Thanks.
The following hunks from 98719d71f58d76616e8ba755f70f39f843ec540f do not apply at ALL.... There is no pi_support in 3.12.19
diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 826eaf5..a171046 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -562,7 +562,15 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
struct isert_device *device;
struct ib_device *ib_dev = cma_id->device;
int ret = 0;
- u8 pi_support = np->tpg_np->tpg->tpg_attrib.t10_pi;
+ u8 pi_support;
+
+ spin_lock_bh(&np->np_thread_lock);
+ if (!np->enabled) {
+ spin_unlock_bh(&np->np_thread_lock);
+ pr_debug("iscsi_np is not enabled, reject connect request\n");
+ return rdma_reject(cma_id, NULL, 0);
+ }
+ spin_unlock_bh(&np->np_thread_lock);
pr_debug("Entering isert_connect_request cma_id: %p, context: %p\n",
cma_id, cma_id->context);
@@ -653,6 +661,7 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
goto out_mr;
}
+ pi_support = np->tpg_np->tpg->tpg_attrib.t10_pi;
if (pi_support && !device->pi_capable) {
pr_err("Protection information requested but not supported\n");
ret = -EINVAL;
> -----Original Message-----
> From: Nicholas A. Bellinger [mailto:nab@xxxxxxxxxxxxxxx]
> Sent: Thursday, May 15, 2014 3:22 PM
> To: Moussa Ba (moussaba)
> Cc: Sagi Grimberg; target-devel@xxxxxxxxxxxxxxx; Nicholas Bellinger; Or
> Gerlitz
> Subject: Re: 3.12.5 Target Errors
>
> On Thu, 2014-05-15 at 21:21 +0000, Moussa Ba (moussaba) wrote:
> > Thanks Nick. I just updated to 3.14.4, does that include the
> necessary patches?I don't have any particular requirement for version.
>
> Minus the three fixes from Sagi that are queued in
> target-pending/master, v3.14.4 has all of the other iser-target related
> fixes.
>
> You'll still want to apply these three mentioned earlier..
>
> > More interested in functionality and stability. 3.14.4 generates the
> following error:
> >
> > [ 271.008891] rdma_accept() failed with: -99
> > [ 271.904227] rdma_accept() failed with: -99
> > [ 271.933538] rdma_accept() failed with: -99
> >
> >
>
> Mmmm, rdma_accept() is returning -EADDRNOTAVAIL here..
>
> It appears that something is failing in drivers/infiniband/core/cma.c
> outside of ULP specific code.. (Or CC'ed).
>
> Note that there was a userspace ABI version bump for libibverbs,
> libmlx4, and librdmacm during v3.14.. You might be hitting some of
> that
> related fallout here somehow..
>
> > 8 of targets have disappeared from vmware 5.5, but they are present
> > and show enabled on the target. Anyhow, what is the recommended
> > version for stability/functionality.
>
> v3.12.19 with the 5 patches applied is probably the best bet here,
> until
> the -EADDRNOTAVAIL issue can be tracked down on v3.14.y.
>
> --nab
��.n��������+%������w��{.n����j� ����{ay� ʇڙ���f���h������_�(�階�ݢj"��������G����?���&��
