CVS log for pvfs2/src/io/flow/flow.c
	Help

Request diff between arbitrary revisions / Display revisions graphically

merge with Orange-Branch

Committed Windows client code to Orange-Branch.

Porting flow etc.

Reverse merged and ported to HEAD.

Get rid of gen_mutex_build.

Merge HEAD into Walt's branch.
Rework new state machines to the new cleanups introduced by Walt.

PVFS2-hint support + minor changes (verbose request scheduler)

Added request statistics to scheduler, reference from datafiles to
parent metafile (needed for migration) and auto migration tool(unfinished).

Reverse merges from HEAD..

reverse merge from trunk.  working for now.

[phil]: flow-proto-tuning: This patch adds "FlowBufferSizeBytes" and "FlowBuffersPerFlow" options to the configuration file format.  They allow you to specify the buffer size that the default flow protocol will use as well as the maximum number of buffers to use per flow.  Note that if you change either of these parameters, then you need to remount any active clients so that they pick up the configuration change before performing any I/O.

[phil]: max-aio: This patch adds "TroveMaxConcurrentIO" to the configuration file format.  It allows you to specify the maximum number of I/O operations that trove will allow to proceed concurrently (currently 16).  Note from the previous email regarding AIO that depending on your access pattern, AIO may queue all of your operations anyway regardless of this setting.  It probably doesn't have much effect unless you are accessing more than one file at a time, or if you are using an alternative to the stock AIO implementation.

Reverse merge changes from HEAD to posix-extensions-branch.

revert style changes back to previous versions.

run maint/pvfs2indent-80col.sh on all .c files to correct style :)

pull externs out of function to quiet warnings

Reverse merge from trunk to lockd-branch...

reverse merge of branch

reverse merge of trunk to branch

return error if any flow proto failed to finalize

merging small I/O changes to trunk.  The majority of this commit includes the
new small I/O state machines on client and server, and the modifications to
the client IO state machine.  It also includes mods to the zero-fill code
to perform zero-fill of holes only if they exist instead of zero-ing the entire
memory region initially.  Also, some debug stuff has been added to the
state machine processing code to print enter/exit of state machines if sm
debug is enabled.  The genconfig now includes ioports/metaports for running
multiple servers on the same host for testing.

fixed a bug in flow caused by outsize being updated correctly in trove now.  Also, s/transfered/transferred/.

mutex can be NULL so assert doesn't work here.

Applying Avery's memleak patch for flow descriptors in sys-io.sm.  Modified slightly to fix invalid free for flow descriptors.  Also made internal function in client core static to remove warnings.

updates to my event changes to bring them inline with trunk

merged acache changes to trunk.

removed refs to flowproto-bmi-trove stuff

Updates for standardized spelling in pint-request.h and new distribution
function signatures.

don't include pthread.h extraneously.  don't compile pthread versions if
NULL_LCOKING selected

- fixed possible flow mutex memory leak (ptr set to null before free
  check was done)
- try to keep the existing mutex (if any) throughout the flow_reset
- make pass over kernel code making sure that out link counts are
  correct again
- fixed directory inode mtime and ctime changes on updates
- misc cleanups

- added configuration option TroveSyncMeta (can be 'yes' or 'no')
- added configuration option TroveSyncData (can be 'yes' or 'no')
- replaced TroveSyncMode configuration option with separate
  TroveSyncMeta and TroveSyncData options.  the defaults are yes to
  both.  this is a change because previously data was NOT being sync'd
  after reads/writes.  for now, only writes are sync'd if
  TroveSyncData is 'yes'.
- updated pvfs2-genconfig to reflect config option changes
- added top-level and flowproto-multiqueue understanding of the
  data sync mode option via setinfo
- added hook to server to pass data sync mode option to the flow
  implementation via setinfo on a per fs/coll_id basis
- added fp_multiqueue tracking of this to be able to pass the
  appropriate data sync flag to trove on a per fs basis
- misc cleanups

- applied Nathan's expandtab patch (expandtab-2.patch.gz)

  first referenced at:
  http://www.beowulf-underground.org/pipermail/pvfs2-developers/2004-July/000745.html

- merging in the pvfs2-nm-nb-branch with the main tree
  see ChangeLog for details, or browse the cvs history of the branch
  for full details

NOTE THAT THIS IS A BRANCH COMMIT (tagged as nm-nb-branch).  Feel free
to ignore it completey as it's a snapshot of a work in progress and it
will crash your computer and reformat your hard drive.

This is almost the initial draft of the pvfs2-client-core that
operates in a non-blocking manner.  While it runs, there are still
issues that need to be resolved (that are keeping it out of the main
CVS trunk).  Many other changes were made along the way, so it's more
than just that.

- added compile time option for disabling thread-safety in the client
  library (enabled by default; --disable-thread-safety to disable)
- improved configure summary information emitted at configure time
- added missing non-blocking sysint declarations to sysint header
- re-wrote pvfs2-client-core to use sysint non-blocking operations
  where possible
- made sysint test and testsome() calls more useable from a user point
  of view
- merged dev unexp polling/handling with system interface
- added PINT_sys_dev_unexp call that allows posting unexpected device
  messages so that they can be returned from the sysint testsome
  method in addition to completed sysint operations
- many memory leaks fixed -- many more to go (started adding macros
  for freeing the server response objects)
- added a id_gen_fast_unregister macro that is a no op, to make the
  api more consistent with the id_gen_safe_* calls
- server-config-mgr: report mutex still in use if it is, but also make
  sure not to unlock an already unlocked mutex (valgrind complains)
- many assertions added
- many formatting changes
- many ptr assignments to NULL after freeing in the job interface;
  done while tracking down a problem
- modified the pint-dev device interface to make sure it can handle
  the pvfs2 device in a non-blocking manner
- pint-dev was using buffers larger than it needed across the device;
  fixed them to be the right size
- added a method to free the memory region mapped into the kernel
  through the device
- freed that mapped memory region on pvfs2-client-core shutdown
  (valgrind complained)
- modified device driver to work properly in non-blocking mode from
  userspace by implementing the character device poll method
- modified pint-dev test function to properly handle non-blocking
  responses from the device driver
- modified PINT_flow_reset to not allocate a new mutex unless the old
  one was destroyed
- modified all job uses of the id-generator to use the safe, rather
  than fast, methods (useful for several reasons including safe
  cancellation)
- modified cancellation methods to be able to handle ops that have
  already completed
- modified the job_dev_unexp method to have (and honor) the no
  immediate completion flag if passed (used in the pvfs2-client-core)
- modified the job completion callbacks to make sure to NOT add a
  completed job desc to the completion queue if it's already been
  added (by checking a flag, not scanning).  this is a safety and
  should only be used when a non-thread-safe client library is being
  used in a thread safe env, but we should handle it gracefully anyway
- replaced all kernel allocations of ops through the slab allocator
  handled op_cache to be replaced by a wrapper method (op_alloc(), as
  suggested by Murali) -- the other cache allocations will probably be
  replaced later.  this allows the removal of the extern op_cache
  declaration
- added macros for freeing some of the most heavily used server
  response messages (readdir, lookup, getattr) -- the others are
  coming later
- freed dirents coming out of sysint response object in the
  pvfs2-client

implemented modification of timeout system for flows: rather than use an
absolute timeout for completion of flows, instead check every "timeout_sec"
seconds to see if progress has been made on it.  Will only cancel if
timeout_sec seconds passes with no data being transfered.  Needs testing
before being turned on

use initialized variable in error message

- consolidate the string split functions in flow/bmi by adding a new
  str util function that does what they did
- some cleanups

give the flow interface the ability to initialize all compiled in protocols
without listing them explicitly

started stubbing in hooks for cancelling previously posted flows

- make flow_mutex a pointer to play nicely with gen_mutex_destroy

exposed a lock used in the flow protocol at a slightly higher level, need it
for some future changes

minor update, changed name of PINT_flow_unpost() -> PINT_flow_cancel(),
still just a stub function

renamed some flowproto-bmi-cache functions and global variables, and made
more stuff static in both bmi-cache and multiqueue; necessary to avoid name
space collisions between the two protocols

big structural change/simplification of the flow code.  job api remains
unchanged.  notification from flow level back to job level is now done
through callbacks rather than by testing.  works for both threaded and
non threaded code path.

print out a helpful error message if the client and server flow protocol
settings don't match

reorganization of the multiqueue code, much more stable now due to simpler
locking model, bug fixes, and more thorough state tracking; added temporary
kludge to flow.c as this protocol cannot be allowed to immediately complete
yet.

stubs for a new flow protocol

changed some types over to named enumerations

renamed some flow protocol functions- still not all that great, but maybe a
little clearer

removed some unused and unimplemented flow functions

compile in support for bmi cache flowprotocol

got rid of FLOWPROTO_ANY and replaced with FLOWPROTO_DEFAULT, no longer
bother searching flow protocols based on endpoint type

added asserts in some places where i've hit null pointers

integrated some request processing updates from walt, mostly to make
memory datatypes work better.  Updated I/O path to expect memory datatypes
and updated all PVFS_sys_read and PVFS_sys_write calls accordingly.
File datatype tiling still isn't quite right, so for now the file datatype
must be large enough to cover whatever it is you are trying to access
in one shot.  Haven't really tested any memory datatypes more
complicated than contiguous with various file datatype offsets- need to
try some fancier stuff...

Put remaining code in place to handle memory datatypes, not ready yet
(pending a fix to some corner case in req processing).  Everything should
still work as it was prior to this commit.  When we do enable memory
datatypes, I'll need to go back and uncomment some code and adjust all
PVFS_sys_{read/write} calls to pass in the right datatype (they are just
taking NULL for now).  PVFS_sys_{read/write} API is in its final form
(I think...) for now, at least.  For now this code is dependent upon
the fact that the file datatype isn't implicitly tiled yet, need to get
memory datatypes going before relaxing that constraint...

changed reference to datatypes called "io_req..." to "file_req...", to make
the distinction clearer between file datatypes and memory datatypes once the
latter is added

Aieeee....

New datatype request processing interface clumsily merged into tree.  For the
most part, things will work as before, but still need to:
- fix bug in precalculation when datatype offset is used
- fix data corruption when running multiple servers
- update flowproto-dump-offsets and enable again
- take out old #ifdef0'd code in bmi trove flowproto
- clean up flowproto's use of new interface

got rid of the PVFS_bitfield type, and replaced with something more
generic everywhere it was used

Anemic memory allocation; caused segfault when efence was involved

put in hooks to manually specify which flowprotocol to use, tested for
a few trial cases

now compile in debugging flow protocol by default.  No longer compile
in template flow protocol.

added the ability to force selection of a specific flow protocol;
untested

implemented contexts

Added contexts to the flow API.  Not doing anything with it yet.

eliminated queue search from test() function; now just check state
directly in descriptor.  removed associated support code

reworked testsome to check flow descriptor states directly, rather than
searching a completion queue; removed old support code

reformatted all of the flow code

made all test functions return 1 if anything completes

fixed some memory leaks

fixed the organization of the gen-locks.h header file so that it
compiles cleanly in more places, fixed several warnings in the
flow code that I somehow missed last week

finished pushing max_idle_time semantics into flow interface

big changes to various I/O API's- there are no longer any "wait"
functions in the BMI, flow, or request scheduling APIs.  In the
request scheduler case, they have simply been discarded.  In BMI
and flows, the testXXX() functions have been augmented with a
timeout argument so that they may perform the role of the
former waitXXX() functions.

added in the first flow test program and modified to work with latest io
description api; made several minor changes to the flow code along the
way, added in a template flowprotocol (renamed from default flow
protocol in old tree)

integrated the flow subsystem into the tree; still need to get matching
test programs in