All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexandra Winter <wintera@linux.ibm.com>
To: David Miller <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Eric Dumazet <edumazet@google.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"D. Wythe" <alibuda@linux.alibaba.com>,
	Dust Li <dust.li@linux.alibaba.com>,
	Sidraya Jayagond <sidraya@linux.ibm.com>,
	Wenjia Zhang <wenjia@linux.ibm.com>,
	Julian Ruess <julianr@linux.ibm.com>
Cc: netdev@vger.kernel.org, linux-s390@vger.kernel.org,
	Heiko Carstens <hca@linux.ibm.com>,
	Vasily Gorbik <gor@linux.ibm.com>,
	Alexander Gordeev <agordeev@linux.ibm.com>,
	Christian Borntraeger <borntraeger@linux.ibm.com>,
	Sven Schnelle <svens@linux.ibm.com>,
	Thorsten Winkler <twinkler@linux.ibm.com>,
	Simon Horman <horms@kernel.org>,
	Mahanta Jambigi <mjambigi@linux.ibm.com>,
	Tony Lu <tonylu@linux.alibaba.com>,
	Wen Gu <guwen@linux.alibaba.com>,
	Halil Pasic <pasic@linux.ibm.com>,
	linux-rdma@vger.kernel.org
Subject: [RFC net-next 04/17] net/smc: Decouple sf and attached send_buf in smc_loopback
Date: Wed,  6 Aug 2025 17:41:09 +0200	[thread overview]
Message-ID: <20250806154122.3413330-5-wintera@linux.ibm.com> (raw)
In-Reply-To: <20250806154122.3413330-1-wintera@linux.ibm.com>

Before this patch there was the following assumption in
smc_loopback.c>smc_lo_move_data():
sf (signalling flag) == 0 : data is already in an attached target dmb
sf == 1 : data is not yet in the target dmb

This is true for the 2 callers in smc client
smcd_cdc_msg_send() : sf=1
smcd_tx_rdma_writes() : sf=0
but should not be a general assumption.

Add a bool to struct smc_buf_desc to indicate whether an SMC-D sndbuf_desc
is an attached buffer. Don't call move_data() for attached send_buffers,
because it is not necessary.

Move the data in smc_lo_move_data() if len != 0 and signal when requested.

Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Mahanta Jambigi <mjambigi@linux.ibm.com>
---
 net/smc/smc_core.h     | 5 +++++
 net/smc/smc_ism.c      | 1 +
 net/smc/smc_loopback.c | 9 +++------
 net/smc/smc_tx.c       | 3 +++
 4 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h
index 48a1b1dcb576..fe5f48d14323 100644
--- a/net/smc/smc_core.h
+++ b/net/smc/smc_core.h
@@ -13,6 +13,7 @@
 #define _SMC_CORE_H
 
 #include <linux/atomic.h>
+#include <linux/types.h>
 #include <linux/smc.h>
 #include <linux/pci.h>
 #include <rdma/ib_verbs.h>
@@ -221,12 +222,16 @@ struct smc_buf_desc {
 					/* virtually contiguous */
 		};
 		struct { /* SMC-D */
+			 /* SMC-D rx buffer: */
 			unsigned short	sba_idx;
 					/* SBA index number */
 			u64		token;
 					/* DMB token number */
 			dma_addr_t	dma_addr;
 					/* DMA address */
+			/* SMC-D tx buffer */
+			bool		is_attached;
+					/* no need for explicit writes */
 		};
 	};
 };
diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c
index a94e1450d095..7363f8be9f94 100644
--- a/net/smc/smc_ism.c
+++ b/net/smc/smc_ism.c
@@ -274,6 +274,7 @@ int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
 		dmb_desc->cpu_addr = dmb.cpu_addr;
 		dmb_desc->dma_addr = dmb.dma_addr;
 		dmb_desc->len = dmb.dmb_len;
+		dmb_desc->is_attached = true;
 	}
 	return rc;
 }
diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c
index 0eb00bbefd17..1853c26fbbbb 100644
--- a/net/smc/smc_loopback.c
+++ b/net/smc/smc_loopback.c
@@ -224,12 +224,6 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
 	struct smc_lo_dev *ldev = smcd->priv;
 	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(&ldev->dmb_ht_lock);
 	hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) {
 		if (tmp_node->token == dmb_tok) {
@@ -244,6 +238,9 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok,
 	memcpy((char *)rmb_node->cpu_addr + offset, data, size);
 	read_unlock_bh(&ldev->dmb_ht_lock);
 
+	if (!sf)
+		return 0;
+
 	conn = smcd->conn[rmb_node->sba_idx];
 	if (!conn || conn->killed)
 		return -EPIPE;
diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c
index 214ac3cbcf9a..3144b4b1fe29 100644
--- a/net/smc/smc_tx.c
+++ b/net/smc/smc_tx.c
@@ -426,6 +426,9 @@ static int smcd_tx_rdma_writes(struct smc_connection *conn, size_t len,
 	int srcchunk, dstchunk;
 	int rc;
 
+	if (conn->sndbuf_desc->is_attached)
+		return 0;
+
 	for (dstchunk = 0; dstchunk < 2; dstchunk++) {
 		for (srcchunk = 0; srcchunk < 2; srcchunk++) {
 			void *data = conn->sndbuf_desc->cpu_addr + src_off;
-- 
2.48.1


  parent reply	other threads:[~2025-08-06 15:41 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-06 15:41 [RFC net-next 00/17] dibs - Direct Internal Buffer Sharing Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 01/17] net/smc: Remove __init marker from smc_core_init() Alexandra Winter
2025-08-07  3:34   ` Dust Li
2025-08-07  7:01     ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 02/17] s390/ism: Log module load/unload Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 03/17] net/smc: Remove error handling of unregister_dmb() Alexandra Winter
2025-08-10 11:03   ` Dust Li
2025-08-11 11:28     ` Alexandra Winter
2025-08-12 22:53       ` Dust Li
2025-08-06 15:41 ` Alexandra Winter [this message]
2025-08-10 14:00   ` [RFC net-next 04/17] net/smc: Decouple sf and attached send_buf in smc_loopback Dust Li
2025-08-11 11:35     ` Alexandra Winter
2025-08-11 12:03       ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 05/17] net/smc: Improve log message for devices w/o pnetid Alexandra Winter
2025-08-10 14:07   ` Dust Li
2025-08-11 14:10     ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 06/17] net/dibs: Create net/dibs Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 07/17] net/dibs: Register smc as dibs_client Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 08/17] net/dibs: Register ism as dibs device Alexandra Winter
2025-08-07 15:05   ` kernel test robot
2025-08-07 16:37   ` Simon Horman
2025-08-07 18:19     ` Simon Horman
2025-08-08 18:36       ` Alexandra Winter
2025-08-07 16:57   ` kernel test robot
2025-08-10 14:46   ` Dust Li
2025-08-11 14:27     ` Alexandra Winter
2025-08-12 22:52       ` Dust Li
2025-08-06 15:41 ` [RFC net-next 09/17] net/dibs: Define dibs loopback Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 10/17] net/dibs: Define dibs_client_ops and dibs_dev_ops Alexandra Winter
2025-08-07 19:47   ` Simon Horman
2025-08-08 18:38     ` Alexandra Winter
2025-08-10 14:53   ` Dust Li
2025-08-11 15:12     ` Alexandra Winter
2025-08-12 22:58       ` Dust Li
2025-08-06 15:41 ` [RFC net-next 11/17] net/dibs: Move struct device to dibs_dev Alexandra Winter
2025-08-07 16:32   ` kernel test robot
2025-08-14  8:51   ` Alexandra Winter
2025-08-15  1:56     ` Dust Li
2025-08-15 11:59       ` Alexandra Winter
2025-08-15 15:18         ` Dust Li
2025-09-01 12:46           ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 12/17] net/dibs: Create class dibs Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 13/17] net/dibs: Local gid for dibs devices Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 14/17] net/dibs: Move vlan support to dibs_dev_ops Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 15/17] net/dibs: Move query_remote_gid() " Alexandra Winter
2025-08-11  9:34   ` Julian Ruess
2025-08-14 14:49     ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 16/17] net/dibs: Move data path to dibs layer Alexandra Winter
2025-08-07 17:20   ` kernel test robot
2025-08-07 20:34   ` Simon Horman
2025-08-08 18:38     ` Alexandra Winter
2025-08-06 15:41 ` [RFC net-next 17/17] net/dibs: Move event handling " Alexandra Winter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250806154122.3413330-5-wintera@linux.ibm.com \
    --to=wintera@linux.ibm.com \
    --cc=agordeev@linux.ibm.com \
    --cc=alibuda@linux.alibaba.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=borntraeger@linux.ibm.com \
    --cc=davem@davemloft.net \
    --cc=dust.li@linux.alibaba.com \
    --cc=edumazet@google.com \
    --cc=gor@linux.ibm.com \
    --cc=guwen@linux.alibaba.com \
    --cc=hca@linux.ibm.com \
    --cc=horms@kernel.org \
    --cc=julianr@linux.ibm.com \
    --cc=kuba@kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=mjambigi@linux.ibm.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=pasic@linux.ibm.com \
    --cc=sidraya@linux.ibm.com \
    --cc=svens@linux.ibm.com \
    --cc=tonylu@linux.alibaba.com \
    --cc=twinkler@linux.ibm.com \
    --cc=wenjia@linux.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.