While this change is technically correct and should remain in the
trunk, in both of these cases we are only using the convertor to get
the address of the users buffer. The data is contiguous in both these
cases so the size should not be changed by the convertor, as far as I
know. The short of this, I don't think this needs to be in v1.0 as it
is essentially a no-op as max_data = size.
Thanks,
Galen
On Nov 22, 2005, at 3:53 PM, Jeff Squyres wrote:
George/Tim/Galen --
Does this need to come over to v1.0?
On Nov 22, 2005, at 4:51 PM, bosi...@osl.iu.edu wrote:
Author: bosilca
Date: 2005-11-22 16:51:04 -0500 (Tue, 22 Nov 2005)
New Revision: 8237
Modified:
trunk/ompi/mca/btl/mvapi/btl_mvapi.c
Log:
Update the correct sended size.
Modified: trunk/ompi/mca/btl/mvapi/btl_mvapi.c
===================================================================
--- trunk/ompi/mca/btl/mvapi/btl_mvapi.c 2005-11-22 17:24:47
UTC (rev
8236)
+++ trunk/ompi/mca/btl/mvapi/btl_mvapi.c 2005-11-22 21:51:04
UTC (rev
8237)
@@ -294,10 +294,10 @@
iov.iov_base = NULL;
ompi_convertor_pack(convertor, &iov, &iov_count, &max_data,
&free_after);
+ *size = max_data;
frag->segment.seg_len = max_data;
frag->segment.seg_addr.pval = iov.iov_base;
-
reg_len = (unsigned char*)vapi_reg->base_reg.bound -
(unsigned char*)iov.iov_base + 1;
frag->sg_entry.len = max_data;
@@ -337,8 +337,8 @@
iov.iov_base = NULL;
ompi_convertor_pack(convertor, &iov, &iov_count, &max_data,
&free_after);
+ *size = max_data;
-
frag->segment.seg_len = max_data;
frag->segment.seg_addr.pval = iov.iov_base;
frag->base.des_flags = 0;
_______________________________________________
svn-full mailing list
svn-f...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/svn-full
--
{+} Jeff Squyres
{+} The Open MPI Project
{+} http://www.open-mpi.org/
_______________________________________________
devel mailing list
de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel