Skip to contentSkip to conversation

net/smc: SMC intra-OS shortcut with loopback-ism

PATCH Superseded Wen Gu 2 participants 17 messages Mar 7, 2024 – Mar 12, 2024

Subsystems (5)

v2 of 7 Superseded Latest is v7 →

lore
Share & tools
lore

Wen Gu

author reviewer
#0 lore raw

v2 cover net/smc: SMC intra-OS shortcut with loopback-ism

This patch set acts as the second part of the new version of [1] (The first part can be referred from [2]), the updated things of this version are listed at the end.

- Background

SMC-D is now used in IBM z with ISM function to optimize network interconnect for intra-CPC communications. Inspired by this, we try to make SMC-D available on the non-s390 architecture through a software-implemented Emulated-ISM device, that is the loopback-ism device here, to accelerate inter-process or inter-containers communication within the same OS instance.

- Design

This patch set includes 3 parts:

- Patch #1-#2: some prepare work for loopback-ism.
- Patch #3-#7: implement loopback-ism device. Noted that loopback-ism now serves only SMC and no userspace interface exposed.
- Patch #10-#15: memory copy optimization for intra-OS scenario.

The loopback-ism device is designed as an ISMv2 device and not be limited to a specific net namespace, ends of both inter-process connection (1/1' in diagram below) or inter-container connection (2/2' in diagram below) can find the same available loopback-ism and choose it during the CLC handshake.

 Container 1 (ns1)                              Container 2 (ns2)
 +-----------------------------------------+    +-------------------------+
 | +-------+      +-------+      +-------+ |    |        +-------+        |
 | | App A |      | App B |      | App C | |    |        | App D |<-+     |
 | +-------+      +---^---+      +-------+ |    |        +-------+  |(2') |
 |     |127.0.0.1 (1')|             |192.168.0.11       192.168.0.12|     |
 |  (1)|   +--------+ | +--------+  |(2)   |    | +--------+   +--------+ |
 |     `-->|   lo   |-` |  eth0  |<-`      |    | |   lo   |   |  eth0  | |
 +---------+--|---^-+---+-----|--+---------+    +-+--------+---+-^------+-+
              |   |           |                                  |
 Kernel       |   |           |                                  |
 +----+-------v---+-----------v----------------------------------+---+----+
 |    |                            TCP                               |    |
 |    |                                                              |    |
 |    +--------------------------------------------------------------+    |
 |                                                                        |
 |                           +--------------+                             |
 |                           | smc loopback |                             |
 +---------------------------+--------------+-----------------------------+

loopback-ism device creates DMBs (shared memory) for each connection peer. Since data transfer occurs within the same kernel, the sndbuf of each peer is only a descriptor and point to the same memory region as peer DMB, so that the data copy from sndbuf to peer DMB can be avoided in loopback-ism case.

 Container 1 (ns1)                              Container 2 (ns2)
 +-----------------------------------------+    +-------------------------+
 | +-------+                               |    |        +-------+        |
 | | App C |-----+                         |    |        | App D |        |
 | +-------+     |                         |    |        +-^-----+        |
 |               |                         |    |          |              |
 |           (2) |                         |    |     (2') |              |
 |               |                         |    |          |              |
 +---------------|-------------------------+    +----------|--------------+
                 |                                         |
 Kernel          |                                         |
 +---------------|-----------------------------------------|--------------+
 | +--------+ +--v-----+                           +--------+ +--------+  |
 | |dmb_desc| |snd_desc|                           |dmb_desc| |snd_desc|  |
 | +-----|--+ +--|-----+                           +-----|--+ +--------+  |
 | +-----|--+    |                                 +-----|--+             |
 | | DMB C  |    +---------------------------------| DMB D  |             |
 | +--------+                                      +--------+             |
 |                                                                        |
 |                           +--------------+                             |
 |                           | smc loopback |                             |
 +---------------------------+--------------+-----------------------------+

- Benchmark Test

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

* Test object:
- TCP: run on TCP loopback.
- SMC lo: run on SMC loopback-ism.

1. ipc-benchmark (see [3])

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

TCP SMC-lo
Message
rate (msg/s) 79287 148946(+87.86%)

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 SMC-lo
Bandwidth(MBps) 5053.47 8195.07(+62.17%)
Latency(us) 6.108 3.404(-44.27%)

3. nginx/wrk

- serv: <smc_run> nginx
- clnt: <smc_run> wrk -t 8 -c 1000 -d 30 http://127.0.0.1:80

TCP SMC-lo
Requests/s 178187.56 247970.92(+39.16%)

4. redis-benchmark

- serv: <smc_run> redis-server
- clnt: <smc_run> redis-benchmark -h 127.0.0.1 -q -t set,get -n 400000 -c 200 -d 1024

TCP SMC-lo

GET(Requests/s)       89067.02                123877.36(+39.08%)
SET(Requests/s)       87700.07                131319.77(+49.73%)

Change log:

v2->v1:
- All the patches: changed the term virtual-ISM to Emulated-ISM as defined by SMCv2.1.
- Patch #3: optimized the description of SMC_LO config. Avoid exposing loopback-ism to sysfs and remove all the knobs until future definition clear.
- Patch #3: try to make lockdep happy by using read_lock_bh() in smc_lo_move_data().
- Patch #6: defaultly use physical contiguous DMB buffers.
- Patch #11: defaultly enable DMB no-copy for loopback-ism and free the DMB in unregister_dmb or detach_dmb when dmb_node->refcnt reaches 0, instead of using wait_event to keep waiting in unregister_dmb.

v1->RFC:
- Patch #9: merge rx_bytes and tx_bytes as xfer_bytes statistics:
/sys/devices/virtual/smc/loopback-ism/xfer_bytes
- Patch #10: add support_dmb_nocopy operation to check if SMC-D device supports merging sndbuf with peer DMB.
- Patch #13 & #14: introduce loopback-ism device control of DMB memory type and control of whether to merge sndbuf and DMB. They can be respectively set by:
/sys/devices/virtual/smc/loopback-ism/dmb_type /sys/devices/virtual/smc/loopback-ism/dmb_copy
The motivation for these two control is that a performance bottleneck was found when using vzalloced DMB and sndbuf is merged with DMB, and there are many CPUs and CONFIG_HARDENED_USERCOPY is set [4]. The bottleneck is caused by the lock contention in vmap_area_lock [5] which is involved in memcpy_from_msg() or memcpy_to_msg(). Currently, Uladzislau Rezki is working on mitigating the vmap lock contention [6]. It has significant effects, but using virtual memory still has additional overhead compared to using physical memory.
So this new version provides controls of dmb_type and dmb_copy to suit different scenarios.
- Some minor changes and comments improvements.

RFC->old version([1]):

- Patch #1: improve the loopback-ism dump, it shows as follows now:

  # smcd d
  FID  Type  PCI-ID        PCHID  InUse  #LGs  PNET-ID
  0000 0     loopback-ism  ffff   No        0
- Patch #3: introduce the smc_ism_set_v2_capable() helper and set
  smc_ism_v2_capable when ISMv2 or virtual ISM is registered,
  regardless of whether there is already a device in smcd device list.
- Patch #3: loopback-ism will be added into /sys/devices/virtual/smc/loopback-ism/.
- Patch #8: introduce the runtime switch /sys/devices/virtual/smc/loopback-ism/active
  to activate or deactivate the loopback-ism.
- Patch #9: introduce the statistics of loopback-ism by
  /sys/devices/virtual/smc/loopback-ism/{{tx|rx}_tytes|dmbs_cnt}.
- Some minor changes and comments improvements.

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

Wen Gu (11):
net/smc: adapt SMC-D device dump for Emulated-ISM net/smc: decouple ism_client from SMC-D DMB registration net/smc: introduce loopback-ism for SMC intra-OS shortcut net/smc: implement ID-related operations of loopback-ism net/smc: implement some unsupported operations of loopback-ism net/smc: implement DMB-related operations of loopback-ism net/smc: register loopback-ism into SMC-D device list net/smc: add operations to merge sndbuf with peer DMB net/smc: attach or detach ghost sndbuf to peer DMB net/smc: adapt cursor update when sndbuf and peer DMB are merged net/smc: implement DMB-merged operations of loopback-ism

 drivers/s390/net/ism_drv.c |   2 +-
 include/net/smc.h          |   7 +-
 net/smc/Kconfig            |  13 +
 net/smc/Makefile           |   2 +-
 net/smc/af_smc.c           |  28 ++-
 net/smc/smc_cdc.c          |  58 ++++-
 net/smc/smc_cdc.h          |   1 +
 net/smc/smc_core.c         |  61 ++++-
 net/smc/smc_core.h         |   1 +
 net/smc/smc_ism.c          |  71 +++++-
 net/smc/smc_ism.h          |   5 +
 net/smc/smc_loopback.c     | 469 +++++++++++++++++++++++++++++++++++++
 net/smc/smc_loopback.h     |  52 ++++
 13 files changed, 741 insertions(+), 29 deletions(-)
 create mode 100644 net/smc/smc_loopback.c
 create mode 100644 net/smc/smc_loopback.h
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#1 lore raw

PATCH v2 1/11 net/smc: adapt SMC-D device dump for Emulated-ISM

The introduction of Emulated-ISM requires adaptation of SMC-D device dump. Software implemented non-PCI device (loopback-ism) should be handled correctly and the CHID reserved for Emulated-ISM should be got from smcd_ops interface instead of PCI information.

---
 net/smc/smc_ism.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
net/smc/smc_ism.c +10 -3
index ac88de2a06a0..b6eca4231913 100644
@@ -252,12 +252,11 @@ static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
 	char smc_pnet[SMC_MAX_PNETID_LEN + 1];
 	struct smc_pci_dev smc_pci_dev;
 	struct nlattr *port_attrs;
+	struct device *device;
 	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);
@@ -272,7 +271,15 @@ 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);
+	device = smcd->ops->get_dev(smcd);
+	if (device->parent)
+		smc_set_pci_values(to_pci_dev(device->parent), &smc_pci_dev);
+	if (smc_ism_is_emulated(smcd)) {
+		smc_pci_dev.pci_pchid = smc_ism_get_chid(smcd);
+		if (!device->parent)
+			snprintf(smc_pci_dev.pci_id, sizeof(smc_pci_dev.pci_id),
+				 "%s", dev_name(device));
+	}
 	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))
Signature · 1 line
2.32.0.3.g01195cf9f
#2 lore raw

Re: [PATCH net-next v2 01/11] net/smc: adapt SMC-D device dump for Emulated-ISM

Quoted 29 lines · Wen Gu

The introduction of Emulated-ISM requires adaptation of SMC-D device dump. Software implemented non-PCI device (loopback-ism) should be handled correctly and the CHID reserved for Emulated-ISM should be got from smcd_ops interface instead of PCI information.

---
  net/smc/smc_ism.c | 13 ++++++++++---
  1 file changed, 10 insertions(+), 3 deletions(-)
net/smc/smc_ism.c +1 -2
index ac88de2a06a0..b6eca4231913 100644
@@ -252,12 +252,11 @@ static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
  	char smc_pnet[SMC_MAX_PNETID_LEN + 1];
  	struct smc_pci_dev smc_pci_dev;
  	struct nlattr *port_attrs;
+	struct device *device;
  	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);

net/smc/smc_ism.c static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd, · line 277

Quoted 8 lines
net/smc/smc_ism.c +4 -1
@@ -272,7 +271,15 @@ 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);
+	device = smcd->ops->get_dev(smcd);
+	if (device->parent)
+		smc_set_pci_values(to_pci_dev(device->parent), &smc_pci_dev);
+	if (smc_ism_is_emulated(smcd)) {

net/smc/smc_ism.c: In function ‘smc_nl_handle_smcd_dev’:
net/smc/smc_ism.c:318:13: error: implicit declaration of function ‘smc_ism_is_emulated’ [-Werror=implicit-function-declaration]

  318 |         if (smc_ism_is_emulated(smcd)) {
      |             ^~~~~~~~~~~~~~~~~~~
Quoted 8 lines
net/smc/smc_ism.c +5 -0
+		smc_pci_dev.pci_pchid = smc_ism_get_chid(smcd);
+		if (!device->parent)
+			snprintf(smc_pci_dev.pci_id, sizeof(smc_pci_dev.pci_id),
+				 "%s", dev_name(device));
+	}
  	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))
author reviewer
#3 lore raw

Re: [PATCH net-next v2 01/11] net/smc: adapt SMC-D device dump for Emulated-ISM

Quoted 36 lines · Jan Karcher
Quoted 33 lines · Wen Gu

The introduction of Emulated-ISM requires adaptation of SMC-D device dump. Software implemented non-PCI device (loopback-ism) should be handled correctly and the CHID reserved for Emulated-ISM should be got from smcd_ops interface instead of PCI information.

---
  net/smc/smc_ism.c | 13 ++++++++++---

  1 file changed, 10 insertions(+), 3 deletions(-)
net/smc/smc_ism.c +2 -3
index ac88de2a06a0..b6eca4231913 100644
@@ -252,12 +252,11 @@ static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
       char smc_pnet[SMC_MAX_PNETID_LEN + 1];
       struct smc_pci_dev smc_pci_dev;
       struct nlattr *port_attrs;
+    struct device *device;
       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);
@@ -272,7 +271,15 @@ 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);
+    device = smcd->ops->get_dev(smcd);
Quoted 9 lines

net/smc/smc_ism.c static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd, · line 277

Quoted 3 lines
net/smc/smc_ism.c +3 -0
+    if (device->parent)
+        smc_set_pci_values(to_pci_dev(device->parent), &smc_pci_dev);
+    if (smc_ism_is_emulated(smcd)) {

net/smc/smc_ism.c: In function ‘smc_nl_handle_smcd_dev’:
net/smc/smc_ism.c:318:13: error: implicit declaration of function ‘smc_ism_is_emulated’ [-Werror=implicit-function-declaration]
  318 |         if (smc_ism_is_emulated(smcd)) {
      |             ^~~~~~~~~~~~~~~~~~~

Hi Jan, the function smc_ism_is_emulated() is introduced by this merged patch:

b27696cd8fcc ("net/smc: change the term virtual ISM to Emulated-ISM")

Could you please check if your code base has this one?

Thanks!

Quoted 8 lines
Quoted 8 lines
net/smc/smc_ism.c +5 -0
+        smc_pci_dev.pci_pchid = smc_ism_get_chid(smcd);
+        if (!device->parent)
+            snprintf(smc_pci_dev.pci_id, sizeof(smc_pci_dev.pci_id),
+                 "%s", dev_name(device));
+    }
       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))
author reviewer
#4 lore raw

PATCH v2 2/11 net/smc: decouple ism_client from SMC-D DMB registration

In reply to Wen Gu

The struct 'ism_client' is specialized for s390 platform firmware ISM. So replace it with 'void' to make SMCD DMB registration helper generic for both Emulated-ISM and existing ISM.

---
 drivers/s390/net/ism_drv.c | 2 +-
 include/net/smc.h          | 4 ++--
 net/smc/smc_ism.c          | 7 ++-----
 3 files changed, 5 insertions(+), 8 deletions(-)
drivers/s390/net/ism_drv.c +1 -1
index 2c8e964425dc..9b2a52913e76 100644
@@ -726,7 +726,7 @@ static int smcd_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid,
 }
 
 static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
-			     struct ism_client *client)
+			     void *client)
 {
 	return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb, client);
 }
include/net/smc.h +2 -2
index c9dcb30e3fd9..6273c3a8b24a 100644
@@ -50,7 +50,6 @@ struct smcd_dmb {
 #define ISM_ERROR	0xFFFF
 
 struct smcd_dev;
-struct ism_client;
 
 struct smcd_gid {
 	u64	gid;
@@ -61,7 +60,7 @@ struct smcd_ops {
 	int (*query_remote_gid)(struct smcd_dev *dev, struct smcd_gid *rgid,
 				u32 vid_valid, u32 vid);
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
-			    struct ism_client *client);
+			    void *client);
 	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);
@@ -81,6 +80,7 @@ struct smcd_ops {
 struct smcd_dev {
 	const struct smcd_ops *ops;
 	void *priv;
+	void *client;
 	struct list_head list;
 	spinlock_t lock;
 	struct smc_connection **conn;
net/smc/smc_ism.c +2 -5
index b6eca4231913..26743a14cf27 100644
@@ -222,7 +222,6 @@ 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;
 
@@ -231,7 +230,7 @@ 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.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);
 	if (!rc) {
 		dmb_desc->sba_idx = dmb.sba_idx;
 		dmb_desc->token = dmb.dmb_tok;
@@ -240,9 +239,6 @@ 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,
@@ -453,6 +449,7 @@ static void smcd_register_dev(struct ism_dev *ism)
 	if (!smcd)
 		return;
 	smcd->priv = ism;
+	smcd->client = &smc_ism_client;
 	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
2.32.0.3.g01195cf9f
author reviewer
#5 lore raw

PATCH v2 3/11 net/smc: introduce loopback-ism for SMC intra-OS shortcut

In reply to Wen Gu

This introduces a kind of Emulated-ISM device named loopback-ism for SMCv2.1. The loopback-ism device is currently exclusive for SMC usage, and aims to provide a SMC shortcut for sockets within the same kernel, leading to improved intra-OS traffic performance. Configuration of this feature is managed through the config SMC_LO.

---
 net/smc/Kconfig        |  13 ++++
 net/smc/Makefile       |   2 +-
 net/smc/af_smc.c       |  12 ++-
 net/smc/smc_loopback.c | 164 +++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_loopback.h |  33 +++++++++
 5 files changed, 222 insertions(+), 2 deletions(-)
 create mode 100644 net/smc/smc_loopback.c
 create mode 100644 net/smc/smc_loopback.h
net/smc/Kconfig +13 -0
index 746be3996768..2f1a8706a372 100644
@@ -20,3 +20,16 @@ config SMC_DIAG
 	  smcss.
 
 	  if unsure, say Y.
+
+config SMC_LO
+	bool "SMC intra-OS shortcut with loopback-ism"
+	depends on SMC
+	default n
+	help
+	  SMC_LO enables the creation of an Emulated-ISM device named
+	  loopback-ism in SMC and the use of it for transferring data
+	  when communication occurs within the same OS. This helps in
+	  convenient testing of SMC-D since loopback-ism is independent
+	  of architecture or hardware.
+
+	  if unsure, say N.
net/smc/Makefile +1 -1
index 875efcd126a2..a8c37111abe1 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 4b52b3b159c0..751a8fdd6eb0 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
@@ -3557,15 +3558,23 @@ 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:
@@ -3603,6 +3612,7 @@ 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_loopback.c new file +164 -0

164 lines not on this page · raw message

net/smc/smc_loopback.h new file +33 -0
new file mode 100644
index 000000000000..9dd44d4c0ca3
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications Direct over loopback-ism device.
+ *
+ *  Provide a SMC-D loopback-ism device.
+ *
+ *  Copyright (c) 2024, Alibaba Inc.
+ *
+ *  Author: Wen Gu <[email protected]>
+ *          Tony Lu <[email protected]>
+ *
+ */
+
+#ifndef _SMC_LOOPBACK_H
+#define _SMC_LOOPBACK_H
+
+#include <linux/device.h>
+#include <linux/err.h>
+#include <net/smc.h>
+
+#if IS_ENABLED(CONFIG_SMC_LO)
+#define SMC_LO_MAX_DMBS		5000
+
+struct smc_lo_dev {
+	struct smcd_dev *smcd;
+	struct device dev;
+};
+#endif
+
+int smc_loopback_init(void);
+void smc_loopback_exit(void);
+
+#endif /* _SMC_LOOPBACK_H */
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#6 lore raw

PATCH v2 4/11 net/smc: implement ID-related operations of loopback-ism

In reply to Wen Gu

This implements GID and CHID related operations of loopback-ism device. loopback-ism acts as an ISMv2. It's GID is generated randomly by UUIDv4 algorithm and CHID is reserved 0xFFFF.

---
 net/smc/smc_loopback.c | 62 ++++++++++++++++++++++++++++++++++++++----
 net/smc/smc_loopback.h |  3 ++
 2 files changed, 60 insertions(+), 5 deletions(-)
net/smc/smc_loopback.c +57 -5
index e9170d86e58f..7656a2474500 100644
@@ -19,11 +19,62 @@
 #include "smc_loopback.h"
 
 #if IS_ENABLED(CONFIG_SMC_LO)
+#define SMC_LO_V2_CAPABLE	0x1 /* loopback-ism acts as ISMv2 */
+
 static const char smc_lo_dev_name[] = "loopback-ism";
 static struct smc_lo_dev *lo_dev;
 
+static void smc_lo_generate_id(struct smc_lo_dev *ldev)
+{
+	struct smcd_gid *lgid = &ldev->local_gid;
+	uuid_t uuid;
+
+	uuid_gen(&uuid);
+	memcpy(&lgid->gid, &uuid, sizeof(lgid->gid));
+	memcpy(&lgid->gid_ext, (u8 *)&uuid + sizeof(lgid->gid),
+	       sizeof(lgid->gid_ext));
+
+	ldev->chid = SMC_LO_CHID;
+}
+
+static int smc_lo_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid,
+			     u32 vid_valid, u32 vid)
+{
+	struct smc_lo_dev *ldev = smcd->priv;
+
+	/* rgid should be the same as lgid */
+	if (!ldev || rgid->gid != ldev->local_gid.gid ||
+	    rgid->gid_ext != ldev->local_gid.gid_ext)
+		return -ENETUNREACH;
+	return 0;
+}
+
+static int smc_lo_supports_v2(void)
+{
+	return SMC_LO_V2_CAPABLE;
+}
+
+static void smc_lo_get_local_gid(struct smcd_dev *smcd,
+				 struct smcd_gid *smcd_gid)
+{
+	struct smc_lo_dev *ldev = smcd->priv;
+
+	smcd_gid->gid = ldev->local_gid.gid;
+	smcd_gid->gid_ext = ldev->local_gid.gid_ext;
+}
+
+static u16 smc_lo_get_chid(struct smcd_dev *smcd)
+{
+	return ((struct smc_lo_dev *)smcd->priv)->chid;
+}
+
+static struct device *smc_lo_get_dev(struct smcd_dev *smcd)
+{
+	return &((struct smc_lo_dev *)smcd->priv)->dev;
+}
+
 static const struct smcd_ops lo_ops = {
-	.query_remote_gid	= NULL,
+	.query_remote_gid = smc_lo_query_rgid,
 	.register_dmb		= NULL,
 	.unregister_dmb		= NULL,
 	.add_vlan_id		= NULL,
@@ -32,10 +83,10 @@ static const struct smcd_ops lo_ops = {
 	.reset_vlan_required	= NULL,
 	.signal_event		= NULL,
 	.move_data		= NULL,
-	.supports_v2		= NULL,
-	.get_local_gid		= NULL,
-	.get_chid		= NULL,
-	.get_dev		= NULL,
+	.supports_v2 = smc_lo_supports_v2,
+	.get_local_gid = smc_lo_get_local_gid,
+	.get_chid = smc_lo_get_chid,
+	.get_dev = smc_lo_get_dev,
 };
 
 static struct smcd_dev *smcd_lo_alloc_dev(const struct smcd_ops *ops,
@@ -95,6 +146,7 @@ static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev)
 
 static int smc_lo_dev_init(struct smc_lo_dev *ldev)
 {
+	smc_lo_generate_id(ldev);
 	return smcd_lo_register_dev(ldev);
 }
 
net/smc/smc_loopback.h +3 -0
index 9dd44d4c0ca3..55b41133a97f 100644
@@ -20,10 +20,13 @@
 
 #if IS_ENABLED(CONFIG_SMC_LO)
 #define SMC_LO_MAX_DMBS		5000
+#define SMC_LO_CHID		0xFFFF
 
 struct smc_lo_dev {
 	struct smcd_dev *smcd;
 	struct device dev;
+	u16 chid;
+	struct smcd_gid local_gid;
 };
 #endif
 
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#7 lore raw

PATCH v2 5/11 net/smc: implement some unsupported operations of loopback-ism

In reply to Wen Gu

vlan operations are not supported currently since the need for vlan in loopback-ism situation does not seem to be strong.

signal_event operation is not supported since no event now needs to be processed by loopback-ism device.

---
 net/smc/smc_loopback.c | 36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)
net/smc/smc_loopback.c +31 -5
index 7656a2474500..253128c77208 100644
@@ -49,6 +49,32 @@ static int smc_lo_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid,
 	return 0;
 }
 
+static int smc_lo_add_vlan_id(struct smcd_dev *smcd, u64 vlan_id)
+{
+	return -EOPNOTSUPP;
+}
+
+static int smc_lo_del_vlan_id(struct smcd_dev *smcd, u64 vlan_id)
+{
+	return -EOPNOTSUPP;
+}
+
+static int smc_lo_set_vlan_required(struct smcd_dev *smcd)
+{
+	return -EOPNOTSUPP;
+}
+
+static int smc_lo_reset_vlan_required(struct smcd_dev *smcd)
+{
+	return -EOPNOTSUPP;
+}
+
+static int smc_lo_signal_event(struct smcd_dev *dev, struct smcd_gid *rgid,
+			       u32 trigger_irq, u32 event_code, u64 info)
+{
+	return 0;
+}
+
 static int smc_lo_supports_v2(void)
 {
 	return SMC_LO_V2_CAPABLE;
@@ -77,11 +103,11 @@ static const struct smcd_ops lo_ops = {
 	.query_remote_gid = smc_lo_query_rgid,
 	.register_dmb		= NULL,
 	.unregister_dmb		= NULL,
-	.add_vlan_id		= NULL,
-	.del_vlan_id		= NULL,
-	.set_vlan_required	= NULL,
-	.reset_vlan_required	= NULL,
-	.signal_event		= NULL,
+	.add_vlan_id = smc_lo_add_vlan_id,
+	.del_vlan_id = smc_lo_del_vlan_id,
+	.set_vlan_required = smc_lo_set_vlan_required,
+	.reset_vlan_required = smc_lo_reset_vlan_required,
+	.signal_event = smc_lo_signal_event,
 	.move_data		= NULL,
 	.supports_v2 = smc_lo_supports_v2,
 	.get_local_gid = smc_lo_get_local_gid,
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#8 lore raw

PATCH v2 6/11 net/smc: implement DMB-related operations of loopback-ism

In reply to Wen Gu

This implements DMB (un)registration and data move operations of loopback-ism device.

---
 net/smc/smc_cdc.c      |   6 ++
 net/smc/smc_cdc.h      |   1 +
 net/smc/smc_loopback.c | 131 ++++++++++++++++++++++++++++++++++++++++-
 net/smc/smc_loopback.h |  13 ++++
 4 files changed, 148 insertions(+), 3 deletions(-)
net/smc/smc_cdc.c +6 -0
index 3c06625ceb20..c820ef197610 100644
@@ -410,6 +410,12 @@ 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);
+
+	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 696cc11f2303..11559d4ebf2b 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_loopback.c +128 -3

171 lines not on this page · raw message

net/smc/smc_loopback.h +13 -0
index 55b41133a97f..24ab9d747613 100644
@@ -20,13 +20,26 @@
 
 #if IS_ENABLED(CONFIG_SMC_LO)
 #define SMC_LO_MAX_DMBS		5000
+#define SMC_LO_DMBS_HASH_BITS	12
 #define SMC_LO_CHID		0xFFFF
 
+struct smc_lo_dmb_node {
+	struct hlist_node list;
+	u64 token;
+	u32 len;
+	u32 sba_idx;
+	void *cpu_addr;
+	dma_addr_t dma_addr;
+};
+
 struct smc_lo_dev {
 	struct smcd_dev *smcd;
 	struct device dev;
 	u16 chid;
 	struct smcd_gid local_gid;
+	rwlock_t dmb_ht_lock;
+	DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS);
+	DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS);
 };
 #endif
 
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#9 lore raw

PATCH v2 7/11 net/smc: register loopback-ism into SMC-D device list

In reply to Wen Gu

After loopback-ism device gets ready, add it to the SMC-D device list as an ISMv2 device.

---
 net/smc/smc_ism.c      | 11 +++++++----
 net/smc/smc_ism.h      |  1 +
 net/smc/smc_loopback.c | 20 +++++++++++++-------
 3 files changed, 21 insertions(+), 11 deletions(-)
net/smc/smc_ism.c +7 -4
index 26743a14cf27..c50910300a03 100644
@@ -91,6 +91,11 @@ bool smc_ism_is_v2_capable(void)
 	return smc_ism_v2_capable;
 }
 
+void smc_ism_set_v2_capable(void)
+{
+	smc_ism_v2_capable = true;
+}
+
 /* Set a connection using this DMBE. */
 void smc_ism_set_conn(struct smc_connection *conn)
 {
@@ -454,11 +459,9 @@ static void smcd_register_dev(struct ism_dev *ism)
 	if (smc_pnetid_by_dev_port(&ism->pdev->dev, 0, smcd->pnetid))
 		smc_pnetid_by_table_smcd(smcd);
 
+	if (smcd->ops->supports_v2())
+		smc_ism_set_v2_capable();
 	mutex_lock(&smcd_dev_list.mutex);
-	if (list_empty(&smcd_dev_list.list)) {
-		if (smcd->ops->supports_v2())
-			smc_ism_v2_capable = true;
-	}
 	/* 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 165cd013404b..2ea169c1301c 100644
@@ -52,6 +52,7 @@ 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);
 bool smc_ism_is_v2_capable(void);
+void smc_ism_set_v2_capable(void);
 int smc_ism_init(void);
 void smc_ism_exit(void);
 int smcd_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb);
net/smc/smc_loopback.c +13 -7
index 39aa35bf38f5..6828e0ad3e90 100644
@@ -275,10 +275,12 @@ static int smcd_lo_register_dev(struct smc_lo_dev *ldev)
 		return -ENOMEM;
 	ldev->smcd = smcd;
 	smcd->priv = ldev;
-
-	/* TODO:
-	 * register loopback-ism to smcd_dev list.
-	 */
+	smc_ism_set_v2_capable();
+	mutex_lock(&smcd_dev_list.mutex);
+	list_add(&smcd->list, &smcd_dev_list.list);
+	mutex_unlock(&smcd_dev_list.mutex);
+	pr_warn_ratelimited("smc: adding smcd device %s\n",
+			    smc_lo_dev_name);
 	return 0;
 }
 
@@ -286,9 +288,13 @@ static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev)
 {
 	struct smcd_dev *smcd = ldev->smcd;
 
-	/* TODO:
-	 * unregister loopback-ism from smcd_dev list.
-	 */
+	pr_warn_ratelimited("smc: removing smcd device %s\n",
+			    smc_lo_dev_name);
+	smcd->going_away = 1;
+	smc_smcd_terminate_all(smcd);
+	mutex_lock(&smcd_dev_list.mutex);
+	list_del_init(&smcd->list);
+	mutex_unlock(&smcd_dev_list.mutex);
 	kfree(smcd->conn);
 	kfree(smcd);
 }
Signature · 1 line
2.32.0.3.g01195cf9f
author reviewer
#10 lore raw

PATCH v2 8/11 net/smc: add operations to merge sndbuf with peer DMB

In reply to Wen Gu

In some scenarios using Emulated-ISM device, sndbuf can share the same physical memory region with peer DMB to avoid data copy from one side to the other. In such case the sndbuf is only a descriptor that describes the shared memory and does not actually occupy memory, it's more like a ghost buffer.

      +----------+                     +----------+
      | socket A |                     | socket B |
      +----------+                     +----------+
            |                               |
       +--------+                       +--------+
       | sndbuf |                       |  DMB   |
       |  desc  |                       |  desc  |
       +--------+                       +--------+
            |                               |
            |                          +----v-----+
            +-------------------------->  memory  |
                                       +----------+

So here introduces three new SMC-D device operations to check if this feature is supported by device, and to {attach|detach} ghost sndbuf to peer DMB. For now only loopback-ism supports this.

---
 include/net/smc.h |  3 +++
 net/smc/smc_ism.c | 40 ++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_ism.h |  4 ++++
 3 files changed, 47 insertions(+)
include/net/smc.h +3 -0
index 6273c3a8b24a..01387631d8a6 100644
@@ -62,6 +62,9 @@ struct smcd_ops {
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
 			    void *client);
 	int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
+	int (*support_dmb_nocopy)(struct smcd_dev *dev);
+	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 +40 -0
index c50910300a03..67ab9bce202a 100644
@@ -246,6 +246,46 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 	return rc;
 }
 
+bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd)
+{
+	/* for now only loopback-ism supports
+	 * merging sndbuf with peer DMB to avoid
+	 * data copies between them.
+	 */
+	return (smcd->ops->support_dmb_nocopy &&
+		smcd->ops->support_dmb_nocopy(smcd));
+}
+
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc)
+{
+	struct smcd_dmb dmb;
+	int rc = 0;
+
+	if (!dev->ops->attach_dmb)
+		return -EINVAL;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_tok = token;
+	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 +4 -0
index 2ea169c1301c..83cf99265160 100644
@@ -48,6 +48,10 @@ int smc_ism_put_vlan(struct smcd_dev *dev, unsigned short vlan_id);
 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_support_dmb_nocopy(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
2.32.0.3.g01195cf9f
author reviewer
#11 lore raw

PATCH v2 9/11 net/smc: attach or detach ghost sndbuf to peer DMB

In reply to Wen Gu

The ghost sndbuf descriptor will be created and attached to peer DMB once peer token is obtained and it will be detach and freed when the connection is freed.

---
 net/smc/af_smc.c   | 16 ++++++++++++
 net/smc/smc_core.c | 61 +++++++++++++++++++++++++++++++++++++++++++++-
 net/smc/smc_core.h |  1 +
 3 files changed, 77 insertions(+), 1 deletion(-)
net/smc/af_smc.c +16 -0
index 751a8fdd6eb0..fce7a5b2ce5c 100644
@@ -1438,6 +1438,14 @@ static int smc_connect_ism(struct smc_sock *smc,
 	}
 
 	smc_conn_save_peer_info(smc, aclc);
+
+	if (smc_ism_support_dmb_nocopy(smc->conn.lgr->smcd)) {
+		rc = smcd_buf_attach(smc);
+		if (rc) {
+			rc = SMC_CLC_DECL_MEM;	/* try to fallback */
+			goto connect_abort;
+		}
+	}
 	smc_close_init(smc);
 	smc_rx_init(smc);
 	smc_tx_init(smc);
@@ -2542,6 +2550,14 @@ 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_support_dmb_nocopy(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 +60 -1

96 lines not on this page · raw message

net/smc/smc_core.h +1 -0
index 1f175376037b..d93cf51dbd7c 100644
@@ -557,6 +557,7 @@ void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *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
2.32.0.3.g01195cf9f
author reviewer
#12 lore raw

PATCH v2 10/11 net/smc: adapt cursor update when sndbuf and peer DMB are merged

In reply to Wen Gu

Since ghost sndbuf shares the same physical memory with peer DMB, the cursor update processing needs to be adapted to ensure that the data to be consumed won't be overwritten.

So in this case, the fin_curs and sndbuf_space that were originally updated after sending the CDC message should be modified to not be update until the peer updates cons_curs.

---
 net/smc/smc_cdc.c | 52 +++++++++++++++++++++++++++++++++++++----------
 1 file changed, 41 insertions(+), 11 deletions(-)
net/smc/smc_cdc.c +41 -11
index c820ef197610..e938fe3bcc7c 100644
@@ -18,6 +18,7 @@
 #include "smc_tx.h"
 #include "smc_rx.h"
 #include "smc_close.h"
+#include "smc_ism.h"
 
 /********************************** send *************************************/
 
@@ -255,17 +256,25 @@ 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_support_dmb_nocopy(conn->lgr->smcd)) {
+		/* Ghost sndbuf shares the same memory region with
+		 * peer DMB, so don't update the tx_curs_fin and
+		 * sndbuf_space until peer has consumed the data.
+		 */
+		/* 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);
+		smc_tx_sndbuf_nonfull(smc);
+	}
 	return rc;
 }
 
@@ -323,7 +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);
@@ -339,6 +348,27 @@ 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_support_dmb_nocopy(conn->lgr->smcd)) {
+			/* Ghost sndbuf shares the same memory region with
+			 * peer RMB, so update tx_curs_fin and sndbuf_space
+			 * when peer has consumed the data.
+			 */
+			/* 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
2.32.0.3.g01195cf9f
author reviewer
#13 lore raw

PATCH v2 11/11 net/smc: implement DMB-merged operations of loopback-ism

In reply to Wen Gu

This implements operations related to merging sndbuf with peer DMB in loopback-ism. The DMB won't be freed until no sndbuf is attached to it.

---
 net/smc/smc_loopback.c | 136 +++++++++++++++++++++++++++++++++++------
 net/smc/smc_loopback.h |   3 +
 2 files changed, 119 insertions(+), 20 deletions(-)
net/smc/smc_loopback.c +116 -20
index 6828e0ad3e90..7e772f3772de 100644
@@ -21,6 +21,7 @@
 
 #if IS_ENABLED(CONFIG_SMC_LO)
 #define SMC_LO_V2_CAPABLE	0x1 /* loopback-ism acts as ISMv2 */
+#define SMC_LO_SUPPORT_NOCOPY	0x1
 #define SMC_DMA_ADDR_INVALID	(~(dma_addr_t)0)
 
 static const char smc_lo_dev_name[] = "loopback-ism";
@@ -82,6 +83,7 @@ static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
 		goto err_node;
 	}
 	dmb_node->dma_addr = SMC_DMA_ADDR_INVALID;
+	refcount_set(&dmb_node->refcnt, 1);
 
 again:
 	/* add new dmb into hash table */
@@ -95,6 +97,7 @@ 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;
@@ -111,13 +114,29 @@ static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,
 	return rc;
 }
 
+static void __smc_lo_unregister_dmb(struct smc_lo_dev *ldev,
+				    struct smc_lo_dmb_node *dmb_node)
+{
+	/* 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);
+	kvfree(dmb_node->cpu_addr);
+	kfree(dmb_node);
+
+	if (atomic_dec_and_test(&ldev->dmb_cnt))
+		wake_up(&ldev->ldev_release);
+}
+
 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 */
-	write_lock(&ldev->dmb_ht_lock);
+	/* find dmb from hash table */
+	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) {
 			dmb_node = tmp_node;
@@ -125,16 +144,76 @@ static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb)
 		}
 	}
 	if (!dmb_node) {
-		write_unlock(&ldev->dmb_ht_lock);
+		read_unlock(&ldev->dmb_ht_lock);
 		return -EINVAL;
 	}
-	hash_del(&dmb_node->list);
-	write_unlock(&ldev->dmb_ht_lock);
+	read_unlock(&ldev->dmb_ht_lock);
 
-	clear_bit(dmb_node->sba_idx, ldev->sba_idx_mask);
-	kfree(dmb_node->cpu_addr);
-	kfree(dmb_node);
+	if (refcount_dec_and_test(&dmb_node->refcnt))
+		__smc_lo_unregister_dmb(ldev, dmb_node);
+	return 0;
+}
+
+static int smc_lo_support_dmb_nocopy(struct smcd_dev *smcd)
+{
+	return SMC_LO_SUPPORT_NOCOPY;
+}
+
+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) {
+			dmb_node = tmp_node;
+			break;
+		}
+	}
+	if (!dmb_node) {
+		read_unlock(&ldev->dmb_ht_lock);
+		return -EINVAL;
+	}
+	read_unlock(&ldev->dmb_ht_lock);
+
+	if (!refcount_inc_not_zero(&dmb_node->refcnt))
+		/* the dmb is being unregistered, but has
+		 * not been removed from the hash table.
+		 */
+		return -EINVAL;
+
+	/* 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))
+		__smc_lo_unregister_dmb(ldev, dmb_node);
 	return 0;
 }
 
@@ -170,8 +249,22 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
 {
 	struct smc_lo_dmb_node *rmb_node = NULL, *tmp_node;
 	struct smc_lo_dev *ldev = smcd->priv;
-
-	read_lock(&ldev->dmb_ht_lock);
+	struct smc_connection *conn;
+
+	if (!sf)
+		/* since sndbuf is merged with peer DMB, there is
+		 * no need to copy data from sndbuf to peer DMB.
+		 */
+		return 0;
+
+	/* read_lock_bh() is used here just to make lockdep
+	 * happy, because spin_(un)lock_bh(&conn->send_lock) wraps
+	 * smc_lo_move_data() and if we use read_lock() here, lockdep
+	 * will complain about SOFTIRQ-safe -> SOFTIRQ-unsafe lock
+	 * order detected, but in fact ldev->dmb_ht_lock will never
+	 * be held in bh context.
+	 */
+	read_lock_bh(&ldev->dmb_ht_lock);
 	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) {
 		if (tmp_node->token == dmb_tok) {
 			rmb_node = tmp_node;
@@ -182,19 +275,14 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
 		read_unlock(&ldev->dmb_ht_lock);
 		return -EINVAL;
 	}
-	read_unlock(&ldev->dmb_ht_lock);
+	read_unlock_bh(&ldev->dmb_ht_lock);
 
 	memcpy((char *)rmb_node->cpu_addr + offset, data, size);
 
-	if (sf) {
-		struct smc_connection *conn =
-			smcd->conn[rmb_node->sba_idx];
-
-		if (conn && !conn->killed)
-			smcd_cdc_rx_handler(conn);
-		else
-			return -EPIPE;
-	}
+	conn = smcd->conn[rmb_node->sba_idx];
+	if (!conn || conn->killed)
+		return -EPIPE;
+	smcd_cdc_rx_handler(conn);
 	return 0;
 }
 
@@ -226,6 +314,9 @@ static const struct smcd_ops lo_ops = {
 	.query_remote_gid = smc_lo_query_rgid,
 	.register_dmb = smc_lo_register_dmb,
 	.unregister_dmb = smc_lo_unregister_dmb,
+	.support_dmb_nocopy = smc_lo_support_dmb_nocopy,
+	.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,
@@ -304,12 +395,17 @@ 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->ldev_release);
+
 	return smcd_lo_register_dev(ldev);
 }
 
 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_release(struct device *dev)
net/smc/smc_loopback.h +3 -0
index 24ab9d747613..9156a6c37e65 100644
@@ -30,6 +30,7 @@ struct smc_lo_dmb_node {
 	u32 sba_idx;
 	void *cpu_addr;
 	dma_addr_t dma_addr;
+	refcount_t refcnt;
 };
 
 struct smc_lo_dev {
@@ -37,9 +38,11 @@ struct smc_lo_dev {
 	struct device dev;
 	u16 chid;
 	struct smcd_gid local_gid;
+	atomic_t dmb_cnt;
 	rwlock_t dmb_ht_lock;
 	DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS);
 	DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS);
+	wait_queue_head_t ldev_release;
 };
 #endif
 
Signature · 1 line
2.32.0.3.g01195cf9f
#14 lore raw

Re: [PATCH net-next v2 11/11] net/smc: implement DMB-merged operations of loopback-ism

Quoted 2 lines · Wen Gu

This implements operations related to merging sndbuf with peer DMB in loopback-ism. The DMB won't be freed until no sndbuf is attached to it.

Hi Wen Gu,

while I'm still reviewing let me drop a lockdep finding.

Quoted 11 lines
---
  net/smc/smc_loopback.c | 136 +++++++++++++++++++++++++++++++++++------
  net/smc/smc_loopback.h |   3 +
  2 files changed, 119 insertions(+), 20 deletions(-)
net/smc/smc_loopback.c +0 -0
index 6828e0ad3e90..7e772f3772de 100644

[...]

Quoted 38 lines
net/smc/smc_loopback.c +17 -6
@@ -170,8 +249,22 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
  {
  	struct smc_lo_dmb_node *rmb_node = NULL, *tmp_node;
  	struct smc_lo_dev *ldev = smcd->priv;
-
-	read_lock(&ldev->dmb_ht_lock);
+	struct smc_connection *conn;
+
+	if (!sf)
+		/* since sndbuf is merged with peer DMB, there is
+		 * no need to copy data from sndbuf to peer DMB.
+		 */
+		return 0;
+
+	/* read_lock_bh() is used here just to make lockdep
+	 * happy, because spin_(un)lock_bh(&conn->send_lock) wraps
+	 * smc_lo_move_data() and if we use read_lock() here, lockdep
+	 * will complain about SOFTIRQ-safe -> SOFTIRQ-unsafe lock
+	 * order detected, but in fact ldev->dmb_ht_lock will never
+	 * be held in bh context.
+	 */
+	read_lock_bh(&ldev->dmb_ht_lock);
  	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) {
  		if (tmp_node->token == dmb_tok) {
  			rmb_node = tmp_node;
@@ -182,19 +275,14 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
  		read_unlock(&ldev->dmb_ht_lock);
  		return -EINVAL;
  	}
-	read_unlock(&ldev->dmb_ht_lock);
+	read_unlock_bh(&ldev->dmb_ht_lock);
 
  	memcpy((char *)rmb_node->cpu_addr + offset, data, size);
 
-	if (sf) {
-		struct smc_connection *conn =
-			smcd->conn[rmb_node->sba_idx];

net/smc/smc_loopback.c static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, · line 285

Quoted 10 lines
net/smc/smc_loopback.c +4 -6
-
-		if (conn && !conn->killed)
-			smcd_cdc_rx_handler(conn);
-		else
-			return -EPIPE;
-	}
+	conn = smcd->conn[rmb_node->sba_idx];
+	if (!conn || conn->killed)
+		return -EPIPE;
+	smcd_cdc_rx_handler(conn);
[ 2385.528515] ============================================
[ 2385.528517] WARNING: possible recursive locking detected
[ 2385.528519] 6.8.0-loopback_ism-g30af186e8a18-dirty #12 Not tainted
[ 2385.528521] --------------------------------------------
[ 2385.528522] smcapp/51326 is trying to acquire lock:
[ 2385.528524] 000000018707a128 (&smc->conn.send_lock){+...}-{2:2}, at: smc_tx_sndbuf_nonempty+0xba/0x1c0 [smc]
[ 2385.528552]
               but task is already holding lock:
[ 2385.528554] 0000000187078728 (&smc->conn.send_lock){+...}-{2:2}, at: smc_cdc_get_slot_and_msg_send+0x66/0xa0 [smc]
[ 2385.528568]
               other info that might help us debug this:
[ 2385.528570]  Possible unsafe locking scenario:
[ 2385.528572]        CPU0
[ 2385.528573]        ----
[ 2385.528574]   lock(&smc->conn.send_lock);
[ 2385.528576]   lock(&smc->conn.send_lock);
[ 2385.528579]
                *** DEADLOCK ***
[ 2385.528580]  May be due to missing lock nesting notation
[ 2385.528582] 3 locks held by smcapp/51326:
[ 2385.528584]  #0: 0000000187078378 (sk_lock-AF_SMC){+.+.}-{0:0}, at: smc_recvmsg+0x3c/0x2b0 [smc]
[ 2385.528598]  #1: 0000000187078728 (&smc->conn.send_lock){+...}-{2:2}, at: smc_cdc_get_slot_and_msg_send+0x66/0xa0 [smc]
[ 2385.528613]  #2: 0000000187079ce8 (slock-AF_SMC){+...}-{2:2}, at: smc_cdc_msg_recv+0x56/0xe0 [smc]
[ 2385.528627]
               stack backtrace:
[ 2385.528660] CPU: 3 PID: 51326 Comm: smcapp Not tainted 6.8.0-loopback_ism-g30af186e8a18-dirty #12
[ 2385.528663] Hardware name: IBM 3906 M04 704 (LPAR)
[ 2385.528664] Call Trace:
[ 2385.528666]  [<000000012db60788>] dump_stack_lvl+0x90/0x120
[ 2385.528671]  [<000000012cc6d088>] validate_chain+0x560/0x960
[ 2385.528677]  [<000000012cc6f644>] __lock_acquire+0x654/0xd58
[ 2385.528680]  [<000000012cc70a04>] lock_acquire.part.0+0xec/0x260
[ 2385.528683]  [<000000012cc70c24>] lock_acquire+0xac/0x170
[ 2385.528687]  [<000000012dba4ccc>] _raw_spin_lock_bh+0x5c/0xb0
[ 2385.528690]  [<000003ff80453b32>] smc_tx_sndbuf_nonempty+0xba/0x1c0 [smc]
[ 2385.528702]  [<000003ff8045428a>] smc_tx_pending+0x32/0x60 [smc]
[ 2385.528712]  [<000003ff80451f02>] smc_cdc_msg_recv_action+0x3c2/0x528 [smc]
[ 2385.528723]  [<000003ff804520cc>] smc_cdc_msg_recv+0x64/0xe0 [smc]
[ 2385.528734]  [<000003ff80452a4c>] smcd_cdc_rx_handler+0x64/0x70 [smc]
[ 2385.528745]  [<000003ff80459f7e>] smc_lo_move_data+0xde/0x100 [smc]
[ 2385.528755]  [<000003ff804533e0>] smcd_tx_ism_write+0x68/0x90 [smc]
[ 2385.528766]  [<000003ff804528a4>] smcd_cdc_msg_send+0x74/0x118 [smc]
[ 2385.528776]  [<000003ff804529b8>] smc_cdc_get_slot_and_msg_send+0x70/0xa0 [smc]
[ 2385.528788]  [<000003ff804543ec>] smc_tx_consumer_update+0xe4/0x1b0 [smc]
[ 2385.528798]  [<000003ff8045458e>] smc_rx_update_consumer+0x86/0x170 [smc]
[ 2385.528809]  [<000003ff80455ba8>] smc_rx_recvmsg+0x3b8/0x6e8 [smc]
[ 2385.528820]  [<000003ff804388a4>] smc_recvmsg+0xdc/0x2b0 [smc]
[ 2385.528831]  [<000000012d8a6d58>] sock_recvmsg+0x70/0xb0
[ 2385.528837]  [<000000012d8aa0c8>] __sys_recvfrom+0xa8/0x128
[ 2385.528840]  [<000000012d8ab3ca>] __do_sys_socketcall+0x1ca/0x398
[ 2385.528844]  [<000000012db8d4c4>] __do_syscall+0x244/0x308
[ 2385.528847]  [<000000012dba6140>] system_call+0x70/0x98
[ 2385.528850] INFO: lockdep is turned off.

I did not investigate deeper, yet. Just an early heads up that there might be something broken.

Thank you
- Jan

Quoted 55 lines
net/smc/smc_loopback.c +8 -0
  	return 0;
  }
 
@@ -226,6 +314,9 @@ static const struct smcd_ops lo_ops = {
  	.query_remote_gid = smc_lo_query_rgid,
  	.register_dmb = smc_lo_register_dmb,
  	.unregister_dmb = smc_lo_unregister_dmb,
+	.support_dmb_nocopy = smc_lo_support_dmb_nocopy,
+	.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,
@@ -304,12 +395,17 @@ 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->ldev_release);
+
  	return smcd_lo_register_dev(ldev);
  }
 
  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_release(struct device *dev)
net/smc/smc_loopback.h +3 -0
index 24ab9d747613..9156a6c37e65 100644
@@ -30,6 +30,7 @@ struct smc_lo_dmb_node {
  	u32 sba_idx;
  	void *cpu_addr;
  	dma_addr_t dma_addr;
+	refcount_t refcnt;
  };
 
  struct smc_lo_dev {
@@ -37,9 +38,11 @@ struct smc_lo_dev {
  	struct device dev;
  	u16 chid;
  	struct smcd_gid local_gid;
+	atomic_t dmb_cnt;
  	rwlock_t dmb_ht_lock;
  	DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS);
  	DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS);
+	wait_queue_head_t ldev_release;
  };
  #endif
 
author reviewer
#15 lore raw

Re: [PATCH net-next v2 11/11] net/smc: implement DMB-merged operations of loopback-ism

Quoted 128 lines · Jan Karcher
Quoted 2 lines · Wen Gu

This implements operations related to merging sndbuf with peer DMB in loopback-ism. The DMB won't be freed until no sndbuf is attached to it.

Hi Wen Gu,

while I'm still reviewing let me drop a lockdep finding.

Quoted 11 lines

---
  net/smc/smc_loopback.c | 136 +++++++++++++++++++++++++++++++++++------ net/smc/smc_loopback.h |   3 +

  2 files changed, 119 insertions(+), 20 deletions(-)
net/smc/smc_loopback.c +0 -0
index 6828e0ad3e90..7e772f3772de 100644

[...]

Quoted 35 lines
net/smc/smc_loopback.c +17 -6
@@ -170,8 +249,22 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
   {
       struct smc_lo_dmb_node *rmb_node = NULL, *tmp_node;
       struct smc_lo_dev *ldev = smcd->priv;
-
-    read_lock(&ldev->dmb_ht_lock);
+    struct smc_connection *conn;
+
+    if (!sf)
+        /* since sndbuf is merged with peer DMB, there is
+         * no need to copy data from sndbuf to peer DMB.
+         */
+        return 0;
+
+    /* read_lock_bh() is used here just to make lockdep
+     * happy, because spin_(un)lock_bh(&conn->send_lock) wraps
+     * smc_lo_move_data() and if we use read_lock() here, lockdep
+     * will complain about SOFTIRQ-safe -> SOFTIRQ-unsafe lock
+     * order detected, but in fact ldev->dmb_ht_lock will never
+     * be held in bh context.
+     */
+    read_lock_bh(&ldev->dmb_ht_lock);
       hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) {
           if (tmp_node->token == dmb_tok) {
               rmb_node = tmp_node;
@@ -182,19 +275,14 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
           read_unlock(&ldev->dmb_ht_lock);
           return -EINVAL;
       }
-    read_unlock(&ldev->dmb_ht_lock);
+    read_unlock_bh(&ldev->dmb_ht_lock);
       memcpy((char *)rmb_node->cpu_addr + offset, data, size);
-    if (sf) {
-        struct smc_connection *conn =
-            smcd->conn[rmb_node->sba_idx];

net/smc/smc_loopback.c static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, · line 283

Quoted 10 lines
net/smc/smc_loopback.c +4 -6
-
-        if (conn && !conn->killed)
-            smcd_cdc_rx_handler(conn);
-        else
-            return -EPIPE;
-    }
+    conn = smcd->conn[rmb_node->sba_idx];
+    if (!conn || conn->killed)
+        return -EPIPE;
+    smcd_cdc_rx_handler(conn);
[ 2385.528515] ============================================
[ 2385.528517] WARNING: possible recursive locking detected
[ 2385.528519] 6.8.0-loopback_ism-g30af186e8a18-dirty #12 Not tainted
[ 2385.528521] --------------------------------------------
[ 2385.528522] smcapp/51326 is trying to acquire lock:
[ 2385.528524] 000000018707a128 (&smc->conn.send_lock){+...}-{2:2}, at: smc_tx_sndbuf_nonempty+0xba/0x1c0 [smc]
[ 2385.528552]
                but task is already holding lock:
[ 2385.528554] 0000000187078728 (&smc->conn.send_lock){+...}-{2:2}, at: smc_cdc_get_slot_and_msg_send+0x66/0xa0 [smc]
[ 2385.528568]
                other info that might help us debug this:
[ 2385.528570]  Possible unsafe locking scenario:
[ 2385.528572]        CPU0
[ 2385.528573]        ----
[ 2385.528574]   lock(&smc->conn.send_lock);
[ 2385.528576]   lock(&smc->conn.send_lock);
[ 2385.528579]
                 *** DEADLOCK ***
[ 2385.528580]  May be due to missing lock nesting notation
[ 2385.528582] 3 locks held by smcapp/51326:
[ 2385.528584]  #0: 0000000187078378 (sk_lock-AF_SMC){+.+.}-{0:0}, at: smc_recvmsg+0x3c/0x2b0 [smc]
[ 2385.528598]  #1: 0000000187078728 (&smc->conn.send_lock){+...}-{2:2}, at: smc_cdc_get_slot_and_msg_send+0x66/0xa0 [smc]
[ 2385.528613]  #2: 0000000187079ce8 (slock-AF_SMC){+...}-{2:2}, at: smc_cdc_msg_recv+0x56/0xe0 [smc]
[ 2385.528627]
                stack backtrace:
[ 2385.528660] CPU: 3 PID: 51326 Comm: smcapp Not tainted 6.8.0-loopback_ism-g30af186e8a18-dirty #12
[ 2385.528663] Hardware name: IBM 3906 M04 704 (LPAR)
[ 2385.528664] Call Trace:
[ 2385.528666]  [<000000012db60788>] dump_stack_lvl+0x90/0x120
[ 2385.528671]  [<000000012cc6d088>] validate_chain+0x560/0x960
[ 2385.528677]  [<000000012cc6f644>] __lock_acquire+0x654/0xd58
[ 2385.528680]  [<000000012cc70a04>] lock_acquire.part.0+0xec/0x260
[ 2385.528683]  [<000000012cc70c24>] lock_acquire+0xac/0x170
[ 2385.528687]  [<000000012dba4ccc>] _raw_spin_lock_bh+0x5c/0xb0
[ 2385.528690]  [<000003ff80453b32>] smc_tx_sndbuf_nonempty+0xba/0x1c0 [smc]
[ 2385.528702]  [<000003ff8045428a>] smc_tx_pending+0x32/0x60 [smc]
[ 2385.528712]  [<000003ff80451f02>] smc_cdc_msg_recv_action+0x3c2/0x528 [smc]
[ 2385.528723]  [<000003ff804520cc>] smc_cdc_msg_recv+0x64/0xe0 [smc]
[ 2385.528734]  [<000003ff80452a4c>] smcd_cdc_rx_handler+0x64/0x70 [smc]
[ 2385.528745]  [<000003ff80459f7e>] smc_lo_move_data+0xde/0x100 [smc]
[ 2385.528755]  [<000003ff804533e0>] smcd_tx_ism_write+0x68/0x90 [smc]
[ 2385.528766]  [<000003ff804528a4>] smcd_cdc_msg_send+0x74/0x118 [smc]
[ 2385.528776]  [<000003ff804529b8>] smc_cdc_get_slot_and_msg_send+0x70/0xa0 [smc]
[ 2385.528788]  [<000003ff804543ec>] smc_tx_consumer_update+0xe4/0x1b0 [smc]
[ 2385.528798]  [<000003ff8045458e>] smc_rx_update_consumer+0x86/0x170 [smc]
[ 2385.528809]  [<000003ff80455ba8>] smc_rx_recvmsg+0x3b8/0x6e8 [smc]
[ 2385.528820]  [<000003ff804388a4>] smc_recvmsg+0xdc/0x2b0 [smc]
[ 2385.528831]  [<000000012d8a6d58>] sock_recvmsg+0x70/0xb0
[ 2385.528837]  [<000000012d8aa0c8>] __sys_recvfrom+0xa8/0x128
[ 2385.528840]  [<000000012d8ab3ca>] __do_sys_socketcall+0x1ca/0x398
[ 2385.528844]  [<000000012db8d4c4>] __do_syscall+0x244/0x308
[ 2385.528847]  [<000000012dba6140>] system_call+0x70/0x98
[ 2385.528850] INFO: lockdep is turned off.
Quoted 2 lines

I did not investigate deeper, yet. Just an early heads up that there might be something broken.

Thank you for reminding, Jan. I think it is because that I used smcd_cdc_rx_handler(), which may acquire conn->send_lock, in smc_lo_move_data() where the send_lock has been held. I reproduced this issue and will fix it in v3.

Thanks!

Quoted 54 lines

Thank you
- Jan

Quoted 50 lines
net/smc/smc_loopback.c +8 -0
       return 0;
   }
@@ -226,6 +314,9 @@ static const struct smcd_ops lo_ops = {
       .query_remote_gid = smc_lo_query_rgid,
       .register_dmb = smc_lo_register_dmb,
       .unregister_dmb = smc_lo_unregister_dmb,
+    .support_dmb_nocopy = smc_lo_support_dmb_nocopy,
+    .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,
@@ -304,12 +395,17 @@ 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->ldev_release);
+
       return smcd_lo_register_dev(ldev);
   }
   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_release(struct device *dev)
net/smc/smc_loopback.h +3 -0
index 24ab9d747613..9156a6c37e65 100644
@@ -30,6 +30,7 @@ struct smc_lo_dmb_node {
       u32 sba_idx;
       void *cpu_addr;
       dma_addr_t dma_addr;
+    refcount_t refcnt;
   };
   struct smc_lo_dev {
@@ -37,9 +38,11 @@ struct smc_lo_dev {
       struct device dev;
       u16 chid;
       struct smcd_gid local_gid;
+    atomic_t dmb_cnt;
       rwlock_t dmb_ht_lock;
       DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS);
       DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS);
+    wait_queue_head_t ldev_release;
   };
   #endif
#16 lore raw
In reply to Wen Gu
Quoted 3 lines · Wen Gu

This patch set acts as the second part of the new version of [1] (The first part can be referred from [2]), the updated things of this version are listed at the end.

Hi Wen Gu,

thanks for the updated patch. Please givr me some time to review and test the latest version.

Thanks
- Jan

Quoted 201 lines

- Background

SMC-D is now used in IBM z with ISM function to optimize network interconnect for intra-CPC communications. Inspired by this, we try to make SMC-D available on the non-s390 architecture through a software-implemented Emulated-ISM device, that is the loopback-ism device here, to accelerate inter-process or inter-containers communication within the same OS instance.

- Design

This patch set includes 3 parts:

- Patch #1-#2: some prepare work for loopback-ism.
- Patch #3-#7: implement loopback-ism device. Noted that loopback-ism now serves only SMC and no userspace interface exposed.
- Patch #10-#15: memory copy optimization for intra-OS scenario.

The loopback-ism device is designed as an ISMv2 device and not be limited to a specific net namespace, ends of both inter-process connection (1/1' in diagram below) or inter-container connection (2/2' in diagram below) can find the same available loopback-ism and choose it during the CLC handshake.

  Container 1 (ns1)                              Container 2 (ns2)
  +-----------------------------------------+    +-------------------------+
  | +-------+      +-------+      +-------+ |    |        +-------+        |
  | | App A |      | App B |      | App C | |    |        | App D |<-+     |
  | +-------+      +---^---+      +-------+ |    |        +-------+  |(2') |
  |     |127.0.0.1 (1')|             |192.168.0.11       192.168.0.12|     |
  |  (1)|   +--------+ | +--------+  |(2)   |    | +--------+   +--------+ |
  |     `-->|   lo   |-` |  eth0  |<-`      |    | |   lo   |   |  eth0  | |
  +---------+--|---^-+---+-----|--+---------+    +-+--------+---+-^------+-+
               |   |           |                                  |
  Kernel       |   |           |                                  |
  +----+-------v---+-----------v----------------------------------+---+----+
  |    |                            TCP                               |    |
  |    |                                                              |    |
  |    +--------------------------------------------------------------+    |
  |                                                                        |
  |                           +--------------+                             |
  |                           | smc loopback |                             |
  +---------------------------+--------------+-----------------------------+

loopback-ism device creates DMBs (shared memory) for each connection peer. Since data transfer occurs within the same kernel, the sndbuf of each peer is only a descriptor and point to the same memory region as peer DMB, so that the data copy from sndbuf to peer DMB can be avoided in loopback-ism case.

  Container 1 (ns1)                              Container 2 (ns2)
  +-----------------------------------------+    +-------------------------+
  | +-------+                               |    |        +-------+        |
  | | App C |-----+                         |    |        | App D |        |
  | +-------+     |                         |    |        +-^-----+        |
  |               |                         |    |          |              |
  |           (2) |                         |    |     (2') |              |
  |               |                         |    |          |              |
  +---------------|-------------------------+    +----------|--------------+
                  |                                         |
  Kernel          |                                         |
  +---------------|-----------------------------------------|--------------+
  | +--------+ +--v-----+                           +--------+ +--------+  |
  | |dmb_desc| |snd_desc|                           |dmb_desc| |snd_desc|  |
  | +-----|--+ +--|-----+                           +-----|--+ +--------+  |
  | +-----|--+    |                                 +-----|--+             |
  | | DMB C  |    +---------------------------------| DMB D  |             |
  | +--------+                                      +--------+             |
  |                                                                        |
  |                           +--------------+                             |
  |                           | smc loopback |                             |
  +---------------------------+--------------+-----------------------------+

- Benchmark Test

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

* Test object:
- TCP: run on TCP loopback.
- SMC lo: run on SMC loopback-ism.

1. ipc-benchmark (see [3])

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

TCP SMC-lo
Message
rate (msg/s) 79287 148946(+87.86%)

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 SMC-lo
Bandwidth(MBps) 5053.47 8195.07(+62.17%)
Latency(us) 6.108 3.404(-44.27%)

3. nginx/wrk

- serv: <smc_run> nginx
- clnt: <smc_run> wrk -t 8 -c 1000 -d 30 http://127.0.0.1:80

TCP SMC-lo
Requests/s 178187.56 247970.92(+39.16%)

4. redis-benchmark

- serv: <smc_run> redis-server
- clnt: <smc_run> redis-benchmark -h 127.0.0.1 -q -t set,get -n 400000 -c 200 -d 1024

TCP SMC-lo

GET(Requests/s)       89067.02                123877.36(+39.08%)
SET(Requests/s)       87700.07                131319.77(+49.73%)

Change log:

v2->v1:
- All the patches: changed the term virtual-ISM to Emulated-ISM as defined by SMCv2.1.
- Patch #3: optimized the description of SMC_LO config. Avoid exposing loopback-ism to sysfs and remove all the knobs until future definition clear.
- Patch #3: try to make lockdep happy by using read_lock_bh() in smc_lo_move_data().
- Patch #6: defaultly use physical contiguous DMB buffers.
- Patch #11: defaultly enable DMB no-copy for loopback-ism and free the DMB in unregister_dmb or detach_dmb when dmb_node->refcnt reaches 0, instead of using wait_event to keep waiting in unregister_dmb.

v1->RFC:
- Patch #9: merge rx_bytes and tx_bytes as xfer_bytes statistics:
/sys/devices/virtual/smc/loopback-ism/xfer_bytes
- Patch #10: add support_dmb_nocopy operation to check if SMC-D device supports merging sndbuf with peer DMB.
- Patch #13 & #14: introduce loopback-ism device control of DMB memory type and control of whether to merge sndbuf and DMB. They can be respectively set by:
/sys/devices/virtual/smc/loopback-ism/dmb_type /sys/devices/virtual/smc/loopback-ism/dmb_copy
The motivation for these two control is that a performance bottleneck was found when using vzalloced DMB and sndbuf is merged with DMB, and there are many CPUs and CONFIG_HARDENED_USERCOPY is set [4]. The bottleneck is caused by the lock contention in vmap_area_lock [5] which is involved in memcpy_from_msg() or memcpy_to_msg(). Currently, Uladzislau Rezki is working on mitigating the vmap lock contention [6]. It has significant effects, but using virtual memory still has additional overhead compared to using physical memory.
So this new version provides controls of dmb_type and dmb_copy to suit different scenarios.
- Some minor changes and comments improvements.

RFC->old version([1]):

- Patch #1: improve the loopback-ism dump, it shows as follows now:

   # smcd d
   FID  Type  PCI-ID        PCHID  InUse  #LGs  PNET-ID
   0000 0     loopback-ism  ffff   No        0
- Patch #3: introduce the smc_ism_set_v2_capable() helper and set
   smc_ism_v2_capable when ISMv2 or virtual ISM is registered,
   regardless of whether there is already a device in smcd device list.
- Patch #3: loopback-ism will be added into /sys/devices/virtual/smc/loopback-ism/.
- Patch #8: introduce the runtime switch /sys/devices/virtual/smc/loopback-ism/active
   to activate or deactivate the loopback-ism.
- Patch #9: introduce the statistics of loopback-ism by
   /sys/devices/virtual/smc/loopback-ism/{{tx|rx}_tytes|dmbs_cnt}.
- Some minor changes and comments improvements.

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

Wen Gu (11):
net/smc: adapt SMC-D device dump for Emulated-ISM net/smc: decouple ism_client from SMC-D DMB registration net/smc: introduce loopback-ism for SMC intra-OS shortcut net/smc: implement ID-related operations of loopback-ism net/smc: implement some unsupported operations of loopback-ism net/smc: implement DMB-related operations of loopback-ism net/smc: register loopback-ism into SMC-D device list net/smc: add operations to merge sndbuf with peer DMB net/smc: attach or detach ghost sndbuf to peer DMB net/smc: adapt cursor update when sndbuf and peer DMB are merged net/smc: implement DMB-merged operations of loopback-ism

  drivers/s390/net/ism_drv.c |   2 +-
  include/net/smc.h          |   7 +-
  net/smc/Kconfig            |  13 +
  net/smc/Makefile           |   2 +-
  net/smc/af_smc.c           |  28 ++-
  net/smc/smc_cdc.c          |  58 ++++-
  net/smc/smc_cdc.h          |   1 +
  net/smc/smc_core.c         |  61 ++++-
  net/smc/smc_core.h         |   1 +
  net/smc/smc_ism.c          |  71 +++++-
  net/smc/smc_ism.h          |   5 +
  net/smc/smc_loopback.c     | 469 +++++++++++++++++++++++++++++++++++++
  net/smc/smc_loopback.h     |  52 ++++
  13 files changed, 741 insertions(+), 29 deletions(-)

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