[openib-general] Re: RFC: e2e credits

[Michael S. Tsirkin]

Quoting r. Sean Hefty <sean.hefty at intel.com>:
> Subject: RFC: e2e credits
> 
> >Sean, just to wrap it up, the API at the verbs layer will look like the below,
> >and then ULPs just put the value they want in the CM and CM will pass
> >it in to low level.
> 
> I'm fine with this, but I do think that it's a minor spec violation/enhancement,
> so I'd like to get agreement with Hal, Roland, and other HCA vendors about this
> change.
> 
> - Sean

I think Hal spoke on this already.
Guys, please review the verbs.h extension below - I'll be submitting a patch
including CM update and mthca implementation next week.
Anyone has a problem with this?

Index: openib/drivers/infiniband/include/rdma/ib_user_verbs.h
===================================================================
--- openib.orig/drivers/infiniband/include/rdma/ib_user_verbs.h	2006-03-02 21:41:01.000000000 +0200
+++ openib/drivers/infiniband/include/rdma/ib_user_verbs.h	2006-03-23 00:50:48.000000000 +0200
@@ -378,7 +378,8 @@ struct ib_uverbs_qp_attr {
 	__u8	rnr_retry;
 	__u8	alt_port_num;
 	__u8	alt_timeout;
-	__u8	reserved[5];
+	__u8    flow_control;
+	__u8	reserved[4];
 };
 
 struct ib_uverbs_create_qp {
Index: openib/drivers/infiniband/include/rdma/ib_verbs.h
===================================================================
--- openib.orig/drivers/infiniband/include/rdma/ib_verbs.h	2006-03-02 21:41:01.000000000 +0200
+++ openib/drivers/infiniband/include/rdma/ib_verbs.h	2006-03-23 00:50:28.000000000 +0200
@@ -573,6 +573,7 @@ struct ib_qp_attr {
 	u8			rnr_retry;
 	u8			alt_port_num;
 	u8			alt_timeout;
+	u8			flow_control;
 };
 
 enum ib_wr_opcode {



-- 
Michael S. Tsirkin
Staff Engineer, Mellanox Technologies