[openib-general] [PATCH] RDMA/addr: Fix some cancellation problems in process_req().

[Roland Dreier]

thanks, queued for 2.6.20.

BTW, I noticed the code after your change:

 		list_del(&req->list);
 		list_add_tail(&req->list, &done_list);

So I also queued the patch below for a savings of a whopping 15 bytes
of .text on x86_64:

commit 922e0c40e423bb52246611b3117399a56e1da318
Author: Roland Dreier <rolandd at cisco.com>
Date:   Mon Nov 27 19:38:32 2006 -0800

    RDMA/addr: list_move() cleanups
    
    Replace a couple list_del()/list_add() combos with list_move().
    
    Signed-off-by: Roland Dreier <rolandd at cisco.com>

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 15ba518..7767a11 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -235,8 +235,7 @@ static void process_req(void *data)
 			else if (req->status == -ENODATA)
 				continue;
 		}
-		list_del(&req->list);
-		list_add_tail(&req->list, &done_list);
+		list_move_tail(&req->list, &done_list);
 	}
 
 	if (!list_empty(&req_list)) {
@@ -346,8 +345,7 @@ void rdma_addr_cancel(struct rdma_dev_ad
 		if (req->addr == addr) {
 			req->status = -ECANCELED;
 			req->timeout = jiffies;
-			list_del(&req->list);
-			list_add(&req->list, &req_list);
+			list_move(&req->list, &req_list);
 			set_timeout(req->timeout);
 			break;
 		}