Does this need to go to v1.3?
On May 27, 2009, at 12:12 PM, <bosi...@osl.iu.edu> wrote:
Author: bosilca
Date: 2009-05-27 12:12:18 EDT (Wed, 27 May 2009)
New Revision: 21300
URL: https://svn.open-mpi.org/trac/ompi/changeset/21300
Log:
Don't add the offset to all segments, only the first one should be
affected. Thanks
to Roberto Ammendola for this bug report and patch.
Text files modified:
trunk/ompi/mca/pml/ob1/pml_ob1_recvreq.h | 7 +++++--
1 files changed, 5 insertions(+), 2 deletions(-)
Modified: trunk/ompi/mca/pml/ob1/pml_ob1_recvreq.h
=
=
=
=
=
=
=
=
======================================================================
--- trunk/ompi/mca/pml/ob1/pml_ob1_recvreq.h (original)
+++ trunk/ompi/mca/pml/ob1/pml_ob1_recvreq.h 2009-05-27 12:12:18
EDT (Wed, 27 May 2009)
@@ -179,8 +179,9 @@
static inline bool
recv_request_pml_complete_check(mca_pml_ob1_recv_request_t *recvreq)
{
+#if OPAL_HAVE_THREAD_SUPPORT
opal_atomic_rmb();
-
+#endif
if(recvreq->req_match_received &&
recvreq->req_bytes_received >= recvreq-
>req_recv.req_bytes_packed &&
lock_recv_request(recvreq)) {
@@ -218,8 +219,9 @@
req->req_recv.req_base.req_ompi.req_status.MPI_SOURCE = hdr-
>hdr_src;
req->req_recv.req_base.req_ompi.req_status.MPI_TAG = hdr-
>hdr_tag;
req->req_match_received = true;
+#if OPAL_HAVE_THREAD_SUPPORT
opal_atomic_wmb();
-
+#endif
if(req->req_recv.req_bytes_packed > 0) {
#if OPAL_ENABLE_HETEROGENEOUS_SUPPORT
if(MPI_ANY_SOURCE == req->req_recv.req_base.req_peer) {
@@ -262,6 +264,7 @@
iov[iov_count].iov_base =
(IOVBASE_TYPE*) \
((unsigned char*)segment->seg_addr.pval +
offset); \
iov_count+
+; \
+ offset =
0; \
}
\
}
\
PERUSE_TRACE_COMM_OMPI_EVENT
(PERUSE_COMM_REQ_XFER_CONTINUE, \
_______________________________________________
svn-full mailing list
svn-f...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/svn-full
--
Jeff Squyres
Cisco Systems