-D options, and exposed as an SMC-D v2 device.

The GID of loopback device is random generated, CHID is 0xFFFF and SEID is SMCD_DEFAULT_V2_SEID.

TODO:
- The hierarchy preference of coexistent SMC-D devices.

---
 include/net/smc.h      |   6 +
 net/smc/Makefile       |   2 +-
 net/smc/af_smc.c       |  12 +-
 net/smc/smc_cdc.c      |   9 +-
 net/smc/smc_cdc.h      |   1 +
 net/smc/smc_ism.h      |   2 +
 net/smc/smc_loopback.c | 406 +++++++++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_loopback.h |  51 +++++++
 8 files changed, 486 insertions(+), 3 deletions(-)
 create mode 100644 net/smc/smc_loopback.c
 create mode 100644 net/smc/smc_loopback.h
include/net/smc.h +6 -0
index 26206d2..021ca42 100644
@@ -41,6 +41,12 @@ struct smcd_dmb {
 	dma_addr_t dma_addr;
 };
 
+struct smcd_seid {
+	u8 seid_string[24];
+	u8 serial_number[4];
+	u8 type[4];
+};
+
 #define ISM_EVENT_DMB	0
 #define ISM_EVENT_GID	1
 #define ISM_EVENT_SWR	2
net/smc/Makefile +1 -1
index 875efcd..a8c3711 100644
@@ -4,5 +4,5 @@ obj-$(CONFIG_SMC)	+= smc.o
 obj-$(CONFIG_SMC_DIAG)	+= smc_diag.o
 smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o
 smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o
-smc-y += smc_tracepoint.o
+smc-y += smc_tracepoint.o smc_loopback.o
 smc-$(CONFIG_SYSCTL) += smc_sysctl.o
net/smc/af_smc.c +11 -1
index 50c38b6..3230309 100644
@@ -53,6 +53,7 @@
 #include "smc_stats.h"
 #include "smc_tracepoint.h"
 #include "smc_sysctl.h"
+#include "smc_loopback.h"
 
 static DEFINE_MUTEX(smc_server_lgr_pending);	/* serialize link group
 						 * creation on server
@@ -3482,15 +3483,23 @@ static int __init smc_init(void)
 		goto out_sock;
 	}
 
+	rc = smc_loopback_init();
+	if (rc) {
+		pr_err("%s: smc_loopback_init fails with %d\n", __func__, rc);
+		goto out_ib;
+	}
+
 	rc = tcp_register_ulp(&smc_ulp_ops);
 	if (rc) {
 		pr_err("%s: tcp_ulp_register fails with %d\n", __func__, rc);
-		goto out_ib;
+		goto out_lo;
 	}
 
 	static_branch_enable(&tcp_have_smc);
 	return 0;
 
+out_lo:
+	smc_loopback_exit();
 out_ib:
 	smc_ib_unregister_client();
 out_sock:
@@ -3528,6 +3537,7 @@ static void __exit smc_exit(void)
 	tcp_unregister_ulp(&smc_ulp_ops);
 	sock_unregister(PF_SMC);
 	smc_core_exit();
+	smc_loopback_exit();
 	smc_ib_unregister_client();
 	smc_ism_exit();
 	destroy_workqueue(smc_close_wq);
net/smc/smc_cdc.c +8 -1
index 89105e9..2f79bac 100644
@@ -410,7 +410,14 @@ static void smc_cdc_msg_recv(struct smc_sock *smc, struct smc_cdc_msg *cdc)
  */
 static void smcd_cdc_rx_tsklet(struct tasklet_struct *t)
 {
-	struct smc_connection *conn = from_tasklet(conn, t, rx_tsklet);
+	struct smc_connection *conn =
+		from_tasklet(conn, t, rx_tsklet);
+
+	smcd_cdc_rx_handler(conn);
+}
+
+void smcd_cdc_rx_handler(struct smc_connection *conn)
+{
 	struct smcd_cdc_msg *data_cdc;
 	struct smcd_cdc_msg cdc;
 	struct smc_sock *smc;
net/smc/smc_cdc.h +1 -0
index 696cc11..11559d4 100644
@@ -301,5 +301,6 @@ int smcr_cdc_msg_send_validation(struct smc_connection *conn,
 				 struct smc_wr_buf *wr_buf);
 int smc_cdc_init(void) __init;
 void smcd_cdc_rx_init(struct smc_connection *conn);
+void smcd_cdc_rx_handler(struct smc_connection *conn);
 
 #endif /* SMC_CDC_H */
net/smc/smc_ism.h +2 -0
index 14d2e77..d18c50a 100644
@@ -15,6 +15,8 @@
 
 #include "smc.h"
 
+#define S390_ISM_IDENT_MASK 0x00FFFF
+
 struct smcd_dev_list {	/* List of SMCD devices */
 	struct list_head list;
 	struct mutex mutex;	/* Protects list of devices */
net/smc/smc_loopback.c new file +406 -0

406 lines not on this page · raw message

net/smc/smc_loopback.h new file +51 -0

51 lines not on this page · raw message

Signature · 1 line
1.8.3.1
#7 lore raw

Re: [RFC PATCH net-next v5 4/9] net/smc: Introduce SMC-D loopback device

Quoted 105 lines · Wen Gu

This patch introduces a kind of loopback device for SMC-D, thus enabling the SMC communication between two local sockets within one OS instance.

The loopback device supports basic capabilities defined by SMC-D options, and exposed as an SMC-D v2 device.

The GID of loopback device is random generated, CHID is 0xFFFF and SEID is SMCD_DEFAULT_V2_SEID.

TODO:
- The hierarchy preference of coexistent SMC-D devices.

---
 include/net/smc.h      |   6 +
 net/smc/Makefile       |   2 +-
 net/smc/af_smc.c       |  12 +-
 net/smc/smc_cdc.c      |   9 +-
 net/smc/smc_cdc.h      |   1 +
 net/smc/smc_ism.h      |   2 +
 net/smc/smc_loopback.c | 406 +++++++++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_loopback.h |  51 +++++++
 8 files changed, 486 insertions(+), 3 deletions(-)
 create mode 100644 net/smc/smc_loopback.c
 create mode 100644 net/smc/smc_loopback.h
include/net/smc.h +6 -0
index 26206d2..021ca42 100644
@@ -41,6 +41,12 @@ struct smcd_dmb {
 	dma_addr_t dma_addr;
 };
 
+struct smcd_seid {
+	u8 seid_string[24];
+	u8 serial_number[4];
+	u8 type[4];
+};
+
 #define ISM_EVENT_DMB	0
 #define ISM_EVENT_GID	1
 #define ISM_EVENT_SWR	2
net/smc/Makefile +1 -1
index 875efcd..a8c3711 100644
@@ -4,5 +4,5 @@ obj-$(CONFIG_SMC)	+= smc.o
 obj-$(CONFIG_SMC_DIAG)	+= smc_diag.o
 smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o
 smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o
-smc-y += smc_tracepoint.o
+smc-y += smc_tracepoint.o smc_loopback.o
 smc-$(CONFIG_SYSCTL) += smc_sysctl.o
net/smc/af_smc.c +11 -1
index 50c38b6..3230309 100644
@@ -53,6 +53,7 @@
 #include "smc_stats.h"
 #include "smc_tracepoint.h"
 #include "smc_sysctl.h"
+#include "smc_loopback.h"
 
 static DEFINE_MUTEX(smc_server_lgr_pending);	/* serialize link group
 						 * creation on server
@@ -3482,15 +3483,23 @@ static int __init smc_init(void)
 		goto out_sock;
 	}
 
+	rc = smc_loopback_init();
+	if (rc) {
+		pr_err("%s: smc_loopback_init fails with %d\n", __func__, rc);
+		goto out_ib;
+	}
+
 	rc = tcp_register_ulp(&smc_ulp_ops);
 	if (rc) {
 		pr_err("%s: tcp_ulp_register fails with %d\n", __func__, rc);
-		goto out_ib;
+		goto out_lo;
 	}
 
 	static_branch_enable(&tcp_have_smc);
 	return 0;
 
+out_lo:
+	smc_loopback_exit();
 out_ib:
 	smc_ib_unregister_client();
 out_sock:
@@ -3528,6 +3537,7 @@ static void __exit smc_exit(void)
 	tcp_unregister_ulp(&smc_ulp_ops);
 	sock_unregister(PF_SMC);
 	smc_core_exit();
+	smc_loopback_exit();
 	smc_ib_unregister_client();
 	smc_ism_exit();
 	destroy_workqueue(smc_close_wq);
net/smc/smc_cdc.c +0 -0
index 89105e9..2f79bac 100644

net/smc/smc_cdc.c static void smc_cdc_msg_recv(struct smc_sock *smc, struct smc_cdc_msg *cdc) · line 414

Quoted 6 lines
net/smc/smc_cdc.c +2 -1
@@ -410,7 +410,14 @@ static void smc_cdc_msg_recv(struct smc_sock *smc, struct smc_cdc_msg *cdc)
  */
 static void smcd_cdc_rx_tsklet(struct tasklet_struct *t)
 {
-	struct smc_connection *conn = from_tasklet(conn, t, rx_tsklet);
+	struct smc_connection *conn =
+		from_tasklet(conn, t, rx_tsklet);

nit: I think the above can be on one line, as it was before this patch.

Quoted 25 lines
net/smc/smc_cdc.c +6 -0
+
+	smcd_cdc_rx_handler(conn);
+}
+
+void smcd_cdc_rx_handler(struct smc_connection *conn)
+{
 	struct smcd_cdc_msg *data_cdc;
 	struct smcd_cdc_msg cdc;
 	struct smc_sock *smc;
net/smc/smc_cdc.h +1 -0
index 696cc11..11559d4 100644
@@ -301,5 +301,6 @@ int smcr_cdc_msg_send_validation(struct smc_connection *conn,
 				 struct smc_wr_buf *wr_buf);
 int smc_cdc_init(void) __init;
 void smcd_cdc_rx_init(struct smc_connection *conn);
+void smcd_cdc_rx_handler(struct smc_connection *conn);
 
 #endif /* SMC_CDC_H */
net/smc/smc_ism.h +0 -0
index 14d2e77..d18c50a 100644

net/smc/smc_ism.h line 18

Quoted 4 lines
net/smc/smc_ism.h +1 -0
@@ -15,6 +15,8 @@
 
 #include "smc.h"
 
+#define S390_ISM_IDENT_MASK 0x00FFFF

nit: I think GENMASK is appropriate here.

Quoted 4 lines
net/smc/smc_ism.h +1 -0
+
 struct smcd_dev_list {	/* List of SMCD devices */
 	struct list_head list;
 	struct mutex mutex;	/* Protects list of devices */

...

author reviewer
#8 lore raw

RFC PATCH v5 5/9 net/smc: Introduce an interface for getting DMB attribute

In reply to Wen Gu

On s390, since all OSs run on a kind of machine level hypervisor which is a partitioning hypervisor without paging, the sndbufs and DMBs in such case are unable to be mapped to the same physical memory.

However, in other scene, such as communication within the same OS instance (loopback) or between guests of a paging hypervisor (eg. KVM), the sndbufs and DMBs can be mapped to the same physical memory to avoid memory copy from sndbufs to DMBs.

So this patch introduces an interface to smcd_ops for users to judge whether DMB-map is available. And for reuse, the interface is designed to return DMB attribute, not only mappability.

---
 include/net/smc.h | 5 +++++
 net/smc/smc_ism.c | 8 ++++++++
 net/smc/smc_ism.h | 1 +
 3 files changed, 14 insertions(+)
include/net/smc.h +5 -0
index 021ca42..e39ac41 100644
@@ -55,6 +55,10 @@ struct smcd_seid {
 
 #define ISM_ERROR	0xFFFF
 
+enum {
+	ISM_DMB_MAPPABLE = 0,
+};
+
 struct smcd_dev;
 
 struct smcd_ops {
@@ -77,6 +81,7 @@ struct smcd_ops {
 	u64 (*get_local_gid)(struct smcd_dev *dev);
 	u16 (*get_chid)(struct smcd_dev *dev);
 	struct device* (*get_dev)(struct smcd_dev *dev);
+	int (*get_dev_dmb_attr)(struct smcd_dev *dev);
 };
 
 struct smcd_dev {
net/smc/smc_ism.c +8 -0
index 8ad4c71..1d97e77 100644
@@ -213,6 +213,14 @@ int smc_ism_unregister_dmb(struct smcd_dev *smcd, struct smc_buf_desc *dmb_desc)
 	return rc;
 }
 
+bool smc_ism_dmb_mappable(struct smcd_dev *smcd)
+{
+	if (smcd->ops->get_dev_dmb_attr &&
+	    (smcd->ops->get_dev_dmb_attr(smcd) & (1 << ISM_DMB_MAPPABLE)))
+		return true;
+	return false;
+}
+
 int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 			 struct smc_buf_desc *dmb_desc)
 {
net/smc/smc_ism.h +1 -0
index d18c50a..0b1913a 100644
@@ -40,6 +40,7 @@ struct smc_ism_vlanid {			/* VLAN id set on ISM device */
 int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size,
 			 struct smc_buf_desc *dmb_desc);
 int smc_ism_unregister_dmb(struct smcd_dev *dev, struct smc_buf_desc *dmb_desc);
+bool smc_ism_dmb_mappable(struct smcd_dev *smcd);
 int smc_ism_signal_shutdown(struct smc_link_group *lgr);
 void smc_ism_get_system_eid(u8 **eid);
 u16 smc_ism_get_chid(struct smcd_dev *dev);
Signature · 1 line
1.8.3.1
#9 lore raw

Re: [RFC PATCH net-next v5 5/9] net/smc: Introduce an interface for getting DMB attribute

Quoted 14 lines · Wen Gu

On s390, since all OSs run on a kind of machine level hypervisor which is a partitioning hypervisor without paging, the sndbufs and DMBs in such case are unable to be mapped to the same physical memory.

However, in other scene, such as communication within the same OS instance (loopback) or between guests of a paging hypervisor (eg. KVM), the sndbufs and DMBs can be mapped to the same physical memory to avoid memory copy from sndbufs to DMBs.

So this patch introduces an interface to smcd_ops for users to judge whether DMB-map is available. And for reuse, the interface is designed to return DMB attribute, not only mappability.

...

net/smc/smc_ism.c int smc_ism_unregister_dmb(struct smcd_dev *smcd, struct smc_buf_desc *dmb_desc) · line 219

Quoted 12 lines
net/smc/smc_ism.c +4 -0
index 8ad4c71..1d97e77 100644
@@ -213,6 +213,14 @@ int smc_ism_unregister_dmb(struct smcd_dev *smcd, struct smc_buf_desc *dmb_desc)
 	return rc;
 }
 
+bool smc_ism_dmb_mappable(struct smcd_dev *smcd)
+{
+	if (smcd->ops->get_dev_dmb_attr &&
+	    (smcd->ops->get_dev_dmb_attr(smcd) & (1 << ISM_DMB_MAPPABLE)))

nit: this could use BIT(ISM_DMB_MAPPABLE)

Quoted 7 lines
net/smc/smc_ism.c +4 -0
+		return true;
+	return false;
+}
+
 int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 			 struct smc_buf_desc *dmb_desc)
 {

...

author reviewer
#10 lore raw

RFC PATCH v5 6/9 net/smc: Introudce interfaces for DMB attach and detach

In reply to Wen Gu

This patch extends smcd_ops, adding two more semantic for SMC-D DMB:

- attach_dmb:
Attach an already registered DMB to a specific buf_desc, so that we can refer to the DMB through this buf_desc.

- detach_dmb:
Reverse operation of attach_dmb. detach the DMB from the buf_desc.

This interface extension is to prepare for the avoidance of memory copy from sndbuf to RMB with SMC-D device whose DMBs has ISM_DMB_MAPPABLE attribute.

---
 include/net/smc.h |  2 ++
 net/smc/smc_ism.c | 31 +++++++++++++++++++++++++++++++
 net/smc/smc_ism.h |  2 ++
 3 files changed, 35 insertions(+)
include/net/smc.h +2 -0
index e39ac41..0132522 100644
@@ -67,6 +67,8 @@ struct smcd_ops {
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
 			    void *client_priv);
 	int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
+	int (*attach_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
+	int (*detach_dmb)(struct smcd_dev *dev, u64 token);
 	int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
 	int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
 	int (*set_vlan_required)(struct smcd_dev *dev);
net/smc/smc_ism.c +31 -0
index 1d97e77..925af13 100644
@@ -243,6 +243,37 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 	return rc;
 }
 
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc)
+{
+	struct smcd_dmb dmb;
+	int rc = 0;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_tok = token;
+
+	if (!dev->ops->attach_dmb)
+		return -EINVAL;
+
+	rc = dev->ops->attach_dmb(dev, &dmb);
+	if (!rc) {
+		dmb_desc->sba_idx = dmb.sba_idx;
+		dmb_desc->token = dmb.dmb_tok;
+		dmb_desc->cpu_addr = dmb.cpu_addr;
+		dmb_desc->dma_addr = dmb.dma_addr;
+		dmb_desc->len = dmb.dmb_len;
+	}
+	return rc;
+}
+
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token)
+{
+	if (!dev->ops->detach_dmb)
+		return -EINVAL;
+
+	return dev->ops->detach_dmb(dev, token);
+}
+
 static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
 				  struct sk_buff *skb,
 				  struct netlink_callback *cb)
net/smc/smc_ism.h +2 -0
index 0b1913a..31eb010 100644
@@ -41,6 +41,8 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size,
 			 struct smc_buf_desc *dmb_desc);
 int smc_ism_unregister_dmb(struct smcd_dev *dev, struct smc_buf_desc *dmb_desc);
 bool smc_ism_dmb_mappable(struct smcd_dev *smcd);
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token, struct smc_buf_desc *dmb_desc);
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token);
 int smc_ism_signal_shutdown(struct smc_link_group *lgr);
 void smc_ism_get_system_eid(u8 **eid);
 u16 smc_ism_get_chid(struct smcd_dev *dev);
Signature · 1 line
1.8.3.1
author reviewer
#11 lore raw

RFC PATCH v5 7/9 net/smc: Avoid data copy from sndbuf to peer RMB in SMC-D

In reply to Wen Gu

This patch aims to avoid data copy from local sndbuf to peer RMB by attaching local sndbuf to peer RMB when DMBs have ISM_DMB_MAPPABLE attribute.

After this, local sndbuf and peer RMB share the same physical memory.

 +----------+                     +----------+
 | socket A |                     | socket B |
 +----------+                     +----------+
       |                               ^
       |         +---------+           |
  regard as      |         | ----------|
  local sndbuf   |  B's    |     regard as
       |         |  RMB    |     local RMB
       |-------> |         |
                 +---------+

1. Actions on local RMB.

     a. Create or reuse RMB when connection is created;
     b. Unuse RMB when connection is freed;
     c. Free RMB when link group is freed;

2. Actions on local sndbuf.

a. Attach local sndbuf to peer RMB by the rtoken exchanged through
CLC message. Since then, accessing local sndbuf is equivalent to accessing peer RMB
b. sndbuf_desc is exclusive to specific connection and won't be
added to lgr buffer pool for reuse.
c. Local sndbuf is detached from peer RMB and freed when connection
is freed.

Therefore, the data written to local sndbuf will directly reach peer RMB.

---
 net/smc/af_smc.c   | 14 +++++++++++
 net/smc/smc_core.c | 70 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
 net/smc/smc_core.h |  1 +
 3 files changed, 84 insertions(+), 1 deletion(-)
net/smc/af_smc.c +14 -0
index 3230309..0956cf5 100644
@@ -1378,6 +1378,12 @@ static int smc_connect_ism(struct smc_sock *smc,
 	}
 
 	smc_conn_save_peer_info(smc, aclc);
+
+	if (smc_ism_dmb_mappable(smc->conn.lgr->smcd)) {
+		rc = smcd_buf_attach(smc);
+		if (rc)
+			goto connect_abort;
+	}
 	smc_close_init(smc);
 	smc_rx_init(smc);
 	smc_tx_init(smc);
@@ -2436,6 +2442,14 @@ static void smc_listen_work(struct work_struct *work)
 		mutex_unlock(&smc_server_lgr_pending);
 	}
 	smc_conn_save_peer_info(new_smc, cclc);
+
+	if (ini->is_smcd &&
+	    smc_ism_dmb_mappable(new_smc->conn.lgr->smcd)) {
+		rc = smcd_buf_attach(new_smc);
+		if (rc)
+			goto out_decl;
+	}
+
 	smc_listen_out_connected(new_smc);
 	SMC_STAT_SERV_SUCC_INC(sock_net(newclcsock->sk), ini);
 	goto out_free;
net/smc/smc_core.c +69 -1

102 lines not on this page · raw message

net/smc/smc_core.h +1 -0
index 1645fba..e52cf70 100644
@@ -524,6 +524,7 @@ void smc_smcd_terminate(struct smcd_dev *dev, u64 peer_gid,
 void smc_smcd_terminate_all(struct smcd_dev *dev);
 void smc_smcr_terminate_all(struct smc_ib_device *smcibdev);
 int smc_buf_create(struct smc_sock *smc, bool is_smcd);
+int smcd_buf_attach(struct smc_sock *smc);
 int smc_uncompress_bufsize(u8 compressed);
 int smc_rmb_rtoken_handling(struct smc_connection *conn, struct smc_link *link,
 			    struct smc_clc_msg_accept_confirm *clc);
Signature · 1 line
1.8.3.1
author reviewer
#12 lore raw

RFC PATCH v5 8/9 net/smc: Modify cursor update logic when using mappable DMB

In reply to Wen Gu

Since local sndbuf shares the same physical memory region with peer RMB when using mappable DMBs, the cursor update logic needs to be adapted.

The main concern is to ensure that the data written by local to this memory region won't overwrite the data that has not been consumed by the peer.

So in this scene, the fin_curs and sndbuf_space that were originally updated when sending out CDC message are not updated until the cons_curs update from the peer is received.

---
 net/smc/smc_cdc.c | 50 +++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 39 insertions(+), 11 deletions(-)
net/smc/smc_cdc.c +39 -11
index 2f79bac..915b8e7 100644
@@ -18,6 +18,7 @@
 #include "smc_tx.h"
 #include "smc_rx.h"
 #include "smc_close.h"
+#include "smc_ism.h"
 
 /********************************** send *************************************/
 
@@ -256,17 +257,24 @@ int smcd_cdc_msg_send(struct smc_connection *conn)
 		return rc;
 	smc_curs_copy(&conn->rx_curs_confirmed, &curs, conn);
 	conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0;
-	/* Calculate transmitted data and increment free send buffer space */
-	diff = smc_curs_diff(conn->sndbuf_desc->len, &conn->tx_curs_fin,
-			     &conn->tx_curs_sent);
-	/* increased by confirmed number of bytes */
-	smp_mb__before_atomic();
-	atomic_add(diff, &conn->sndbuf_space);
-	/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
-	smp_mb__after_atomic();
-	smc_curs_copy(&conn->tx_curs_fin, &conn->tx_curs_sent, conn);
+	if (!smc_ism_dmb_mappable(conn->lgr->smcd)) {
+		/* If local sndbuf has been mapped to peer RMB, then
+		 * don't update the tx_curs_fin and sndbuf_space until
+		 * peer has consumed the data in RMB.
+		 */
 
-	smc_tx_sndbuf_nonfull(smc);
+		/* Calculate transmitted data and increment free send buffer space */
+		diff = smc_curs_diff(conn->sndbuf_desc->len, &conn->tx_curs_fin,
+				     &conn->tx_curs_sent);
+		/* increased by confirmed number of bytes */
+		smp_mb__before_atomic();
+		atomic_add(diff, &conn->sndbuf_space);
+		/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+		smp_mb__after_atomic();
+		smc_curs_copy(&conn->tx_curs_fin, &conn->tx_curs_sent, conn);
+
+		smc_tx_sndbuf_nonfull(smc);
+	}
 	return rc;
 }
 
@@ -324,7 +332,7 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc,
 {
 	union smc_host_cursor cons_old, prod_old;
 	struct smc_connection *conn = &smc->conn;
-	int diff_cons, diff_prod;
+	int diff_cons, diff_prod, diff_tx;
 
 	smc_curs_copy(&prod_old, &conn->local_rx_ctrl.prod, conn);
 	smc_curs_copy(&cons_old, &conn->local_rx_ctrl.cons, conn);
@@ -340,6 +348,26 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc,
 		atomic_add(diff_cons, &conn->peer_rmbe_space);
 		/* guarantee 0 <= peer_rmbe_space <= peer_rmbe_size */
 		smp_mb__after_atomic();
+
+		if (conn->lgr->is_smcd &&
+		    smc_ism_dmb_mappable(conn->lgr->smcd)) {
+			/* If local sndbuf has been mapped to peer RMB, then
+			 * update tx_curs_fin and sndbuf_space when peer has
+			 * consumed the data in it's RMB.
+			 */
+
+			/* calculate peer rmb consumed data */
+			diff_tx = smc_curs_diff(conn->sndbuf_desc->len, &conn->tx_curs_fin,
+						&conn->local_rx_ctrl.cons);
+			/* increase local sndbuf space and fin_curs */
+			smp_mb__before_atomic();
+			atomic_add(diff_tx, &conn->sndbuf_space);
+			/* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */
+			smp_mb__after_atomic();
+			smc_curs_copy(&conn->tx_curs_fin, &conn->local_rx_ctrl.cons, conn);
+
+			smc_tx_sndbuf_nonfull(smc);
+		}
 	}
 
 	diff_prod = smc_curs_diff(conn->rmb_desc->len, &prod_old,
Signature · 1 line
1.8.3.1
author reviewer
#13 lore raw

RFC PATCH v5 9/9 net/smc: Add interface implementation of loopback device

In reply to Wen Gu

This patch completes the specific implementation of loopback device for the newly added SMC-D DMB-related interface.

The loopback device always provides mappable DMB because the device users are in the same OS instance.

---
 net/smc/smc_loopback.c | 101 +++++++++++++++++++++++++++++++++++++++++++++----
 net/smc/smc_loopback.h |   5 +++
 2 files changed, 98 insertions(+), 8 deletions(-)
net/smc/smc_loopback.c +93 -8
index 4197201..dc1909b 100644
@@ -110,6 +110,7 @@ static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
 	}
 	dmb_node->len = dmb->dmb_len;
 	dmb_node->dma_addr = SMC_LO_DMA_ADDR_INVALID;
+	refcount_set(&dmb_node->refcnt, 1);
 
 again:
 	/* add new dmb into hash table */
@@ -124,6 +125,7 @@ static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
 	}
 	hash_add(ldev->dmb_ht, &dmb_node->list, dmb_node->token);
 	write_unlock(&ldev->dmb_ht_lock);
+	atomic_inc(&ldev->dmb_cnt);
 
 	dmb->sba_idx = dmb_node->sba_idx;
 	dmb->dmb_tok = dmb_node->token;
@@ -145,11 +147,12 @@ static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)
 	struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node;
 	struct smc_lo_dev *ldev = smcd->priv;
 
-	/* remove dmb from hash table */
+	/* find dmb from hash table */
 	write_lock(&ldev->dmb_ht_lock);
 	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb->dmb_tok) {
 		if (tmp_node->token == dmb->dmb_tok) {
 			dmb_node = tmp_node;
+			dmb_node->freeing = 1;
 			break;
 		}
 	}
@@ -157,16 +160,85 @@ static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)
 		write_unlock(&ldev->dmb_ht_lock);
 		return -EINVAL;
 	}
+	write_unlock(&ldev->dmb_ht_lock);
+
+	/* wait for dmb refcnt to be 0 */
+	if (!refcount_dec_and_test(&dmb_node->refcnt))
+		wait_event(ldev->dmbs_release, !refcount_read(&dmb_node->refcnt));
+
+	/* remove dmb from hash table */
+	write_lock(&ldev->dmb_ht_lock);
 	hash_del(&dmb_node->list);
 	write_unlock(&ldev->dmb_ht_lock);
 
 	clear_bit(dmb_node->sba_idx, ldev->sba_idx_mask);
+
 	kfree(dmb_node->cpu_addr);
 	kfree(dmb_node);
 
+	if (atomic_dec_and_test(&ldev->dmb_cnt))
+		wake_up(&ldev->ldev_release);
 	return 0;
 }
 
+static int smc_lo_attach_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)
+{
+	struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node;
+	struct smc_lo_dev *ldev = smcd->priv;
+
+	/* find dmb_node according to dmb->dmb_tok */
+	read_lock(&ldev->dmb_ht_lock);
+	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb->dmb_tok) {
+		if (tmp_node->token == dmb->dmb_tok && !tmp_node->freeing) {
+			dmb_node = tmp_node;
+			break;
+		}
+	}
+	if (!dmb_node) {
+		read_unlock(&ldev->dmb_ht_lock);
+		return -EINVAL;
+	}
+	refcount_inc(&dmb_node->refcnt);
+	read_unlock(&ldev->dmb_ht_lock);
+
+	/* provide dmb information */
+	dmb->sba_idx = dmb_node->sba_idx;
+	dmb->dmb_tok = dmb_node->token;
+	dmb->cpu_addr = dmb_node->cpu_addr;
+	dmb->dma_addr = dmb_node->dma_addr;
+	dmb->dmb_len = dmb_node->len;
+	return 0;
+}
+
+static int smc_lo_detach_dmb(struct smcd_dev *smcd, u64 token)
+{
+	struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node;
+	struct smc_lo_dev *ldev = smcd->priv;
+
+	/* find dmb_node according to dmb->dmb_tok */
+	read_lock(&ldev->dmb_ht_lock);
+	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, token) {
+		if (tmp_node->token == token) {
+			dmb_node = tmp_node;
+			break;
+		}
+	}
+	if (!dmb_node) {
+		read_unlock(&ldev->dmb_ht_lock);
+		return -EINVAL;
+	}
+	read_unlock(&ldev->dmb_ht_lock);
+
+	if (refcount_dec_and_test(&dmb_node->refcnt))
+		wake_up_all(&ldev->dmbs_release);
+	return 0;
+}
+
+static int smc_lo_get_dev_dmb_attr(struct smcd_dev *smcd)
+{
+	return (1 << ISM_DMB_MAPPABLE);
+}
+
 static int smc_lo_add_vlan_id(struct smcd_dev *smcd, u64 vlan_id)
 {
 	return -EOPNOTSUPP;
@@ -199,7 +271,15 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, unsigned int idx
 {
 	struct smc_lo_dmb_node *rmb_node = NULL, *tmp_node;
 	struct smc_lo_dev *ldev = smcd->priv;
-
+	struct smc_connection *conn;
+
+	if (!sf) {
+		/* local sndbuf shares the same physical memory with
+		 * peer RMB, so no need to copy data from local sndbuf
+		 * to peer RMB.
+		 */
+		return 0;
+	}
 	read_lock(&ldev->dmb_ht_lock);
 	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) {
 		if (tmp_node->token == dmb_tok) {
@@ -215,13 +295,10 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, unsigned int idx
 
 	memcpy((char *)rmb_node->cpu_addr + offset, data, size);
 
-	if (sf) {
-		struct smc_connection *conn =
-			smcd->conn[rmb_node->sba_idx];
+	conn = smcd->conn[rmb_node->sba_idx];
+	if (conn && !conn->killed)
+		smcd_cdc_rx_handler(conn);
 
-		if (conn && !conn->killed)
-			smcd_cdc_rx_handler(conn);
-	}
 	return 0;
 }
 
@@ -254,6 +331,8 @@ static struct device *smc_lo_get_dev(struct smcd_dev *smcd)
 	.query_remote_gid = smc_lo_query_rgid,
 	.register_dmb = smc_lo_register_dmb,
 	.unregister_dmb = smc_lo_unregister_dmb,
+	.attach_dmb = smc_lo_attach_dmb,
+	.detach_dmb = smc_lo_detach_dmb,
 	.add_vlan_id = smc_lo_add_vlan_id,
 	.del_vlan_id = smc_lo_del_vlan_id,
 	.set_vlan_required = smc_lo_set_vlan_required,
@@ -265,6 +344,7 @@ static struct device *smc_lo_get_dev(struct smcd_dev *smcd)
 	.get_local_gid = smc_lo_get_local_gid,
 	.get_chid = smc_lo_get_chid,
 	.get_dev = smc_lo_get_dev,
+	.get_dev_dmb_attr = smc_lo_get_dev_dmb_attr,
 };
 
 static struct smcd_dev *smcd_lo_alloc_dev(const struct smcd_ops *ops,
@@ -345,6 +425,9 @@ static int smc_lo_dev_init(struct smc_lo_dev *ldev)
 	smc_lo_generate_id(ldev);
 	rwlock_init(&ldev->dmb_ht_lock);
 	hash_init(ldev->dmb_ht);
+	atomic_set(&ldev->dmb_cnt, 0);
+	init_waitqueue_head(&ldev->dmbs_release);
+	init_waitqueue_head(&ldev->ldev_release);
 
 	return smcd_lo_register_dev(ldev);
 }
@@ -383,6 +466,8 @@ static int smc_lo_dev_probe(void)
 static void smc_lo_dev_exit(struct smc_lo_dev *ldev)
 {
 	smcd_lo_unregister_dev(ldev);
+	if (atomic_read(&ldev->dmb_cnt))
+		wait_event(ldev->ldev_release, !atomic_read(&ldev->dmb_cnt));
 }
 
 static void smc_lo_dev_remove(void)
net/smc/smc_loopback.h +5 -0
index 9d34aba..e0bf044 100644
@@ -33,6 +33,8 @@ struct smc_lo_dmb_node {
 	u32 sba_idx;
 	void *cpu_addr;
 	dma_addr_t dma_addr;
+	refcount_t refcnt;
+	u8 freeing : 1;
 };
 
 struct smc_lo_dev {
@@ -43,6 +45,9 @@ struct smc_lo_dev {
 	DECLARE_BITMAP(sba_idx_mask, SMC_LODEV_MAX_DMBS);
 	rwlock_t dmb_ht_lock;
 	DECLARE_HASHTABLE(dmb_ht, SMC_LODEV_MAX_DMBS_BUCKETS);
+	atomic_t dmb_cnt;
+	wait_queue_head_t dmbs_release;
+	wait_queue_head_t ldev_release;
 };
 
 int smc_loopback_init(void);
Signature · 1 line
1.8.3.1
author reviewer
#14 lore raw
In reply to Wen Gu

Hi Wenjia & Jan:

Does this version work fine on your platform?

And any comments on this version? :)

Best regards,
Wen Gu

Quoted 171 lines · Wen Gu

Hi, all

# Background

The background and previous discussion can be referred from [1]~[3].

We found SMC-D can be used to accelerate OS internal communication, such as loopback or between two containers within the same OS instance. So this patch set provides a kind of SMC-D dummy device (we call it the SMC-D loopback device) to emulate an ISM device, so that SMC-D can also be used on architectures other than s390. The SMC-D loopback device are designed as a system global device, visible to all containers.

# Design

This patch set basically follows the design of the previous version.

Patch #1/9 ~ #3/9 attempt to decouple ISM-related structures from the SMC-D generalized code and extract some helpers to make SMC-D protocol compatible with devices other than s390 ISM device.

Patch #4/9 introduces a kind of loopback device, which is defined as SMC-Dv2 device and designed to provide communication between SMC sockets on the same OS instance.

  +-------------------------------------------+
  |  +--------------+       +--------------+  |
  |  | SMC socket A |       | SMC socket B |  |
  |  +--------------+       +--------------+  |
  |       ^                         ^         |
  |       |    +----------------+   |         |
  |       |    |   SMC stack    |   |         |
  |       +--->| +------------+ |<--|         |
  |            | |   dummy    | |             |
  |            | |   device   | |             |
  |            +-+------------+-+             |
  |                   OS                      |
  +-------------------------------------------+

Patch #5/9 ~ #8/9 expand SMC-D protocol interface (smcd_ops) for scenarios where SMC-D is used to communicate within VM (loopback here) or between VMs on the same host (based on virtio-ism device, see [4]). What these scenarios have in common is that the local sndbuf and peer RMB can be mapped to same physical memory region, so the data copy between the local sndbuf and peer RMB can be omitted. Performance improvement brought by this extension can be found in # Benchmark Test.

  +----------+                     +----------+
  | socket A |                     | socket B |
  +----------+                     +----------+
        |                               ^
        |         +---------+           |
   regard as      |         | ----------|
   local sndbuf   |  B's    |     regard as
        |         |  RMB    |     local RMB
        |-------> |         |
                  +---------+

Patch #9/9 realizes the support of loopback device for the above-mentioned expanded SMC-D protocol interface.

# Benchmark Test

* Test environments:
- VM with Intel Xeon Platinum 8 core 2.50GHz, 16 GiB mem.
- SMC sndbuf/RMB size 1MB.

* Test object:
- TCP lo: run on TCP loopback.
- domain: run on UNIX domain.
- SMC lo: run on SMC loopback device with patch #1/9 ~ #4/9.
- SMC lo-nocpy: run on SMC loopback device with patch #1/9 ~ #9/9.

1. ipc-benchmark (see [5])

- ./<foo> -c 1000000 -s 100

TCP-lo domain SMC-lo SMC-lo-nocpy
Message
rate (msg/s) 79025 115736(+46.45%) 146760(+85.71%) 149800(+89.56%)

2. sockperf

- serv: <smc_run> taskset -c <cpu> sockperf sr --tcp
- clnt: <smc_run> taskset -c <cpu> sockperf { tp | pp } --tcp --msg-size={ 64000 for tp | 14 for pp } -i 127.0.0.1 -t 30

TCP-lo SMC-lo SMC-lo-nocpy
Bandwidth(MBps) 4822.388 4940.918(+2.56%) 8086.67(+67.69%)
Latency(us) 6.298 3.352(-46.78%) 3.35(-46.81%)

3. iperf3

- serv: <smc_run> taskset -c <cpu> iperf3 -s
- clnt: <smc_run> taskset -c <cpu> iperf3 -c 127.0.0.1 -t 15

TCP-lo SMC-lo SMC-lo-nocpy
Bitrate(Gb/s) 40.7 40.5(-0.49%) 72.4(+77.89%)

4. nginx/wrk

- serv: <smc_run> nginx
- clnt: <smc_run> wrk -t 8 -c 500 -d 30 http://127.0.0.1:80

TCP-lo SMC-lo SMC-lo-nocpy
Requests/s 155994.57 214544.79(+37.53%) 215538.55(+38.17%)

v5->v4
1. The loopback device generates SEID in the same way as the ISM devices when coexisting with ISM devices and uses a default fixed SEID in other cases.
2. Ensure each DMB token of the same loopback device is unique.
3. Fixe a crash caused by setting smcd_ops->signal_event interface to NULL.
4. Fixe a compilation warning complained by kernel test rebot.

v4->v3
1. Rebase to the latest net-next;
2. Introduce SEID helper. SMC-D loopback will return SMCD_DEFAULT_V2_SEID. And if it coexist with ISM device, the SEID of ISM device will overwrite SMCD_DEFAULT_V2_SEID as smc_ism_v2_system_eid.
3. Won't remove dmb_node from hashtable until no sndbuf attaching to it.

Something postponed in this version
1. Hierarchy perference of SMC-D devices when loopback and ISM devices coexist, which will be determinated after comparing the performance of loopback and ISM.

v3->v2
1. Adapt new generalized interface provided by [2];
2. Select loopback device through SMC-D v2 protocol;
3. Split the loopback-related implementation and generic implementation into different patches more reasonably.

v1->v2
1. Fix some build WARNINGs complained by kernel test rebot

2. Add iperf3 test data.

[1] https://lore.kernel.org/netdev/[email protected]/
[2] https://lore.kernel.org/netdev/[email protected]/
[3] https://lore.kernel.org/netdev/[email protected]/
[4] https://lore.kernel.org/all/[email protected]/
[5] https://github.com/goldsborough/ipc-bench

Wen Gu (9):
net/smc: Decouple ism_dev from SMC-D device dump net/smc: Decouple ism_dev from SMC-D DMB registration net/smc: Extract v2 check helper from SMC-D device registration net/smc: Introduce SMC-D loopback device
net/smc: Introduce an interface for getting DMB attribute net/smc: Introudce interfaces for DMB attach and detach net/smc: Avoid data copy from sndbuf to peer RMB in SMC-D net/smc: Modify cursor update logic when using mappable DMB net/smc: Add interface implementation of loopback device

  drivers/s390/net/ism_drv.c |   5 +-
  include/net/smc.h          |  18 +-
  net/smc/Makefile           |   2 +-
  net/smc/af_smc.c           |  26 ++-
  net/smc/smc_cdc.c          |  59 ++++--
  net/smc/smc_cdc.h          |   1 +
  net/smc/smc_core.c         |  70 ++++++-
  net/smc/smc_core.h         |   1 +
  net/smc/smc_ism.c          |  79 ++++++--
  net/smc/smc_ism.h          |   6 +
  net/smc/smc_loopback.c     | 491 +++++++++++++++++++++++++++++++++++++++++++++
  net/smc/smc_loopback.h     |  56 ++++++
  12 files changed, 777 insertions(+), 37 deletions(-)

create mode 100644 net/smc/smc_loopback.c
create mode 100644 net/smc/smc_loopback.h

#15 lore raw
Quoted 5 lines · Wen Gu

Hi Wenjia & Jan:

Does this version work fine on your platform?

And any comments on this version? :)

Hi Wen Gu,

as background for the others: IBM & Alibaba are having calls regarding SMC since Alibaba shows a lot of interest and there is a lot to discuss (as you may have noticed).

As we have agreed in our calls, the information shared about SMC-D in specific doesn't make it easy to understand the underlying concepts (e.g. role of GID, EIDs, DMB tokens, etc.). That's why one of the next calls is dedicated to explain some of the concepts in depth to you and see where we can improve our flow of information to the community in the future. After that we can go over this RFC with a shared understanding of the concepts.

Thank you
- Jan

Quoted 181 lines

Best regards,
Wen Gu

Quoted 173 lines · Wen Gu

Hi, all

# Background

The background and previous discussion can be referred from [1]~[3].

We found SMC-D can be used to accelerate OS internal communication, such as loopback or between two containers within the same OS instance. So this patch set provides a kind of SMC-D dummy device (we call it the SMC-D loopback device) to emulate an ISM device, so that SMC-D can also be used on architectures other than s390. The SMC-D loopback device are designed as a system global device, visible to all containers.

# Design

This patch set basically follows the design of the previous version.

Patch #1/9 ~ #3/9 attempt to decouple ISM-related structures from the SMC-D generalized code and extract some helpers to make SMC-D protocol compatible with devices other than s390 ISM device.

Patch #4/9 introduces a kind of loopback device, which is defined as SMC-Dv2 device and designed to provide communication between SMC sockets on the same OS instance.

  +-------------------------------------------+
  |  +--------------+       +--------------+  |
  |  | SMC socket A |       | SMC socket B |  |
  |  +--------------+       +--------------+  |
  |       ^                         ^         |
  |       |    +----------------+   |         |
  |       |    |   SMC stack    |   |         |
  |       +--->| +------------+ |<--|         |
  |            | |   dummy    | |             |
  |            | |   device   | |             |
  |            +-+------------+-+             |
  |                   OS                      |
  +-------------------------------------------+

Patch #5/9 ~ #8/9 expand SMC-D protocol interface (smcd_ops) for scenarios where SMC-D is used to communicate within VM (loopback here) or between VMs on the same host (based on virtio-ism device, see [4]). What these scenarios have in common is that the local sndbuf and peer RMB can be mapped to same physical memory region, so the data copy between the local sndbuf and peer RMB can be omitted. Performance improvement brought by this extension can be found in # Benchmark Test.

  +----------+                     +----------+
  | socket A |                     | socket B |
  +----------+                     +----------+
        |                               ^
        |         +---------+           |
   regard as      |         | ----------|
   local sndbuf   |  B's    |     regard as
        |         |  RMB    |     local RMB
        |-------> |         |
                  +---------+

Patch #9/9 realizes the support of loopback device for the above-mentioned expanded SMC-D protocol interface.

# Benchmark Test

  * Test environments:
       - VM with Intel Xeon Platinum 8 core 2.50GHz, 16 GiB mem.
       - SMC sndbuf/RMB size 1MB.

  * Test object:
       - TCP lo: run on TCP loopback.
       - domain: run on UNIX domain.
       - SMC lo: run on SMC loopback device with patch #1/9 ~ #4/9.
       - SMC lo-nocpy: run on SMC loopback device with patch #1/9 ~ #9/9.

1. ipc-benchmark (see [5])

  - ./<foo> -c 1000000 -s 100

                     TCP-lo              domain SMC-lo          SMC-lo-nocpy
Message
rate (msg/s)         79025      115736(+46.45%) 146760(+85.71%)       149800(+89.56%)

2. sockperf

  - serv: <smc_run> taskset -c <cpu> sockperf sr --tcp
  - clnt: <smc_run> taskset -c <cpu> sockperf { tp | pp } --tcp --msg-size={ 64000 for tp | 14 for pp } -i 127.0.0.1 -t 30

                     TCP-lo                  SMC-lo SMC-lo-nocpy
Bandwidth(MBps)   4822.388        4940.918(+2.56%) 8086.67(+67.69%)
Latency(us)          6.298          3.352(-46.78%) 3.35(-46.81%)

3. iperf3

  - serv: <smc_run> taskset -c <cpu> iperf3 -s
  - clnt: <smc_run> taskset -c <cpu> iperf3 -c 127.0.0.1 -t 15

                     TCP-lo                  SMC-lo SMC-lo-nocpy
Bitrate(Gb/s)         40.7            40.5(-0.49%) 72.4(+77.89%)

4. nginx/wrk

  - serv: <smc_run> nginx
  - clnt: <smc_run> wrk -t 8 -c 500 -d 30 http://127.0.0.1:80

                     TCP-lo                  SMC-lo SMC-lo-nocpy
Requests/s       155994.57      214544.79(+37.53%) 215538.55(+38.17%)

v5->v4
  1. The loopback device generates SEID in the same way as the ISM devices when coexisting with ISM devices and uses a default fixed SEID in other cases.
  2. Ensure each DMB token of the same loopback device is unique.
  3. Fixe a crash caused by setting smcd_ops->signal_event interface to NULL.
  4. Fixe a compilation warning complained by kernel test rebot.

v4->v3
  1. Rebase to the latest net-next;
  2. Introduce SEID helper. SMC-D loopback will return SMCD_DEFAULT_V2_SEID. And if it coexist with ISM device, the SEID of ISM device will overwrite SMCD_DEFAULT_V2_SEID as smc_ism_v2_system_eid.
  3. Won't remove dmb_node from hashtable until no sndbuf attaching to it.

  Something postponed in this version
  1. Hierarchy perference of SMC-D devices when loopback and ISM devices coexist, which will be determinated after comparing the performance of loopback and ISM.

v3->v2
  1. Adapt new generalized interface provided by [2];
  2. Select loopback device through SMC-D v2 protocol;
  3. Split the loopback-related implementation and generic implementation into different patches more reasonably.

v1->v2
  1. Fix some build WARNINGs complained by kernel test rebot

  2. Add iperf3 test data.

[1] https://lore.kernel.org/netdev/[email protected]/
[2] https://lore.kernel.org/netdev/[email protected]/
[3] https://lore.kernel.org/netdev/[email protected]/
[4] https://lore.kernel.org/all/[email protected]/
[5] https://github.com/goldsborough/ipc-bench

Wen Gu (9):
   net/smc: Decouple ism_dev from SMC-D device dump net/smc: Decouple ism_dev from SMC-D DMB registration net/smc: Extract v2 check helper from SMC-D device registration net/smc: Introduce SMC-D loopback device
   net/smc: Introduce an interface for getting DMB attribute net/smc: Introudce interfaces for DMB attach and detach net/smc: Avoid data copy from sndbuf to peer RMB in SMC-D net/smc: Modify cursor update logic when using mappable DMB net/smc: Add interface implementation of loopback device

  drivers/s390/net/ism_drv.c |   5 +-
  include/net/smc.h          |  18 +-
  net/smc/Makefile           |   2 +-
  net/smc/af_smc.c           |  26 ++-
  net/smc/smc_cdc.c          |  59 ++++--
  net/smc/smc_cdc.h          |   1 +
  net/smc/smc_core.c         |  70 ++++++-
  net/smc/smc_core.h         |   1 +
  net/smc/smc_ism.c          |  79 ++++++--
  net/smc/smc_ism.h          |   6 +
  net/smc/smc_loopback.c     | 491
+++++++++++++++++++++++++++++++++++++++++++++
  net/smc/smc_loopback.h     |  56 ++++++

  12 files changed, 777 insertions(+), 37 deletions(-)

  create mode 100644 net/smc/smc_loopback.c
  create mode 100644 net/smc/smc_loopback.h

author reviewer
#16 lore raw
Quoted 9 lines · Jan Karcher
Quoted 5 lines · Wen Gu

Hi Wenjia & Jan:

Does this version work fine on your platform?

And any comments on this version? :)

Hi Wen Gu,

Quoted 9 lines

as background for the others: IBM & Alibaba are having calls regarding SMC since Alibaba shows a lot of interest and there is a lot to discuss (as you may have noticed).

As we have agreed in our calls, the information shared about SMC-D in specific doesn't make it easy to understand the underlying concepts (e.g. role of GID, EIDs, DMB tokens, etc.). That's why one of the next calls is dedicated to explain some of the concepts in depth to you and see where we can improve our flow of information to the community in the future. After that we can go over this RFC with a shared understanding of the concepts.

Thank you
- Jan

OK, Jan, Let's follow this plan and thank you for organizing this.

Thanks,
Wen Gu