mpi-external Mailing List Archive by subject
Starting: Wed 24 May 1995 - 14:19:55 PDT
Ending: Tue 24 Jun 1997 - 14:37:13 PDT
Messages: 460
- "inetd" example
- 'Missing' in 'Accessing MPI Datatypes'
- (corrected) Simplified version of generalized requests
- (no subject)
- 1st class datatypes (was: use of MPI_Status and MPI_Status)
- 3rd party datatype instantiation
- 3rd party datatype instantiation (was: mpi-io, mpi-dynamic generalizations)
- [Richard Frost: mpi-io, mpi-dynamic generalizations]
- Alternative thread initialization scheme
- ANSI C
- Attribute caching fuzzyness
- C++ class hierarchy for handles ?
- caching and naming other handles
- Caching file related information using info or hints?
- caching on file handles
- caching on MPI handles
- callbacks
- Cancelling (yuk) generalized requests...
- canonical data representations
- canonical section removed
- changes at Forum meeting
- Chnages to Chapter 7 (May 8 draft)
- clarification requested
- comments
- comments of Steve
- Comments on Chapter 7 (External Interfaces)
- comments on datatype encode/decode
- comments on Marc's proposal
- comments on Marc's threads section
- comments on new packing proposals
- comments on: Proposal for User Access to an MPI Data type
- Communicator IDs & Dynamic Processes
- Completion of nonblocking operations
- Datatype access
- datatype accessor functions
- datatype decoding
- datatype decoding & GR return codes
- deadlines and "XDR-like" functionalities
- error classes and codes -- function names
- Error handlers for files
- Event Callbacks Proposal
- Example use of MPI_Type_compact
- Exit code (aka call-backs in finalize)
- Explicit Job Scheduler Interface
- ext. chapter
- External (formerly Canonical) Proposal
- External changes
- external chapter
- external data representation
- external data representation - DIGITAL Unix on Alpha
- extra-state of attribute keys
- final draft
- freeing error code, classes, strings
- fundamental vs basic datatype
- generalized request
- Generalized requests
- generalized requests -- cancel
- generalized requests -- get_status and request_free
- generalized requests -- postscript
- generalized requests changes
- generalized requests postscript
- generalized requests without communication handlers
- Getting information out of a request object
- GR cancel_fn return value
- GR proposal
- GRs and request completion
- Handlers & Request completion
- Hanlders
- information for profiling
- Is non-blocking collective I/O desirable?
- lastest version of chapter
- latest chapter
- length in string
- low level datatype accessing
- low level MPI data type accessors ("merged" TMC/NEC proposal)
- Marc's new GR section
- more changes
- More clarification
- More on GRs
- More requirements from external for I/O layerability
- MPI and Threads
- MPI Data type accessors -- proposal
- mpi datatype access revisited (was: MPI and MPI-IO)
- mpi on a single thread
- MPI thread-compliant and malloc()...
- MPI-2 proposal to pack datatypes and communicators
- MPI-IO requirements/ portable datatypes
- mpi-io, mpi-dynamic generalizations
- mpi-io, mpi-dynamic generalizations]
- mpi_get_char_data_type implementation?
- MPI_Name_get twice specified
- MPI_Name_put clarification
- MPI_Set_request_{tag,source,error,count}
- MPI_STATUS_IGNORE implications
- MPI_TRUE_EXTENT
- MPI_TYPE_DUP
- MPI_TYPE_DUP (Actually, what I really wanted)
- MPI_Type_get_contents
- MPI_TYPE_LB
- My apology for too succint previous message (mailers...)
- My suggested answer
- Naming proposal
- new chapter
- new chapter available
- new ei chapter
- New errhandler text
- new external chapter
- new external chapter available
- new external chapter done
- new external chapter ready
- new external interfaces draft
- new gen req proposal
- new generalized request proposal
- new GR proposal(s)
- new partial version of chapter
- New request access function
- new threads initialization section
- new version of chapter
- new version of chapter ready
- new version of thread section
- new version of threads section
- new version out
- new version(s) of generalized requests
- non-blocking collective I/O
- notes from meeting
- portable mpi debuggers
- print_mpi_datatype (user access to mpi data types)
- progress with GR and handlers
- Proposal for communicator id
- Proposal for datatype access and instantiation suite
- Proposal for Datatype and Request Caching
- Proposal for Datatype Compacting
- proposal for fns. to add new error codes and classes
- PROPOSAL FOR INTERLANGUAGE ACCESS TO HANDLES
- PROPOSAL FOR USER ACCESS TO AN MPI DATA TYPE
- proposal of Thakur
- proposed text for complete and MPI_REQUEST_FREE case for GR
- ps file of proposal
- question on threads proposal
- question that need answers asap
- questions of Dick
- RE: Proposal for communicator id
- RE: 3rd party datatype instantiation
- RE: information for profiling
- RE: low level datatype accessing
- RE: MPI and Threads
- RE: mpi on a single thread
- RE: MPI_TYPE_DUP
- RE: My apology for too succint previous message (mailers...)
- RE: notes from meeting
- RE: Proposal for datatype access and instantiation suite
- Request Handler return codes
- Revised generalized cacheing proposal
- Simplified version of generalized requests
- summary, changes to chapter - resend
- tabled issues for generalized requests
- test
- test of mpi-external archive
- Thread cancellation points
- Thread query/init
- Threads and I/O calls
- threads and libc
- Threads and MPI
- threads and signals
- Threads yet again...
- two-face collective scenario
- type-simple-struct
- use of MPI_Status and MPI_Status
- user access to mpi data types
- Welcome to mpi-external
- workaround for communicator id
- Yet another clarification request
Last message date: Tue 24 Jun 1997 - 14:37:13 PDT
Archived on: Tue Jun 24 1997 - 14:37:23 PDT
This archive was generated by hypermail 1.02.