diff mbox series

[v2,1/2] IB/srp: Add parse function for maximum initiator to target IU size

Message ID 20190917032421.13000-1-honli@redhat.com (mailing list archive)
State Superseded
Headers show
Series [v2,1/2] IB/srp: Add parse function for maximum initiator to target IU size | expand

Commit Message

Honggang LI Sept. 17, 2019, 3:24 a.m. UTC 
From: Honggang Li <honli@redhat.com>

According to SRP specifications 'srp-r16a' and 'srp2r06',
IOControllerProfile attributes for SRP target port include
the maximum initiator to target IU size.

SRP connection daemons, such as srp_daemon, can get the value
from subnet manager. The SRP connection daemon can pass this
value to kernel.

This patch add parse function for it.

Upstream commit [1] enables the kernel parameter, 'use_imm_data',
by default. [1] also use (8 * 1024) as the default value for
kernel parameter 'max_imm_data'. With those default values, the
maximum initiator to target IU size will be 8260.

In case the SRPT modules, which include the in-tree 'ib_srpt.ko'
module, do not support SRP-2 'immediate data' feature, the default
maximum initiator to target IU size is significantly samller than
8260. For 'ib_srpt.ko' module, which built from source before
[2], the default maximum initiator to target IU is 2116.

[1] introduces a regression issue for old srp target with default
kernel parameters, as the connection will be reject because of
too large maximum initiator to target IU size.

[1] commit 882981f4a411 ("RDMA/srp: Add support for immediate data")
[2] commit 5dabcd0456d7 ("RDMA/srpt: Add support for immediate data")

Signed-off-by: Honggang Li <honli@redhat.com>
---
 drivers/infiniband/ulp/srp/ib_srp.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

Comments

Bart Van Assche Sept. 17, 2019, 5:44 p.m. UTC | #1 
On 9/16/19 8:24 PM, Honggang LI wrote:
> In case the SRPT modules, which include the in-tree 'ib_srpt.ko'
> module, do not support SRP-2 'immediate data' feature, the default
> maximum initiator to target IU size is significantly samller than
                                                        ^^^^^^^
                                                        smaller?
> 8260. For 'ib_srpt.ko' module, which built from source before
> [2], the default maximum initiator to target IU is 2116.
[ ... ]
> diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
> index b5960351bec0..2eadb038b257 100644
> --- a/drivers/infiniband/ulp/srp/ib_srp.c
> +++ b/drivers/infiniband/ulp/srp/ib_srp.c
> @@ -75,6 +75,7 @@ static bool prefer_fr = true;
>   static bool register_always = true;
>   static bool never_register;
>   static int topspin_workarounds = 1;
> +static uint32_t srp_opt_max_it_iu_size;

Each SCSI host can represent a connection to another SRP target, and the 
max_it_iu_size parameter can differ per target. So I think this variable 
should be moved into struct srp_target_port instead of being global. See 
also srp_max_it_iu_len().

Thanks,

Bart.
Honggang LI Sept. 19, 2019, 3:29 a.m. UTC | #2 
On Tue, Sep 17, 2019 at 10:44:56AM -0700, Bart Van Assche wrote:
> On 9/16/19 8:24 PM, Honggang LI wrote:
> > In case the SRPT modules, which include the in-tree 'ib_srpt.ko'
> > module, do not support SRP-2 'immediate data' feature, the default
> > maximum initiator to target IU size is significantly samller than
>                                                        ^^^^^^^
>                                                        smaller?

Yes, will fix this typo.

> > 8260. For 'ib_srpt.ko' module, which built from source before
> > [2], the default maximum initiator to target IU is 2116.
> [ ... ]
> > diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
> > index b5960351bec0..2eadb038b257 100644
> > --- a/drivers/infiniband/ulp/srp/ib_srp.c
> > +++ b/drivers/infiniband/ulp/srp/ib_srp.c
> > @@ -75,6 +75,7 @@ static bool prefer_fr = true;
> >   static bool register_always = true;
> >   static bool never_register;
> >   static int topspin_workarounds = 1;
> > +static uint32_t srp_opt_max_it_iu_size;
> 
> Each SCSI host can represent a connection to another SRP target, and the
> max_it_iu_size parameter can differ per target. So I think this variable
> should be moved into struct srp_target_port instead of being global. See
> also srp_max_it_iu_len().

Yes, will do as you said.

thanks
diff mbox series

Patch

diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index b5960351bec0..2eadb038b257 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -75,6 +75,7 @@  static bool prefer_fr = true;
 static bool register_always = true;
 static bool never_register;
 static int topspin_workarounds = 1;
+static uint32_t srp_opt_max_it_iu_size;
 
 module_param(srp_sg_tablesize, uint, 0444);
 MODULE_PARM_DESC(srp_sg_tablesize, "Deprecated name for cmd_sg_entries");
@@ -3411,6 +3412,7 @@  enum {
 	SRP_OPT_IP_SRC		= 1 << 15,
 	SRP_OPT_IP_DEST		= 1 << 16,
 	SRP_OPT_TARGET_CAN_QUEUE= 1 << 17,
+	SRP_OPT_MAX_IT_IU_SIZE  = 1 << 18,
 };
 
 static unsigned int srp_opt_mandatory[] = {
@@ -3443,6 +3445,7 @@  static const match_table_t srp_opt_tokens = {
 	{ SRP_OPT_QUEUE_SIZE,		"queue_size=%d"		},
 	{ SRP_OPT_IP_SRC,		"src=%s"		},
 	{ SRP_OPT_IP_DEST,		"dest=%s"		},
+	{ SRP_OPT_MAX_IT_IU_SIZE,	"max_it_iu_size=%d"	},
 	{ SRP_OPT_ERR,			NULL 			}
 };
 
@@ -3736,6 +3739,14 @@  static int srp_parse_options(struct net *net, const char *buf,
 			target->tl_retry_count = token;
 			break;
 
+		case SRP_OPT_MAX_IT_IU_SIZE:
+			if (match_int(args, &token) || token < 0) {
+				pr_warn("bad maximum initiator to target IU size '%s'\n", p);
+				goto out;
+			}
+			srp_opt_max_it_iu_size = token;
+			break;
+
 		default:
 			pr_warn("unknown parameter or missing value '%s' in target creation request\n",
 				p);