[PATCH 1/2] iw_cxgb4: limit IRD/ORD advertised to ULP by device max.

[PATCH 1/2] iw_cxgb4: limit IRD/ORD advertised to ULP by device max.

[Steve Wise]

Problem Description: i40iw initiator sends MPA-request with ird =
63, ord = 63. The cxgb4 responder sends a RST.  Since the inbound ord=63
and it exceeds the max_ird/c4iw_max_read_depth (=32 default), chelsio
decides to abort.

Instead, cxgb4 should adjust the ord/ird down before presenting it to
the ULP.

Reported-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 3aca7f6..a698efc 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1827,8 +1827,12 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 				(ep->mpa_pkt + sizeof(*mpa));
 			ep->ird = ntohs(mpa_v2_params->ird) &
 				MPA_V2_IRD_ORD_MASK;
+			ep->ird = min_t(u32, ep->ird,
+					cur_max_read_depth(ep->com.dev));
 			ep->ord = ntohs(mpa_v2_params->ord) &
 				MPA_V2_IRD_ORD_MASK;
+			ep->ord = min_t(u32, ep->ord,
+					cur_max_read_depth(ep->com.dev));
 			PDBG("%s initiator ird %u ord %u\n", __func__, ep->ird,
 			     ep->ord);
 			if (ntohs(mpa_v2_params->ird) & MPA_V2_PEER2PEER_MODEL)
-- 
2.7.0

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 2/2] iw_cxgb4: use the MPA initiator's IRD if < our ORD

[Steve Wise]

Problem Description: The i40iw initiator sends an MPA-request with ird=16
and ord=16. The cxgb4 responder sends an MPA-reply with ord = 32 causing
i40iw to terminate due to insufficient resources.

The logic to reduce the ORD to <= peer's IRD was wrong.

Reported-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index a698efc..b6a953a 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -3140,7 +3140,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
 		if (conn_param->ord > ep->ird) {
 			if (RELAXED_IRD_NEGOTIATION) {
-				ep->ord = ep->ird;
+				conn_param->ord = ep->ird;
 			} else {
 				ep->ird = conn_param->ird;
 				ep->ord = conn_param->ord;
-- 
2.7.0

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 0/2] cxgb4/i40iw interoperability fixes

[Steve Wise]

Doug, these would be good for 4.8-rc.  They fix some interoperability
problems between Intel and Chelsio iWARP devices.

Thanks,

Steve.

----

Steve Wise (2):
  iw_cxgb4: limit IRD/ORD advertised to ULP by device max.
  iw_cxgb4: use the MPA initiator's IRD if < our ORD

 drivers/infiniband/hw/cxgb4/cm.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

-- 
2.7.0

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/2] cxgb4/i40iw interoperability fixes

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 840 bytes --]

On Wed, Aug 17, 2016 at 10:07:41AM -0700, Steve Wise wrote:
> Doug, these would be good for 4.8-rc.  They fix some interoperability
> problems between Intel and Chelsio iWARP devices.

Hi Steve,
Is there any chance you could remove "Problem description:" words from commit messages?

Thanks

> 
> Thanks,
> 
> Steve.
> 
> ----
> 
> Steve Wise (2):
>   iw_cxgb4: limit IRD/ORD advertised to ULP by device max.
>   iw_cxgb4: use the MPA initiator's IRD if < our ORD
> 
>  drivers/infiniband/hw/cxgb4/cm.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> -- 
> 2.7.0
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCH 0/2] cxgb4/i40iw interoperability fixes

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 551 bytes --]

On Thu, Aug 18, 2016 at 08:54:39AM -0500, Steve Wise wrote:
> > On Wed, Aug 17, 2016 at 10:07:41AM -0700, Steve Wise wrote:
> > > Doug, these would be good for 4.8-rc.  They fix some interoperability
> > > problems between Intel and Chelsio iWARP devices.
> > 
> > Hi Steve,
> > Is there any chance you could remove "Problem description:" words from commit
> > messages?
> 
> I can.  But why is that a problem? (pun intended :) )

It is not a problem if you can. We are thrilled to see it happening,
please show us :)

> 
> Steve.
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]