linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org,
	Krishnamraju Eraparaju <krishna2@chelsio.com>,
	Sagi Grimberg <sagi@grimberg.me>,
	Max Gurtovoy <mgurtovoy@nvidia.com>,
	Jason Gunthorpe <jgg@nvidia.com>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.10 06/29] IB/isert: add module param to set sg_tablesize for IO cmd
Date: Mon, 22 Feb 2021 13:13:00 +0100	[thread overview]
Message-ID: <20210222121021.163596444@linuxfoundation.org> (raw)
In-Reply-To: <20210222121019.444399883@linuxfoundation.org>

From: Max Gurtovoy <mgurtovoy@nvidia.com>

[ Upstream commit dae7a75f1f19bffb579daf148f8d8addd2726772 ]

Currently, iser target support max IO size of 16MiB by default. For some
adapters, allocating this amount of resources might reduce the total
number of possible connections that can be created. For those adapters,
it's preferred to reduce the max IO size to be able to create more
connections. Since there is no handshake procedure for max IO size in iser
protocol, set the default max IO size to 1MiB and add a module parameter
for enabling the option to control it for suitable adapters.

Fixes: 317000b926b0 ("IB/isert: allocate RW ctxs according to max IO size")
Link: https://lore.kernel.org/r/20201019094628.17202-1-mgurtovoy@nvidia.com
Reported-by: Krishnamraju Eraparaju <krishna2@chelsio.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 27 ++++++++++++++++++++++++-
 drivers/infiniband/ulp/isert/ib_isert.h |  6 ++++++
 2 files changed, 32 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 436e17f1d0e53..bd478947b93a5 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -28,6 +28,18 @@ static int isert_debug_level;
 module_param_named(debug_level, isert_debug_level, int, 0644);
 MODULE_PARM_DESC(debug_level, "Enable debug tracing if > 0 (default:0)");
 
+static int isert_sg_tablesize_set(const char *val,
+				  const struct kernel_param *kp);
+static const struct kernel_param_ops sg_tablesize_ops = {
+	.set = isert_sg_tablesize_set,
+	.get = param_get_int,
+};
+
+static int isert_sg_tablesize = ISCSI_ISER_DEF_SG_TABLESIZE;
+module_param_cb(sg_tablesize, &sg_tablesize_ops, &isert_sg_tablesize, 0644);
+MODULE_PARM_DESC(sg_tablesize,
+		 "Number of gather/scatter entries in a single scsi command, should >= 128 (default: 256, max: 4096)");
+
 static DEFINE_MUTEX(device_list_mutex);
 static LIST_HEAD(device_list);
 static struct workqueue_struct *isert_comp_wq;
@@ -47,6 +59,19 @@ static void isert_send_done(struct ib_cq *cq, struct ib_wc *wc);
 static void isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc);
 static void isert_login_send_done(struct ib_cq *cq, struct ib_wc *wc);
 
+static int isert_sg_tablesize_set(const char *val, const struct kernel_param *kp)
+{
+	int n = 0, ret;
+
+	ret = kstrtoint(val, 10, &n);
+	if (ret != 0 || n < ISCSI_ISER_MIN_SG_TABLESIZE ||
+	    n > ISCSI_ISER_MAX_SG_TABLESIZE)
+		return -EINVAL;
+
+	return param_set_int(val, kp);
+}
+
+
 static inline bool
 isert_prot_cmd(struct isert_conn *conn, struct se_cmd *cmd)
 {
@@ -101,7 +126,7 @@ isert_create_qp(struct isert_conn *isert_conn,
 	attr.cap.max_send_wr = ISERT_QP_MAX_REQ_DTOS + 1;
 	attr.cap.max_recv_wr = ISERT_QP_MAX_RECV_DTOS + 1;
 	factor = rdma_rw_mr_factor(device->ib_device, cma_id->port_num,
-				   ISCSI_ISER_MAX_SG_TABLESIZE);
+				   isert_sg_tablesize);
 	attr.cap.max_rdma_ctxs = ISCSI_DEF_XMIT_CMDS_MAX * factor;
 	attr.cap.max_send_sge = device->ib_device->attrs.max_send_sge;
 	attr.cap.max_recv_sge = 1;
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index 7fee4a65e181a..6c5af13db4e0d 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -65,6 +65,12 @@
  */
 #define ISER_RX_SIZE		(ISCSI_DEF_MAX_RECV_SEG_LEN + 1024)
 
+/* Default I/O size is 1MB */
+#define ISCSI_ISER_DEF_SG_TABLESIZE 256
+
+/* Minimum I/O size is 512KB */
+#define ISCSI_ISER_MIN_SG_TABLESIZE 128
+
 /* Maximum support is 16MB I/O size */
 #define ISCSI_ISER_MAX_SG_TABLESIZE	4096
 
-- 
2.27.0




  parent reply	other threads:[~2021-02-22 12:28 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-22 12:12 [PATCH 5.10 00/29] 5.10.18-rc1 review Greg Kroah-Hartman
2021-02-22 12:12 ` [PATCH 5.10 01/29] vdpa_sim: remove hard-coded virtq count Greg Kroah-Hartman
2021-02-22 12:12 ` [PATCH 5.10 02/29] vdpa_sim: add struct vdpasim_dev_attr for device attributes Greg Kroah-Hartman
2021-02-22 12:12 ` [PATCH 5.10 03/29] vdpa_sim: store parsed MAC address in a buffer Greg Kroah-Hartman
2021-02-22 19:54   ` Pavel Machek
2021-02-23  4:49     ` Greg Kroah-Hartman
2021-02-23  8:06     ` Stefano Garzarella
2021-02-24  8:29       ` Pavel Machek
2021-02-24  8:36         ` Stefano Garzarella
2021-02-22 12:12 ` [PATCH 5.10 04/29] vdpa_sim: make config generic and usable for any device type Greg Kroah-Hartman
2021-02-22 12:12 ` [PATCH 5.10 05/29] vdpa_sim: add get_config callback in vdpasim_dev_attr Greg Kroah-Hartman
2021-02-22 12:13 ` Greg Kroah-Hartman [this message]
2021-02-22 12:13 ` [PATCH 5.10 07/29] net: qrtr: Fix port ID for control messages Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 08/29] mptcp: skip to next candidate if subflow has unacked data Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 09/29] net/sched: fix miss init the mru in qdisc_skb_cb Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 10/29] mt76: mt7915: fix endian issues Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 11/29] mt76: mt7615: fix rdd mcu cmd endianness Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 12/29] net: sched: incorrect Kconfig dependencies on Netfilter modules Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 13/29] net: openvswitch: fix TTL decrement exception action execution Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 14/29] net: bridge: Fix a warning when del bridge sysfs Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 15/29] net: fix proc_fs init handling in af_packet and tls Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 16/29] Xen/x86: dont bail early from clear_foreign_p2m_mapping() Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 17/29] Xen/x86: also check kernel mapping in set_foreign_p2m_mapping() Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 18/29] Xen/gntdev: correct dev_bus_addr handling in gntdev_map_grant_pages() Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 19/29] Xen/gntdev: correct error checking " Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 20/29] xen/arm: dont ignore return errors from set_phys_to_machine Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 21/29] xen-blkback: dont "handle" error by BUG() Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 22/29] xen-netback: " Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 23/29] xen-scsiback: " Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 24/29] xen-blkback: fix error handling in xen_blkbk_map() Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 25/29] tty: protect tty_write from odd low-level tty disciplines Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 26/29] Bluetooth: btusb: Always fallback to alt 1 for WBS Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 27/29] btrfs: fix backport of 2175bf57dc952 in 5.10.13 Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 28/29] btrfs: fix crash after non-aligned direct IO write with O_DSYNC Greg Kroah-Hartman
2021-02-22 12:13 ` [PATCH 5.10 29/29] media: pwc: Use correct device for DMA Greg Kroah-Hartman
2021-02-22 17:17 ` [PATCH 5.10 00/29] 5.10.18-rc1 review Florian Fainelli
2021-02-24 18:42   ` Greg Kroah-Hartman
2021-02-22 18:42 ` Pavel Machek
2021-02-24 18:42   ` Greg Kroah-Hartman
2021-02-22 21:28 ` Guenter Roeck
2021-02-22 21:34 ` Igor
2021-02-23  2:49 ` Naresh Kamboju
2021-02-23 21:06 ` Shuah Khan

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=20210222121021.163596444@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=jgg@nvidia.com \
    --cc=krishna2@chelsio.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgurtovoy@nvidia.com \
    --cc=sagi@grimberg.me \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).