The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system.
Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed.
Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief.
With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn.
Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put".
This commit was SVN r14711.
1. Remove a unneeded field, bytes_avail, from orte_buffer_t. It is a calcualed value, and updating it everywhere is worse then just calculating it in the one place it is acutally used.
2. Change it so the default size of a orte_buffer is 128 bytes instead of 1024 bytes. We then double the size of the buffer up to 1024 bytes, then we additively increase the size by 1024 bytes at a time as was done before.
This commit was SVN r14252.
This merge adds Checkpoint/Restart support to Open MPI. The initial
frameworks and components support a LAM/MPI-like implementation.
This commit follows the risk assessment presented to the Open MPI core
development group on Feb. 22, 2007.
This commit closes trac:158
More details to follow.
This commit was SVN r14051.
The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
r13912
The following Trac tickets were found above:
Ticket 158 --> https://svn.open-mpi.org/trac/ompi/ticket/158
and George on these refinements):
* Rename the static OBJ initializer macro to be
OPAL_OBJ_STATIC_INIT(class)
* Ensure that all static OBJ initializations get a refcount of 1
(doesn't ''really'' matter, since they're static, it should never
get to the point where the OBJ is DESTRUCTed, but more correct
nonetheless)
* Add a "magic number" to the OBJ when compiling with debug support.
The magic number does some rudimentary support to ensure that
you're operating on a valid OBJ (and fails an assertion if you're
not). Check to ensure that the memory contains the magic number
when performing actions of OBJ's. Also remove the magic number
when DESTRUCTing OBJs, so that if, for example, an OBJ is
DESTRUCTed more than once, we'll fail the magic number assert.
This commit was SVN r13338.
The following SVN revision numbers were found above:
r13227 --> open-mpi/ompi@96030de97b
r13228 --> open-mpi/ompi@c2e9075d29
* When using the load/unload interface, stash away the current buffer
type so that it can be properly unpacked on the receiving side if
the buffer type is other than the receiver default
* Include type information for unsized types (bool, int, size_t,
pid_t) so that they can be properly unpacked by the receiver
in the heterogeneous case.
* Restore the NON_DESC type as the default for optimized builds,
since it looks like this fixes the known issues with the
non-described buffers
Refs trac:587
This commit was SVN r12784.
The following Trac tickets were found above:
Ticket 587 --> https://svn.open-mpi.org/trac/ompi/ticket/587
Also, change the dss buffer type mca param to something more easily remembered (it is now "dss_buffer_type"). Heck, even I had to keep looking at the darn code to remember it.
This commit was SVN r12728.
1. implement and enable the non-described buffer operations. I will send out a more detailed explanation separately. However, this mode of operation (which is now the default) significantly reduces message size during startup. If you want the described buffers, set the mca param "-mca dss_describe_buffer 1".
2. revise the xcast system to support both linear and binomial tree broadcast methods. Since we are seeing scenarios where the binomiall tree can cause problems, I have made the linear method the default. To run with the binomial tree, set the mca param "-mca oob_xcast_mode binomial".
3. add some detailed timing reports to the xcast operation. These are enabled via "-mca oob_xcast_timing 1".
4. add some more unit tests for the dss and gpr (focused on support for the non-described buffer)
This commit was SVN r12722.
Accordingly, there are new APIs to the name service to support the ability to get a job's parent, root, immediate children, and all its descendants. In addition, the terminate_job, terminate_orted, and signal_job APIs for the PLS have been modified to accept attributes that define the extent of their actions. For example, doing a "terminate_job" with an attribute of ORTE_NS_INCLUDE_DESCENDANTS will terminate the given jobid AND all jobs that descended from it.
I have tested this capability on a MacBook under rsh, Odin under SLURM, and LANL's Flash (bproc). It worked successfully on non-MPI jobs (both simple and including a spawn), and MPI jobs (again, both simple and with a spawn).
This commit was SVN r12597.
- use the OPAL functions for PATH and environment variables
- make all headers C++ friendly
- no unamed structures
- no implicit cast.
Plus a full implementation for the orte_wait functions.
This commit was SVN r11347.
different macros, one for each project. Therefore, now we have OPAL_DECLSPEC,
ORTE_DECLSPEC and OMPI_DECLSPEC. Please use them based on the sub-project.
This commit was SVN r11270.
Clean up the remainder of the size_t references in the runtime itself. Convert to orte_std_cntr_t wherever it makes sense (only avoid those places where the actual memory size is referenced).
Remove the obsolete oob barrier function (we actually obsoleted it a long time ago - just never bothered to clean it up).
I have done my best to go through all the components and catch everything, even if I couldn't test compile them since I wasn't on that type of system. Still, I cannot guarantee that problems won't show up when you test this on specific systems. Usually, these will just show as "warning: comparison between signed and unsigned" notes which are easily fixed (just change a size_t to orte_std_cntr_t).
In some places, people didn't use size_t, but instead used some other variant (e.g., I found several places with uint32_t). I tried to catch all of them, but...
Once we get all the instances caught and fixed, this should once and for all resolve many of the heterogeneity problems.
This commit was SVN r11204.
1. Introduces a flag for the type of buffer that now allows a user to either have a fully described or a completely non-described buffer. In the latter case, no data type descriptions are included in the buffer. This obviously limits what we can do for debugging purposes, but the intent here was to provide an optimized communications capability for those wanting it.
Note that individual buffers can be designated for either type using the orte_dss.set_buffer_type command. In other words, the buffer type can be set dynamically - it isn't a configuration setting at all. The type will default to fully described. A buffer MUST be empty to set its type - this is checked by the set_buffer_type command, and you will receive an error if you violate that rule.
IMPORTANT NOTE: ORTE 1.x actually will NOT work with non-described buffers. This capability should therefore NOT be used until we tell you it is okay. For now, it is here simply so we can begin bringing over parts of ORTE 2.0. The problem is that ORTE 1.x depends upon the transmission of non-hard-cast data types such as size_t. These "soft" types currently utilize a "peek" function to see their actual type in the buffer - obviously, without description, the system has no idea how to unpack these "soft" types. We will deal with this later - for now, please don't use the non-described buffer option.
2. Introduces the orte_std_cntr_t type. This will become the replacement for the size_t's used throughout ORTE 1.x. At the moment, it is actually typedef'd to size_t for backward compatibility.
3. Introduces the orte_dss.arith API that supports arbitrary arithmetic functions on numeric data types. Calling the function with any other data type will generate an error.
This commit was SVN r11075.
- The constant 1 is a signed int by default. Explicitly say that
it is an unsigned value so we can't overflow
- Fix unreachable statement warnings in dss_arith by breaking out
of switch statements instead of returning - this should have
no impact on performance, since it's a non-conditional jump
- A couple of the GPR files had carriage returns and were in
DOS mode - put them in unix mode...
These should all probably go to the v1.1 branch...
This commit was SVN r9664.
for the type to be packed / unpacked when dealing with sized types (like
size_t) so that the dss_unpack code to deal with types of different sizes is
activated. Necessary for proper 32/64 interoperability.
This commit was SVN r9475.
* cleanup the unpack_size_mismatch macros a little bit
* ad comment about endianness of size_mismatch cleanup code so that I don't
think I've found a bug that really isn't and lose an hour tracking it down
again...
This commit was SVN r9458.
to match the receiving process's setup. sizeof(bool) is different on
i386 OS X and PowerPC OS X, so need this to do endian testing between
the two
This commit was SVN r9140.
- 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.
