From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Guay Wei Lin" Subject: Questions on RTS->SQD Date: Tue, 22 Dec 2009 10:56:57 +0100 Message-ID: <000001ca82ed$1c67b8a0$553729e0$@no> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: References: In-Reply-To: Content-Language: en-us Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org Hi, As for an experiment for my research to try out the SQD state, I've reused the perftest code (send_bw.c) to modify the QP state from RTS->SQD, and query_qp to make sure the QP is in the SQD state. The codes were added before ibv_post_send function. (to make sure no outstanding WR in the SQ). When the ibv_modify_qp is executed to modify QP from RTS->SQD, it returns a 0 without any error code, but when the ibv_query_qp to query on the QP_state, the QP is still in RTS state. I'm using the mlx4 HCAs. Any ideas/suggestions why this will happen or I miss out something? Thanks, Wei Lin PhD student Simula School of Research and Innovation AS -- 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