glusterfs.git/rpc/rpc-transport/rdma, branch v3.1.6

LICENSE: s/GNU Affero General Public/GNU General Public/

2011-08-06T13:36:02+00:00

Change-Id: I30b1807f4493a9b4ffa9899548daf7d29a0a6364
BUG: 3348
Reviewed-on: http://review.gluster.com/184
Tested-by: Gluster Build System 
Reviewed-by: Anand Avati

rpc-transport/rdma: call ibv_fork_init to make rdma work with fork.

2011-07-01T08:26:42+00:00

Signed-off-by: Raghavendra G 
Signed-off-by: Anand Avati 

BUG: 3091 (rebalance fails with "transport endpoint not connected" in 3.2.1 rdma set-up)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=3091

rpc-transport/rdma: don't return '0' in case of un-initiated rdma_connect()

2011-06-19T09:16:23+00:00

needed to handle the re-configuration of the port in client protocol. If
there is a successful 'connect()' after the query port, the remote-port
value gets reset to '0'. In this case, if the connect() return 0 for
already connected socket, the query_portmap will keep happening in a loop.

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand Avati 

BUG: 2258 (enhance gluster volume rebalance)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2258

rpc-transport/rdma: handle cases where peer is NULL in send_completion.

2011-03-12T13:10:39+00:00

- There can be valid cases where peer is NULL - For eg., race b/w pollin
    received on handshake socket and as a result transport getting disconnected
    and handling of a send/recv completion on the same qp for some previous
    read/write - in send/recv_completion_process. Hence we should be checking
    for a valid peer pointer before accessing it.

Signed-off-by: Raghavendra G 
Signed-off-by: Vijay Bellur 

BUG: 2514 ()
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2514

rdma : prevent confusing logs when both 'poll_in' and 'poll_err' are received.

2010-12-28T04:51:11+00:00

Signed-off-by: Amar Tumballi 
Signed-off-by: Anand V. Avati 

BUG: 875 (Implement a new protocol to provide proper backward/forward compatibility)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=875

rpc-transport/rdma: QP configuration changes.

2010-12-13T04:07:45+00:00

- set send and recv count to 4K.
  - set max_send_sge of QP to 2 since for msgs of type RDMA_NOMSG, there
    can be two vectors - rpc header and message from program.

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 2197 (applications running on large number of clients simultaneously result in ibv_post_send errors on server.)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2197

rpc-transport/rdma: add informative debug messages when posting of send requests fail.

2010-12-13T04:07:39+00:00

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 2197 (applications running on large number of clients simultaneously result in ibv_post_send errors on server.)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2197

rpc-transport: don't merge payload iobuf and iobuf which holds program header into single iobref.

2010-11-25T08:34:09+00:00

- io-cache holds a reference on iobref passed from transport layer. Hence,
   two iobufs are accounted instead of one in calculated used cache size.

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 2135 ()
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2135

rpc-transport/rdma: set the default quota of recv buffers to 128.

2010-11-09T10:12:11+00:00

- since quota is returned once the reply is recieved, 32 is too less value.

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 2060 (Infiniteband errors on large file find)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=2060

rpc-transport/rdma: create request_context only if message contains read chunks or write chunks.

2010-11-08T04:15:13+00:00

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 1992 (glusterd crashed while submitting a callback.)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1992