- 04 Mar, 2015 1 commit
-
-
The implementations of sendNoncontig for intra-node communication in Nemesis and inter-node communication in network modules (except for TCP and SCIF) assume that req->dev.segment_first is zero and req->dev.segment_size is the size of data, which is not always true. If we stream an RMA operation and issue partial of derived data, req->dev.segment_first specifies the current starting location of the data and req->dev.segment_size specifies the current ending location of the data. Also, the data size should be (req->dev.segment_size - req->dev.segment_first). This patch corrects this issue in Nemesis and network modules. Signed-off-by:
Pavan Balaji <balaji@anl.gov>
-
- 10 Oct, 2012 1 commit
-
-
David Goodell authored
By setting "indent-tabs-mode:nil" we should hopefully begin to slowly squeeze out hard tabs from the source without a disruptive (to downstream projects) whitespace-fixing change. No reviewer.
-
- 10 Jul, 2012 1 commit
-
-
Darius Buntinas authored
-
- 15 Apr, 2012 1 commit
-
-
David Goodell authored
Without this change we are vulnerable to a nasty bit of compiler behavior. MPIDI_CH3_PktGeneric_t sometimes included padding between the "kind" and "pktptrs" fields. Then whole-structure assignments using the generic packet (e.g., ch3_istartmsg.c:96) of buffers that are actually MPIDI_CH3_Pkt_t unions under the hood will can fail to copy key parts of the MPIDI_CH3_Pkt_t. This specifically manifested itself for #1599 by transmitting garbage "source_win_handle" values in a LOCK_GRANTED packet, which in turn caused segfaults. Removing this second type and directly using the actual packet union eliminates this source of bugs. The minor downside is that the mpidpkt.h header becomes more tightly integrated into the rest of the device. Reviewed by buntinas@.
-
- 09 Dec, 2011 1 commit
-
-
Guillaume Mercier authored
-
- 08 Dec, 2011 1 commit
-
-
Guillaume Mercier authored
-
- 07 Dec, 2011 1 commit
-
-
Guillaume Mercier authored
[svn-r9301] VC now check for pensing send messages - fix for memory leak with iov - unexpected callback issue fixed
-
- 10 Nov, 2011 1 commit
-
-
Guillaume Mercier authored
-
- 23 Mar, 2011 1 commit
-
-
Guillaume Mercier authored
-
- 28 Oct, 2010 2 commits
-
-
Guillaume Mercier authored
-
Guillaume Mercier authored
-
- 27 Oct, 2010 1 commit
-
-
Guillaume Mercier authored
-
- 31 Mar, 2010 1 commit
-
-
Darius Buntinas authored
-
- 08 Mar, 2010 1 commit
-
-
Guillaume Mercier authored
-
- 05 Mar, 2010 3 commits
-
-
Guillaume Mercier authored
-
Guillaume Mercier authored
-
Guillaume Mercier authored
-
- 01 Dec, 2009 1 commit
-
-
Guillaume Mercier authored
-
- 12 Jun, 2009 1 commit
-
-
Darius Buntinas authored
-
- 30 Mar, 2009 1 commit
-
-
Guillaume Mercier authored
-
- 26 Mar, 2009 1 commit
-
-
Guillaume Mercier authored
-
- 13 Mar, 2009 1 commit
-
-
Jayesh Krishna authored
[svn-r4062] 1) Modifying the datatype code to find the number of contig blocks in an instance of an MPI derived datatype - Since finding the real number of contig blocks is not easy we find a reasonable upper bound instead. This also fixes the case where the number of contig blocks was uninitialized for contiguous datatypes. Refer ticket #428 for details (2) Adding a test case, test1_dt.c, to test the fix - Review @ rross, thakur
-
- 05 Mar, 2009 1 commit
-
-
Darius Buntinas authored
[svn-r3945] Merging mx netmod branch into trunk. This adds override functions to the vc for overriding mpid_ send functions, as well as hooks for posting receives.
-