1
1
Граф коммитов

422 Коммитов

Автор SHA1 Сообщение Дата
Tim Woodall
6523c12e4b - decrease eager limit to 12K (improves latency)
- trigger event library while setting up connections

This commit was SVN r9645.
2006-04-14 22:28:05 +00:00
Tim Woodall
c6489cb5aa - turn on eager rdma by default
This commit was SVN r9641.
2006-04-14 21:11:14 +00:00
George Bosilca
b3cc3d82d3 Activate the OOB while we setup connections for MVAPI. Same thing should be done for the
Open IB ...

This commit was SVN r9640.
2006-04-14 20:53:42 +00:00
Gleb Natapov
98282a3567 fix spelling. threashold -> threshold.
This commit was SVN r9577.
2006-04-08 08:13:37 +00:00
Andrew Friedley
d461b55696 - Implement OOB connection handshaking via the ORTE RML. To start a connect,
we send our local addr_t OOB.  Remote side then matches endpoints and calls
  dat_ep_connect().  Everything should be the same as before from here, except
  that client/server roles are reversed.
- Properly set our buffer size when posting receives.  When the frag used to
  transfer address information is recycled by the free list, the wrong buffer
  size was being used, which caused buffer overflow errors.
- Finally put the uDAPL error handling stuff in the mpool component.
- Remove a few more OPAL_OUTPUTs.

This commit was SVN r9569.
2006-04-07 15:26:05 +00:00
Gleb Natapov
b6ab1f4262 fix compilation warnings.
This commit was SVN r9515.
2006-04-02 11:32:25 +00:00
Andrew Friedley
74b2f77a4c The expected cleanup/refactoring commit..
Not much got tested that wasn't already - I've uncovered a connection
establishment deadlock and wanted to get these changes committed before I
attack it.

The big changes:
 - Moved much of the connection code from btl_udapl_component.c to
   btl_udapl_endpoint.c.
 - Cleaned up initialization of various fragment members.
 - MCA_BTL_UDAPL_ERROR macro, which is compiled in/out appropriately.

This commit was SVN r9496.
2006-03-31 16:25:19 +00:00
Gleb Natapov
256bf70530 Forgot to add file to previous commit
This commit was SVN r9480.
2006-03-30 17:37:52 +00:00
Gleb Natapov
79bcfb096f Add type to frag. Sometimes we need to know that a frag is from short rdma area.
I used hack for this that doesn't work for mvapi, so changing it to something more sane.

This commit was SVN r9477.
2006-03-30 15:26:21 +00:00
Gleb Natapov
ea11582191 Porting of short message RDMA from openib BTL. Endpoint registers circular buffer and sends its address and rkey to the peer. Peer uses this buffer to eagerly RDMA small message into it. Endpoint polls the buffer for message arrival before checking HP/LP QPs. Set btl_mvapi_use_eager_rdma to 1 to enable it.
This commit was SVN r9474.
2006-03-30 12:55:31 +00:00
Andrew Friedley
0eba366b07 Various pieces all over to make basic small message send/recv work. Next step
is clean up the code.. it is in need of refactoring and testing.

Thanks to Brian for help in troubleshooting!

This commit was SVN r9466.
2006-03-29 21:55:41 +00:00
Gleb Natapov
590c992a7e fix recursive lock of openib_btl->ib_lock.
This commit was SVN r9427.
2006-03-26 15:02:43 +00:00
Gleb Natapov
01a119c3c5 fix compilation bug with --enable-mpi-threads
This commit was SVN r9426.
2006-03-26 13:24:10 +00:00
Gleb Natapov
a5a78b10cc Implementation of short message RDMA. Endpoint registers circular buffer and sends its address and rkey to the peer. Peer uses this buffer to eagerly RDMA small message into it. Endpoint polls the buffer for message arrival before checking HP/LP QPs. Set btl_openib_use_eager_rdma to 1 to enable it.
This commit was SVN r9425.
2006-03-26 08:30:50 +00:00
Andrew Friedley
48d61cd99a Mostly fragment/LMR handling fixes:
- Grab the mpool_registration in _frag_common_constructor()
 - Save the LMR context in the segment key
 - No need for cookie variables - can just cast the frag
 - No need to memcpy() data when recv'ing
 - Add an LMR triplet to the fragment structure and initialize it
   in btl_udapl_alloc().
 - Whitespace/typo fixes, remove some opal_output() calls

Looks like I can use triplets describing sub-regions of registered LMR's.  So I
do this - prior to this patch I was sending the entire free list memory over,
which isn't correct :)

Back to an earlier problem - when sending address information right after
connection establishment, the receiving end receives a DTO completion event and
appears to have good data.  But the sending end never receives a DTO completion
event indicating the send completed, and never completes the client side of the
connection.

This commit was SVN r9386.
2006-03-23 16:21:08 +00:00
Tim Woodall
c7ee5e13bc simplification - dont swap src/dst pointers - always leave both
src/dst pointing to same segments

This commit was SVN r9357.
2006-03-21 18:20:17 +00:00
George Bosilca
f7a5a582c5 Diagnostic function for mvapi. It print all the credits used for the flow control.
This commit was SVN r9355.
2006-03-21 17:02:14 +00:00
Andrew Friedley
cf9246f7b9 Long overdue commit.. many changes.
In short, I'm very close to having connection establishment and eager send/recv working.

Part of the connection process involves sending address information from the
client to server.  For some reason, I am never receiving an event indicating
completetion of the send on the client side.  Otherwise, connection
establishment is working and eager send/recv should be trivial from here.


Some more detailed changes:
 - Send partially implemented, just handles starting up new connections.
 - Several support functions implemented for establishing connection.  Client
   side code went in btl_udapl_endpoint.c, server side in btl_udapl_component.c
 - Frags list and send/recv locks added to the endpoint structure.
 - BTL sets up a public service point, which listens for new connections.
   Steps over ports that are already bound, iterating through a range of ports.
 - Remove any traces of recv frags, don't think I need them after all.
 - Pieces of component_progress() implemented for connection establishment.
 - Frags have two new types for connection establishment - CONN_SEND and
   CONN_RECV.
 - Many other minor cleanups not affecting functionality

This commit was SVN r9345.
2006-03-21 00:12:55 +00:00
George Bosilca
e181153f16 Remove the bogus prototype.
This commit was SVN r9333.
2006-03-19 19:22:35 +00:00
George Bosilca
a0d25ab6ef Add missing prototype for the mvapi diagnostic function.
This commit was SVN r9331.
2006-03-18 19:38:56 +00:00
Tim Woodall
bd870519fd - modified convertor copy_and_prepare routines to accept an addition
flag, new flags to be included when convertor is initialized
- modified pml/btl module defs and added stub functions for diagnostic
  output routines to dump state of queues / endpoints
- updates to data reliability pml

This commit was SVN r9329.
2006-03-17 18:46:48 +00:00
Tim Woodall
712468dbef add diagnostic interface
This commit was SVN r9328.
2006-03-17 17:39:41 +00:00
Tim Woodall
c34f4c2cb7 correct cleanup for threaded case
This commit was SVN r9291.
2006-03-16 00:05:39 +00:00
Galen Shipman
440417e92c Add max_btls option
This commit was SVN r9263.
2006-03-13 17:03:21 +00:00
Sven Stork
12b94972e2 Fix comment of a paramter.
This commit was SVN r9261.
2006-03-13 09:11:46 +00:00
Brian Barrett
d041558f85 * protect lock when not building threaded
This commit was SVN r9254.
2006-03-11 03:20:50 +00:00
Brian Barrett
3e2c51dea8 * fix some silly commenting done by a previous developer that are good for
a laugh but probably not good for usability ;)

This commit was SVN r9253.
2006-03-11 03:09:24 +00:00
Tim Woodall
9ae910044b resolve threading issue
This commit was SVN r9234.
2006-03-09 17:59:05 +00:00
Tim Woodall
8bf6ed7a36 - corrected locking in gm btl - gm api is not thread safe
- initial support for gm progress thread
- corrected threading issue in pml
- added polling progress for a configurable number of cycles to wait for threaded case

This commit was SVN r9188.
2006-03-02 00:39:07 +00:00
Brian Barrett
579e74290f * make gm wire-up endian safe
This commit was SVN r9179.
2006-02-28 02:03:46 +00:00
Brian Barrett
bfd49d248b * (hopefully) fix MPI_BOTTOM for portals, same way as oll the other RDMA btls from
eons ago...

This commit was SVN r9172.
2006-02-27 17:07:24 +00:00
Brian Barrett
9b19e3fef0 * remove some debugging output that shouldn't have been committed. Doh!
This commit was SVN r9171.
2006-02-27 16:23:52 +00:00
Brian Barrett
285581dff2 More endian-related cleanups:
- moved hton64 and ntoh64 from the bunch of places it had been copied
    into one header file
  - properly set and use the btl_tcp's nbo option to put things in
    network byte order on the wire if both sides don't have the same
    endianness
  - Put the OB1 PML's headers (with a couple exceptions I need to discuss
    with Tim) in network byte order on the wire if both sides don't have
    the same endianness
  - since it was needed for the TCP BTL, move the orte_process_name_t
    HTON and NTOH macros from the TCP OOB to ns_types.h

This commit was SVN r9145.
2006-02-26 00:45:54 +00:00
Jeff Squyres
628125599d Fix the TCL btl module endpoint matching during setup for the scenario
when running an MPI job spanning a node that has two TCP NICs and a
node that has one TCP NIC.  Previously, for the 2 NIC/module process,
we would return the first peer IP address if we couldn't find a subnet
match with any of the peer's published IP addresses -- this was to
support running OMPI across subnet boundaries.  Changed the behavior
to only do that behavior if the IP address we're trying to match is
public (i.e., not 10.x.y.z, 192.168.x.y, or 172.16.x.y) *and* any of
the remote peer's addresses are public (working on the assumption that
if we both have public addresses, they're routable to each other).

This definitely will not work in all scenarios, such as when we go to
WAN kinds of executions, and will need to be revisited at that time.

This commit was SVN r9119.
2006-02-23 02:02:19 +00:00
Galen Shipman
e58b758031 standardize behavior of btl_alloc, if the size is larger than the max send
size, btl_alloc returns NULL. 

This commit was SVN r9114.
2006-02-22 17:37:59 +00:00
Brian Barrett
0d098c9d57 * after talking with galen, take into account we might truncate
This commit was SVN r9113.
2006-02-22 17:03:04 +00:00
Brian Barrett
765d2ffc29 * the self btl should set the segment size field on alloc like the other btls
* clean up duplicate free in long message accumulates that looks like it was
  a cut-n-paste error

This commit was SVN r9112.
2006-02-22 16:20:13 +00:00
Brian Barrett
08747dcaf8 * Throttle the number of incoming receives so that we don't overrun our receive
event queue and lose receive messages.

This commit was SVN r9006.
2006-02-13 15:59:54 +00:00
Brian Barrett
20d06e889e * revert out some of the attempts to better use the Portals 3.3.2-2 user-space
run-time support, as it appears to be doing bad things to memory.  Update
  the hack to get the local nid to match the recent TCP nal changes, and
  update the P3RT api useage

This commit was SVN r9005.
2006-02-13 15:41:00 +00:00
George Bosilca
ecc3e00362 Various cleanups.
This commit was SVN r9002.
2006-02-12 21:36:07 +00:00
Brian Barrett
566a050c23 Next step in the project split, mainly source code re-arranging
- move files out of toplevel include/ and etc/, moving it into the
    sub-projects
  - rather than including config headers with <project>/include, 
    have them as <project>
  - require all headers to be included with a project prefix, with
    the exception of the config headers ({opal,orte,ompi}_config.h
    mpi.h, and mpif.h)

This commit was SVN r8985.
2006-02-12 01:33:29 +00:00
Andrew Friedley
b37e18916f Many different things, the big ones:
- Start filling in the progress function, focusing on connection establishment.
 - Initialize udapl mpool and free lists
 - Create/destroy a protection zone with each IA
 - Misc organization as I learn how things work

This commit was SVN r8969.
2006-02-10 21:49:15 +00:00
George Bosilca
9f1357fb89 Remove all the useless includes. Most of the endpoint do not depend on the
orte includes.

This commit was SVN r8932.
2006-02-08 05:10:48 +00:00
Galen Shipman
c8045bf397 Fixup for ORTE datatype checkin,
- use appropriate header files 
- change calls from orte_dps to orte_dss 

This commit was SVN r8920.
2006-02-07 15:20:44 +00:00
George Bosilca
20fd358327 A BTL cannot depend on the orte data-types.
This commit was SVN r8915.
2006-02-07 06:05:36 +00:00
Ralph Castain
4b9f015c0b Merge in the new data support subsystem for ORTE. MPI folks should not notice a difference. Longer explanation will be sent to developers mailing list.
This commit was SVN r8912.
2006-02-07 03:32:36 +00:00
Rainer Keller
7ac0ffc349 - Instead of doing the unlock inside the if, just move the if-statement
later after the mandatory unlock.

This commit was SVN r8885.
2006-02-02 17:32:22 +00:00
Tim Woodall
a2fde48f2f changes from release branch
This commit was SVN r8858.
2006-01-31 16:17:18 +00:00
Tim Woodall
bcd6c525f8 removed duplicate locks
This commit was SVN r8857.
2006-01-31 16:12:37 +00:00
Tim Woodall
9d484916db remove locks already held
This commit was SVN r8853.
2006-01-31 14:23:08 +00:00
Brian Barrett
b1d2424013 Merge in present work on the MPI-2 onesided chapter. The current code is not
complete, but stable enough that it will have no impact on general development,
so into the trunk it goes.  Changes in this commit include:

 - Remove the --with option for disabling MPI-2 onesided support.  It
   complicated code, and has no real reason for existing
 - add a framework osc (OneSided Communication) for encapsulating
   all the MPI-2 onesided functionality
 - Modify the MPI interface functions for the MPI-2 onesided chapter
   to properly call the underlying framework and do the required
   error checking
 - Created an osc component pt2pt, which is layered over the BML/BTL
   for communication (although it also uses the PML for long message
   transfers).  Currently, all support functions, all communication
   functions (Put, Get, Accumulate), and the Fence synchronization
   function are implemented.  The PWSC active synchronization
   functions and Lock/Unlock passive synchronization functions are
   still not implemented

This commit was SVN r8836.
2006-01-28 15:38:37 +00:00
George Bosilca
0f1c6d79e8 Make the MVAPI BTL thread safe again. The problem was a double locking on the endpoint mutex.
It's still not very clean as we still lock the mvapi_btl mutex inside a critical section
protected by the endpoint mutex ...

This commit was SVN r8810.
2006-01-25 23:14:06 +00:00
Galen Shipman
ddc22d8c7e Use endpoint_lock, not ib_lock (copy paste error from openib btl)
This commit was SVN r8806.
2006-01-25 15:04:37 +00:00
Andrew Friedley
ec995160e6 Checkpoint for switch to mpool work:
- Remove printing of CFLAGS in configure.m4
 - Set MCA_BTL_FLAGS_SEND flag
 - Improved error handling during module initialization
 - Extract the address of each interface with dat_ia_query
 - Start playing around with fragment stuff - probably wrong
 - Misc code cleanup (removal of GM-specific code)

This commit was SVN r8801.
2006-01-25 02:21:34 +00:00
Tim Woodall
51ec050647 port of revised flow control from openib
This commit was SVN r8799.
2006-01-24 23:44:30 +00:00
Tim Woodall
e861158fcd - removed debug code
- removed extraneous memset

This commit was SVN r8798.
2006-01-24 23:38:41 +00:00
Galen Shipman
1e0ea9dd6d Major fixes for the RDMA registration cache (leave_pinned).
This commit fixes issues with HPL runs on node counts > 4. 

This commit was SVN r8793.
2006-01-23 22:51:50 +00:00
Brian Barrett
8b2a285f8f * update reference implementation support to match latest release
This commit was SVN r8783.
2006-01-22 04:56:18 +00:00
Galen Shipman
d657052510 misc cleanup..
This commit was SVN r8731.
2006-01-18 16:20:50 +00:00
Galen Shipman
84a09e4f4e use #if not #ifdef..
This commit was SVN r8720.
2006-01-17 21:07:34 +00:00
Galen Shipman
0c81c0a6ce use ibv_get_device_list if present
(submitted from roland)

This commit was SVN r8712.
2006-01-17 16:23:35 +00:00
Andrew Friedley
5ccab7bcda Checkpoint:
- Move mca_btl_udapl_error/mca_btl_module_init to mca_btl_udapl.c and rename it
 - White space cleanups
 - Free the uDAPL evd and ia handles in mca_btl_udapl_finalize

This commit was SVN r8705.
2006-01-16 21:54:50 +00:00
Andrew Friedley
a4abe3bdbe Checkpoint:
- Borrow configure.m4 from the mvapi btl.  One of the uDAPL headers emits a
   warning when -pedantic is enabled, so strip it out.
 - Change function check in ompi_check_dapl.m4 from dat_ia_open to
   dat_registry_list_providers.. dat_ia_open wasn't working right
 - Make the references to prepare_dst, put, and get NULL for now
 - Add opal_output() calls in all the udapl interface functions for debugging
 - Add evd_qlen component parameter to control event dispatcher queue length
 - First stab at component_init and module_init
 - Misc cleanups - whitespace, dead code removal
 - Update copyrights to 2006

This commit was SVN r8701.
2006-01-16 03:01:12 +00:00
George Bosilca
d4699037f7 Protect an assert if the endpoint cache is not activated.
This commit was SVN r8695.
2006-01-14 21:10:09 +00:00
George Bosilca
3317bf81ad A better implementation for the TCP endpoint cache + few comments.
This commit was SVN r8692.
2006-01-14 20:21:44 +00:00
Tim Woodall
a584c60dbe re-worked flow control logic to take into account the return
of credits from the peer prior to local completion, so that
we don't overrun the number of send wqes available.

This commit was SVN r8683.
2006-01-12 23:42:44 +00:00
Andrew Friedley
c0bad339af - Use the GM BTL as a template instead, per Tim's suggestion
- Begin adding uDAPL-specific stuff
- Added config/ompi_check_udapl.m4 - hopefully I did this right

This commit was SVN r8681.
2006-01-12 04:05:02 +00:00
Tim Woodall
63d0438991 merge in changes from release branch
This commit was SVN r8637.
2006-01-04 16:34:45 +00:00
George Bosilca
479d510eaf Use the common SM component to unmap the shared memory file.
This commit was SVN r8623.
2005-12-31 15:07:48 +00:00
George Bosilca
228f966798 A little trick to force qsort to order the sm modules in the order we expect.
On some systems (like windows) qsort can modify the order of modules when the
comparaison function return 0. As we expect to have the
mca_btl_sm_add_procs_same_base_addr called before mca_btl_sm_add_procs we have
force the qsort to return the SM modules in the correct order. Giving to the
same_addr module a slightly higher priority solve this problem.

This commit was SVN r8620.
2005-12-31 14:59:53 +00:00
Tim Woodall
4ff5316b2d correct copy/paste error
This commit was SVN r8601.
2005-12-22 18:07:46 +00:00
Tim Woodall
5d91c492d6 improve diagnostic error messages
This commit was SVN r8600.
2005-12-22 18:01:47 +00:00
Tim Woodall
e9498f7a75 improve error reporting when registrations fail
This commit was SVN r8598.
2005-12-22 16:05:28 +00:00
Andrew Friedley
f402854a96 Initial commit of uDAPL BTL component.
- Copied the template BTL and renamed everything
 - Compiles and shows up correctly in ompi_info, not tested past that
 - Should be ignored for everyone but me

This commit was SVN r8544.
2005-12-19 16:37:05 +00:00
Brian Barrett
a5af07cd6b fixes suggested by Ralf for supporting both Libtool 1 and 2 in Open MPI...
This commit was SVN r8538.
2005-12-19 03:10:23 +00:00
George Bosilca
1b667067d6 I need to know the number of iovec attached to the fragment.
This commit was SVN r8447.
2005-12-10 23:28:16 +00:00
George Bosilca
e5158142b9 The lb should be extracted from the datatype not from the convertor.
This commit was SVN r8446.
2005-12-10 23:27:20 +00:00
George Bosilca
01b0db91ae Get the lower-bound from the data not from the convertor.
This commit was SVN r8444.
2005-12-10 22:38:25 +00:00
George Bosilca
7baae4f394 Protect the headers and remove the unused ones.
This commit was SVN r8439.
2005-12-10 22:04:28 +00:00
Tim Woodall
1929a97d2f corrections for MPI_BOTTOM
This commit was SVN r8429.
2005-12-09 23:27:55 +00:00
George Bosilca
8888bfb063 And the thread-safe version. The lock/unlock macros are supposed to be
empty for non threaded builds, but somehow just by moving the code a
little bit around and removing 2 call to lock/unlock the latency for TCP
went down by 2 micro-seconds ...

This commit was SVN r8426.
2005-12-09 05:16:50 +00:00
Jeff Squyres
6fbd321442 Fix a bunch of install locations for header files
This commit was SVN r8406.
2005-12-08 00:54:44 +00:00
George Bosilca
5851b55647 Improve the latency for small and medium messages. The idea is to decrease the
number of recv system call by caching the data. Each endpoint has a buffer
(the size is an MCA parameter) that can be use as a cache. Before each receive
operation this buffer is added at the end of the iovec list. All data that are
not expected by the fragment will go in this cache. If the cache contain data
all subsequent receive will just memcpy the data into the BTL buffers.

The only drawback is that we will spin around the receive_handle until all the
cached data is readed by the PML layer. This limitation come from the fact that
the event library is unable to call us if there is no events on the socket.
Therefore we are unable to keep the data in the cache until the next loop
into the progress engine.

This commit was SVN r8398.
2005-12-07 00:12:59 +00:00
Brian Barrett
38391e3406 disable shared receive queue support at compile time if the mvapi implementation
does not support shared receive queues (such as the one shipped by SilverStorm / 
Infinicon for OS X).  Reviewed by Galen.

This commit was SVN r8389.
2005-12-06 15:46:30 +00:00
Tim Woodall
3f396aeae9 fix send to self for large messages
This commit was SVN r8379.
2005-12-05 23:36:33 +00:00
Tim Woodall
8c443832ae add a parameter to limit max number of btls (HCA ports used)
This commit was SVN r8342.
2005-11-30 22:18:21 +00:00
Tim Woodall
5db38b38f5 corrections for latency issue
- don't do additional select until non-blocking read fails 
- don't do an additional read for 0 byte message

This commit was SVN r8312.
2005-11-29 17:33:01 +00:00
George Bosilca
9d990af4a5 Remove 2 useless functions. They have been replaced by the mca_base version few commits ago.
This commit was SVN r8287.
2005-11-28 20:14:23 +00:00
Galen Shipman
55a9fbefd8 fix misc compiler warnings..
This commit was SVN r8263.
2005-11-27 22:53:30 +00:00
George Bosilca
bfa4a40983 Cast it the a well known type to remove a warning.
This commit was SVN r8261.
2005-11-26 21:17:15 +00:00
George Bosilca
b9a739e2b6 Remove 2 useless assignments (they are done at the end before the return).
This commit was SVN r8260.
2005-11-26 21:16:30 +00:00
George Bosilca
00c10a6372 Make the MX BTL startup scalable. When the number of processes involved in the MPI application
increase the previous connection code was broken. It can take as much as 60 seconds to connect
64 processes. Now we do not create the connections when we add the procs but only when we send
them the first message. Now it take only 1.6 seconds to setup a 64 procs MPI job over MX (doing a 2 steps barrier in order to insure that we create all the connections).

This commit was SVN r8252.
2005-11-23 23:48:56 +00:00
George Bosilca
7c73095440 Update the correct sended size.
This commit was SVN r8237.
2005-11-22 21:51:04 +00:00
Brian Barrett
20cea60b82 * fix "make distclean" error in PML
* turns out (duh!) that there was a reason that the <projectdir>dir
  variable was set in the AM conditional.  If not, stupid directories
  are created and not needed...  duh.

This commit was SVN r8205.
2005-11-20 07:41:09 +00:00
Brian Barrett
8faa1884f0 * The last of the build system optimizations. Combine the component and
component/base Makefile.am files, reducing the time configure spends
  stamping out Makefiles at the end
* Install base_impl.h file when devel-headers are being installed

This commit was SVN r8200.
2005-11-20 01:03:01 +00:00
Galen Shipman
eb3ccdb4d8 make compiler happy on false postive warning..
This commit was SVN r8192.
2005-11-18 18:48:11 +00:00
Galen Shipman
4fce90a37b one last warning fixed on 32 bit platforms.
This commit was SVN r8191.
2005-11-18 17:27:09 +00:00
Galen Shipman
635e7a682b fix for 32bit compile warnings.
This commit was SVN r8190.
2005-11-18 17:08:51 +00:00
George Bosilca
bba42f5e49 We are allowed to call mx_set_error_handler before any other MX functions, even before mx_init.
With the errors set to return mx_init will not force the application to exit if there is no MX kernel
module loaded.

This commit was SVN r8184.
2005-11-17 18:47:27 +00:00
Galen Shipman
dde38d4119 reset sg_entry->addr to point at header when sending control messages.
cast to uint64_t (the correct datatype per verbs.h) instead of uintptr_t. 

This commit was SVN r8175.
2005-11-17 05:45:33 +00:00