CVS log for pvfs2/Makefile.in
	Help

Request diff between arbitrary revisions / Display revisions graphically

Merged robust security code into Orange-Branch.

actual foxes for ucached (in Makefile.in, etc)

Added include file for pvfs_xxx user interface
Added script to build tools with PVFS support
Added dependency for libofs.so to Makefile
Modified stat and mknod to pass correct version
Now builds both static and shared libs on 32 and 64
systems.

initial working versionof usrint code
auto configures usrint on an ucache off (not working yet)
new acl repimplemented - old one has #def
some minor format edits

clean up existing gzip manpages in install location before gzipping new

gzip manpages to be nice

added support for usrint and usrint-cache to build system

merge with Orange-Branch

small build change for libuuid and printing of temp change for handle printing in pvfs2-db-display

add target for installing just 2.4 kernel apps to prevent combined path of destdir and prefix

define _GNU_SOURCE in CFLAGS in place of __USE_GNU throughout the code. Using the internal glibc macro of __USE_GNU isn't correct and in the case of newer platforms was the cause of backtrace functionality not building correct.

merge with Orange-Branch

merge with Orange-Branch

fix silly mistake assigning version to wrong variable

Change of utsrelease.h location to include/generated/

remove db-4.8 from develtools target

remove db-4.8 from server libs

JNI patch applied.

corrected library name

temporarily added libdb-4.8

temporarily added libdb-4.8

Merged in changes from Orange-Branch. There were a number of bugs fixed there since this branch was created.

merge with Orange-Branch

Update Makefile.in to check for kernel version in include/linux/utsrelease.h as suggested by Michael Stroucken

Update Makefile.in to check for kernel version in include/linux/utsrelease.h as suggested by Michael Stroucken

add -lm to LIBS in addition to SERVERLIBS

incorporate dist-dir-struct to the source tree. It's compilable but not runnable. still have a lot to change and add. Commit as a touch base.

merging Orange Branch changes in

initial merge with Orange-Branch. much will be broken

tweaked OpenSSL support in build process

added stubbed security functions so security can be disabled

fixed credential generation bugs. changed default credential key path.

brought create credential utility functions up to date

revert cu-security-branch to before the attempted merge with Orange-Branch

Addition of src/apps/devel directory and pvfs2-display-db to give a dump of berkeley dbs in text form to assist in development/debugging. Also made configure and make changes with new build target develtools

merge of B2O-Blue-Sync,HEAD,and latest of Orange.

Added --with-cuda and --with-cuda-sdk optional compilation to the configure script.

commiting changes from merge of Orange and Blue into test branch.

Added a conditional compilation check for enabling/disabling CUDA, --with-cuda=<dir>
Modified the kmeans.sm to make it compile depending on --with-cuda option.

The remaining pieces of BMI ZOID support.

Added reduction kernels in CUDA (from NWU) into source tree.
Revised test programs.

*** empty log message ***

*** empty log message ***

Merged NWU's CUDA code.
Modified configure script and Makefile to enable CUDA compilation.

merged in changes from latest Orange-Branch

merged in changes from summer at LANL

Initial import of branch supporting keyval attribute/value lookup

Merged Phil's tcp-peer-comm-branch that allows simultaneous server to server communications.
Added basic allreduce module:
  - implemented a recursive distance doubling algorithm.
  - works only when the # of servers are power of 2.
  - when activated, servers simply exchange their id to each other.

TODO
  - handle arbitrary sizes of buffers of various data types.

Setting up Orange Branch

initial support for solaris.  doesn't include /dev/epoll support.  See trac ticket #97.

fix make dist problem

merging another fix from trunk

fix the world's most annoying symlink (again), this time with help from
Dries Kimpe

merging over bug fix from trunk

Committing fix from Kevin Harms for pvfs2-lsplus symlink path so that rpm
builds work correctly.  Problem was also reported by Tony Kew with a similar
fix.

merge changes from 2.8 branch to HEAD.

fix weird dependency between cleaner and saveconfig.

merging BUILD_KERNEL updates and sys-io.sm logging updates to 2-8 branch

updates to BUILD_KERNEL flag: move to positive section of --with-kernel
test in configure.in, add to --with-kernel24 test, and add Makefile.in
flag to pick up at AC_SUBST time.

merge of Sumit's pvfs2fuse code to HEAD.

merge pvfs2-lsplus functionality and delete pvfs2-lsplus again

applied bmidist patch from Dries.  Also added a number of cleanups of my own.

Compatibility header from Dries Kimpe for system inteface changes.  Thanks Dries!

Make pvfs2.h include the version numbers to be exported.  This allows consumers of the sysint calls to determine which parameters to include.

merging hints/events code into trunk.

bmi changes to allow separate bmi packaging

silence configure warning about ignored datarootdir setting

reverse merge of direct IO changes to he-branch.

add tau formatting code from Aroon.

merge of direct IO code to PVFS HEAD.

cleanup

replace pvfs2-ls with pvfs2-lsplus; install symlink in case anyone has
scripts etc. that use pvfs2-lsplus explicitly

merge trunk updates down to small-file-branch.  Passes basic tests but needs
some double checking of pint-cached-config and sys-create conflicts.

Added missing FUSE changes.

adding threaded odirect impl

db-rep-send state machine added

Reverse merged and ported to HEAD.

Added the heap path.

reverse merge of trunk changes since branch.

Committed patch from Tony Kew (tonykew@ccr.buffalo.edu) to install the
shared libraries if you use the --enable-threaded-kmod-helper option

reverse merge latest trunk changes to small-file-branch and update msgpair
usage where needed

reverse merge of latest changes from HEAD to he branch.

Updated branch with code from HEAD

updated build system for security module

merge make dist changes

allow make dist to work with config args specified.

merge make dist tweaks to 2.7  branch

syncing small-file-branch back up with trunk at small-file-branch-point2 tag
(reverse merge)

sync up make dist from trunk to pvfs-2-7-branch

some tweaks to make dist process:
- don't include .d files (this fixes hung make bug reports)
- make dist now fails if any build steps fail (such as making docs)
- distclean no longer removes pregenerated docs in release tree
- make dist now relies on configure and distclean being run in constructed
  (tmp) directory rather than local directory.  This allows it to produce
  same results whether make dist is invoked from in tree or out of tree build.

library versioning

This patch adds a version code to the installed shared libraries,
libpvfs2.so and possibly libpvfs2-threaded.so.  Library versioning
is considered a good thing, in that apps can know what version of the
library they need.  Use the actual PVFS version numbers to build
the library version string.  This is not always the right thing to do,
but works for us because we pay attention to the meaning of "major",
"minor" and "release" changes.  Adapted from a patch by Justin Bronder.

destdir

Use variables like prefix that come from the configure process as given;
they should not have DESTDIR glued on the front.  In the install
process, however, pay attention to DESTDIR when putting all the files
where they go.  This works well with package scripts that may set DESTDIR
only in the install phase.  Adapted from a patch by Justin Bronder.

shared lib deps

For proper symbol versioning, at shared library build time, it is necessary
to specify all the other shared libraries that will be required to run
the one we are about to create.  The particular place this shows up as a
problem is for libibverbs.so when building bmi ib.  If you don't tell the
linker that you'll need libibverbs.so later, at runtime, it will pickup
the 1.0 rather than the 1.1 symbols.

Big pile-o' changes to make Pete go mad.

The main changes are the implementation of the stuffed create path (now just the create state machine and operation), and the beginnings of the "unstuff" operation.

This also includes the request scheduler and state machine cleanups (required for using prelude with batch_remove) and whatever else I may have happened to change along the way.

This code is able to create files, but writing to them is almost guaranteed to break.

threaded kmod helper

Add --enable-threaded-kmod-helper option to build a threaded
pvfs2-client-core, that is, an executable linked against
pvfs2-threaded.so.  Take care to make sure only what libraries
are needed based on --enable-static, --enable-shared, and this
new option.

Remove the "--threaded" argument from pvfs2-client.  Whatever version
of client-core that was configured is all that is available at runtime.
It will always be installed as pvfs2-client-core.

Sam's prelude and scheduler updates

gm-mx-autoconf

Convert GM and MX to use autoconf tests to find includes and libs.  Move
these tests into new config files to keep the top-level Makefile clean.
This fixes a configuration problem with GM on x86_64 where libs are
in lib64/, and generalizes both GM and MX to be a bit more clever at
autodetection.

Signed-off-by: Pete Wyckoff <pw@osc.edu>
---
 Makefile.in                    |   32 +++-----
 configure                      |  167 ++++++++++++++++++++++++++++++++--------
 configure.in                   |   49 ++----------
 maint/config/gm.m4             |   72 +++++++++++++++++
 maint/config/mx.m4             |   73 +++++++++++++++++
 src/apps/admin/pvfs2-config.in |    8 +-
 src/io/bmi/bmi_gm/module.mk.in |   45 ++++++++----
 src/io/bmi/bmi_mx/module.mk.in |   40 +++++++---
 8 files changed, 363 insertions(+), 123 deletions(-)
 create mode 100644 maint/config/gm.m4
 create mode 100644 maint/config/mx.m4

The normal path for a server request operation is to jump to the prelude state machine, which tries schedule the request based on the handle+fsid.  This meant that the request scheduler code had a switch/case returning the handle+fsid for each request we defined.  The function was also overloaded with a readonly parameter (whether the request modified any objects on the server), and a scheduling policy parameter (as the return code), which specified whether the request scheduling could be bypassed for this request.

I moved all that to a per-request parameter structure, that gets specified within the server state machine for the request.  Now there is one array of server requests (moved to pvfs2-server-req.c) which the request scheduler and prelude state machine can poke at to get parameters (defined in the state machine for that request), and based decisions on.

The patch includes some other minor changes:

* A fix to the BMI addr ref count increment/decrement.  Previously, the addr ref count was being incremented/decremented in the prelude/final-response state machines.  That requires that those state machines get invoked, which I don't want to have to depend on.  I've moved the INC call to the unexpected_sm, and the DEC call to where the state machine gets finished in the server.  That way we don't have to worry about whether we're using prelude/final-response.

* Cleanup of the request scheduler.  The way we were changing the mode from normal to admin was sort of embedded into the request scheduling code, so we had to hack the mode change from prelude_sm instead of in setparam.  Also, the request scheduler had to be aware of the setparam fields in the request.  I've moved the mode change code to separate request scheduler and job calls.  I think something similar could be done for the timer code (it doesn't really belong in the request scheduler), but I'll leave that for another patch.

* Changed the way we compile the server binary in Makefile.in to allow for multiple object files.

once and for all.  Really.

update to Makefile for htmlfixup

This seems to be the rule for both in-tree and out-of-tree html versions

doc changes, changes to script that fixes the latex2html gen-ed docs.

commit of Murali's unifying config file patch.  Also adding a convert script for converting to the new single fs.conf format.

Keep bmi-tcp epoll issue out of global cflags.

Move gcc options for --enable-strict from configure to Makefile, so they
are near the other build flags and so non-GCC compilers can be handled
differently.  Also redefine --enable-strict to apply only warning options,
not to turn on -g too.  And the option works with CFLAGS= override on the
command line too, unlike --enable-fast.  Finally, set default CFLAGS to
"-g -O2", adding the -g in hopes we will get better tracebacks on errors.
This will increase code size somewhat.  Builds that use --enable-fast or
custom CFLAGS are not affected.

Redo configure args to allow different choices of Portals libs depending
on the local machine.  Some hacking around incompatibilities between
Cray Portals and the reference TCP implemenation.

uh-oh, no more tcp.  Looks like this got removed on portals commit?

Do not build threaded client libs unless something asks for it, like kmod build.

Add no portals libs unless requested.

Propagate separate build cflags and ldflags through configure.

Make BMI TCP optional, but configure it on by default for old times sake.

Initial addition of portals BMI method.

Remove both client-core versions on clean.

Another swat at ctags.  This time getting it right.

mx fixes

Adding .c files for .sm in tags generation was just a bad idea; revert.  Will fix later for out of tree builds.

Merge HEAD changes to TAS-branch.

oops, too much commit, revert

typo

help out packagers (rpm, deb) and honor DESTDIR env var

walt3 reverse merge from head includes merges of pw's sm changes (no state declarations), cleanup of state machine code, and other general merging/fixes.

fixes for darwin

update to current CVS version because statecomp modifications are neat for my work, thanks!

Build statecomp from top-level makefile.

Update migration branch to current CVS version

Sync hint-branch to current CVS version

Fix my previous tags generation fix so that it does the right thing on out-of-tree builds too.

Initial commit of BMI MX implementation provided by Scott and Myricom.  Thanks Scott!

looks like linewrap bit this one

tags: ignore .c files that were auto-generated from state machine .sm files

merge a bunch of minor fixes pete made to head recently.  See the changelogs from Dec 15th to Jan 15th for more info (esp. Dec 19th).  Also updating the changelog and minor release version number for impending release (2.6.2).

Synchronization with HEAD

pay attention to CPPFLAGS: include paths and such as produced by configure or users

clean threaded libraries too; teensy ctags addition

merge pw's fixes to the 2.6 branch.

add the LIBRARIES_THREADED dep to the KERNAPPSTHR target for the 2.6 branch as well.

Upgrade to current CVS version

add device libs to threaded library

added KERNAPPSTHR target

* merge make clean bugfix for removing -threaded.o
* merge Murali's immutable file kernel module support

fix clean to remove -threaded.o

reverse merge of HEAD to WALT3 branch.

add comms libs to new libs_threaded make variable

-threaded option to client and changed Makefile to build a pvfs2-client-core-threaded version.

correct path to new tarball name

remove corrected tarball name

Added itree and rbtree headers.

get CFLAGS before SERVERCFLAGS

add -I$(srcdir)/include to front of CFLAGS

backmerging of HEAD to branch...

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

add new variable to make autoconf 2.60 happy, unused

fix for out-of-tree builds, spotted by Pete..

As per Sam's suggestions, change the LDFLAGS in top-level makefile to have @SRC_ABSOLUTE_TOP@ instead
of just the relative path to lib.
client-core segfaults if there is no secret key installed on the FS (i.e. generated by
genconfig without the --genkey option)
this was caused by a NULL pointer to secret_key field in the fs_conf structure.
Now we should be able to handle 0 length fs secret key fields.

Merge posix-extensions-branch to HEAD
This branch implemented patches to the 2.6.16 kernel for the proposed POSIX I/O extensions and those patches are under
the patches subdirectory.
It also implements the PVFS2 specific hooks for these system calls.

Tools that may be of immediate use to the pvfs2 general audience is the pvfs2-lsplus utility in src/apps/admin
that should be noticeably faster than the pvfs2-ls utility if there are a lot of objects in a single directory.
Other features are left out by configure and are not even built if the kernels do not support those features
and callbacks.

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

Reverse merges from HEAD..

build fixes on Chiba.

Backmerge with HEAD

dependancy was not taking db cflags into account

Added tools for mpi and mpi test program.

server-config.c is only user of dotconf, remove some unnecessary includes

Reverse merge from trunk to lockd-branch...

reverse merge of branch

reverse merge of trunk to branch

quote CC when passing it through the shell to preserve possible spaces

build support for openib

disable many intel compiler warnings to avoid overwhelming build output; add comments about some gcc options but do not enable anything new

pass compiler to dependency generator script

Includes reverse merges from trunk
in addition, it includes fixes for readdir and readdirplus to have a variable length
number of directory entries passed in from client-core using a trailer page (currently).
However, there is nothing that prevents us from sharing those pages with the kernel using vmap
and friends which is the next logical step.
So client-core now does a readdir/readdirplus and writev's the trailer page (which is essentially
an encoded version of the readdir/readdirplus response) to the kmod.
The kernel module decodes that and copies it to the user-space app (which issued the getdents/getdents_plus
system call)
Added/edited the getdents.c test program in the test/posix sub directory to issue
getdents/getdents64/getdents_plus/getdents64_plus system calls
Also fixed a critical bug that was somehow never triggered in readdir if the buffer size
provided by glibc/user was not sufficient enough and we advance the f_pos token beyond where we stopped.

remove configure-generated file

Added a few more missing files and fixes from HEAD to lockd-branch.
lockd-branch should build cleanly now.

commit of versioning stuff to branch

make target and supporting script to publish the automatically generated pages

onekeyval merge

fix compile error and merge from trunk

gotta have the -ls in the right order and fix the admin target

mostly working onekey impl

ib: separate config options for includes and libraries, optionally

branch needs SERVER_LDFLAGS from trunk

added --enable-static-server to config

zero-fill appears to work now (finally).

added all dep to install target.  now I can just type make install and magic stuff happens!

stuff from trunk.

get the verbose but functional make dist back into the branch

commit fixes to 1.3.1 branch as well

fix for berkdb checking when --with-db is used

reverting 1.204.  This means "make dist" might not do the right thing, but
until i can figure out why "make all" now removes generated .c files,  this
will get the nightly tests running again.  Need this patch (or something like
it) before next release after 1.3.1.

fixes to small io stuff to get test-zero-fill working again.  Also committing updates from HEAD to make things easier for merge.

more tweaks on the "ship state machine parser, not generated c" fix

* malloc.h is obsolete in freebsd, check for it
* No error.h in freebsd
* linux allows Lu,Lx, and Ld to be interchangeable with llu,llx, and
lld.  Other platforms (solaris, bsd) don't.  Replaced all the L*
with ll*.
* bsd doesn't have mntent.h.  use fstab.h and our own funcs.
* bsd doesn't know about ssize_t.  Use size_t instead.
* bsd doesn't have asm errnos.  Check for each of those
errno macros and if not defined define to -1
* check for db3/db.h db4/db.h header if db path is specified using
--with-db
* darwin has aio support builtin, doesn't need -lrt
* remove kernel stuff in pvfs2-types.h that sets WORDSIZE.  gcc
defines __LP64__ on archs with 64bit longs so we use that instead.
* darwin's open doesn't accept O_LARGEFILE.  off_t is an int64 by
default.  define O_LARGEFILE if it hasn't been.
* need to run ranlib on darwin after libpvfs2.a gets installed to regenerate
table of contents

slang's suggestion: we don't want to build-depend on lexers and parsers, so
ship the generated statecomp code.

Attached patches implements trusted ports and networks similar to
what PVFS1 implemented.
It adds a new <Security> </Security> tag to the fs.conf config file
and parse utilities accordingly. We can edit these later if there
is a strong need.

Within these tags, you have to specify TrustedPorts <Range of ports> and
TrustedNetwork tcp://<network> tcp://<mask>.
I dont know how to support this cleanly on  non tcp/ip bmi
architectures, so currently this only enables it for the bmi_tcp module
alone. (i guess we could do it for bmi_ib module as well but not done
until someone says that is the right thing to do...)

Also adds support for a --trusted argument to the pvfs2-genconfig script
that will prompt for the trusted port ranges and the network address/masks and write it out
to the fs.conf and server.conf file.

Configure now supports a --enable-trusted to enable all these features.
(NOTE that the fs.conf tags (or lack of one) still allow you to not
select these features even if configure was asked to compile with this
option).

this patch also cleans up all the warnings that arises with gcc
(3.4) and --enable-strict build mode.

parallel to just_kmod builds, add targets to install without building everything again too

updates to my event changes to bring them inline with trunk

merged acache changes to trunk.

once in a rare while (BGL) we need to build the kernel module separately from
the kernel apps (pvfs2-client and pvfs2-client-core

merge changes from trunk since branch point into branch.  Effectively moves the branchpoint to the latest trunk.

removed refs to flowproto-bmi-trove stuff

[pcarns]: Introduce mount.pvfs2.  At the expense of an additional helper
utility, makes pvfs2 mount commands and fstab entries the same for linux-2.4
and linux-2.6.  Update documents accordingly

install pvfs2 to a pvfs2/ directory in the kernel (like all the other file
systems)

in the "make kmod_install" and "make kmod24_install" cases, make the sbin
directory before stuffing kernapps there.  helpful for packagers who might
package kernel stuff separately from userspace.

- make kmod_install and kmod24_install depend on kmod/kmod24 and kernapps
- honor the kmod_prefix environment variable.  packagers might want to put
  kernel module in a nonstandard place

continue disabling trove autogen

phil's patch to separate NPTL workaround from 2.4-with-2.6-backports workaround

don't attempt to install ADMINTOOLS_SERVER if we did not opt to build the server

don't build the kernel tools as part of 'all'

we have to build statecomp with the host compiler, not the cross compiler

More work to make building of just server possible.  There is a hack here
in that we still build server/request-scheduler stuff, which needs to be
moved out of src/server.

More little tweaks to the configure/make process to help with BG/L builds.

first stab at a "disable server" rule

merge statecomp process into build; remove old CLEANFILES and PRIVATEDEPENDS vars

simplify karma and vis builds

remove unused flag

point to source dir, not target, for def local include

eliminate global src/server include

reduce flow.h include usage

reduce llist.h include usage

- build: remove duplicate commands used for QUIET_COMPILE
- build: fix test configure library detection

hide karma details in its module.mk

fix for out-of-tree: server stub lives in srcdir

fix karma ldso dependency, add another source dir to tags

'make install' needs to install karma

- added a configure test looking specifically for a redhat 9 release,
  as many things have to be changed on that platform:
  - use some 2.6.x kernel code on redhat 9 2.4.x kernels, if detected
  (this means we no longer have to use the externally referenced
  redhat patch)
  - do not mark pages as reserved on redhat 9 2.4.x kernels, as the
  kernel will oops eventually due to the different virtual memory
  manager
  - since redhat 9 has an incompatible NPTL implementation (and the
  server mysteriously crashes on it), install a wrapper script in
  place of the pvfs2-server that exports the LD_ASSUME_KERNEL
  environment variable before running the server
  - addition of above mentioned src/server/pvfs2-server-stub
  - in redhat 9 only cases, install pvfs2-server-stub as pvfs2-server
  and install pvfs2-server as pvfs2-server.bin
  - make sure that make kmod{24}_install works properly even if there
  are multiple UTS_RELEASE definitions in the version.h file (as there
  are in redhat 9).  fallback to using uname -r for kernel
  installation path in /lib/modules if that's the case

NOTE: at this time, we should be able to work properly on redhat 9
installations "out of the box".  in my testing, this is the case.
VS: ----------------------------------------------------------------------

update top level build to skip generating the .libs file that was
previously used to relay library information (in particular, which db lib
to use) to the test subdir build

break db libraries into a seperate variable from LIBS, to make sure that it
only gets used on server side

undo ib-disables-tcp hack

dur. i meant 'pvfs2-config'

initial cut of pvfs2-config, mostly stolen from sdl-config

make sure that MISCSRC makefile category is initialized; otherwise
pvfs2-server.c will not be covered in the dependency rules

- added an optional socket management component that uses epoll() instead of
  poll() for its work cycle (epoll is a linux 2.6 specific alternative to
  poll that should result in much lower latency and better scalability)
- added a --enable-epoll configure option to enable it
- fixed a bug in configure.in; AC_PROG_CC was being called too late, which
  made it possible for several earlier tests to (if triggered) accidentally
  discard the default CFLAGS that we would normally expect
- epoll implementation needs more testing (both for correctness and
  performance)

some shell hackery to automatically add what used to be added by hand. Add
comments to remind future authors to make updates in two places.

some tweaks to allow msgpairarray to be reused with non-client state
machines

tweak distclean target so that it does not remove autogenerated files in
release snapshots (which cannot be recreated)

slight reorganization of how .c files generated from .sm files are tracked
so that the client and server rules are more consistent

- 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

- pull mainline changes into this branch

makefile typo, fix suggested by
sven@pvs.informatik.uni-heidelberg.de

- merge in as much as possible from the main tree to keep up to date

make LD selection honor the compiler found by configure; suggested by
Sven Marnach sven@pvs.informatik.uni-heidelberg.de

(branch commit) added rules for creating a multithreaded client library in
addition to normal library
- "make lib/libpvfs2-threaded.a" must be done manually to build it
- resulting lib is buggy right now, but may be useful for testing later

whoops- added makefile fix that was accidentally applied in wrong branch
before

update makefile to make sure that a few special case tools (admin tools that
link against the server lib) are installed

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

get rid of generated config files on make clean

- initial addition of linux kernel 2.4.x support code and build requirements
   (thanks to Murali for working on a prototype and reviewing the merge)

avoid multi-method TCP and IB build since it does not work and adds much operation latency; remove icky -rpath

Makefile bugfix by Ti Leggett:

"However, there is a real bug. KMOD_DIR is supposed to be the dir where
the module(s) get installed, but the Makefile ignores it if it's set and
resets it. I've attached a patch."

while we're at it, generate html versions as well

Break the build process for the test subdirectory out into its own
configure and makefile setup.  Various tweaks along the way.  Some stuff
still needs to be streamlined.  It should work for both in tree and out of
tree builds without any special options to configure as long as you build
the main pvfs2 tree first.

more makefile tweaks after automated test run pointed out some problems :)
also recategorized pvfs2-client stuff one more time so that it can be
installed in prefix/sbin instead of prefix/bin where the admin tools are

update makefile.in to install pvfs2-client from the proper directory

add comments for kmod and kmod_install targets

Remove -D__STATIC_FLOWPROTO_BMI_CACHE__ from the "CFLAGS".

Add "-D__STATIC_FLOWPROTO_BMI_CACHE__" to enable the client side
to use the BMI_CACHE flow protocol.

compile in support for Jiesheng's caching flow protocol; enable at runtime
with config file options

turn off one intel compiler warning, auto-install kernel module directory per nathan

fix kmod_install target

fix up the 'make cscope' rule so it works out-of-tree even if configure called
w/ a relative path

do not build/install kernel module automatically

- Build system:
    - list explicit include directories
    - add cflags provided by module makefile stubs
    - install kernel module if configured
    - simplify dependency generator script, add .po

- configure cleanups (replaced inconsistent uses of AC_HELP_STRING, as it's
  not supported on autoconf 2.13)
- made the mmap-ra-cache a configure time option (enabled by default but can
  now be disables with --disable-mmap-racache configure option)
- fix mmap-ra-cache bug that didn't flush data at a time when we know it's
  no longer valid; explicitly flush at these times

- Kernel module builds:
    - build the module with everything else, if configured
    - clean up properly
    - work out-of-tree

add shared and static libraries build options; default static only

can now use MPE or Pablo to profile pvfs2.  still needs more work.

notice when compiler is Intel C; adjust compile flags appropriately

Karma GUI in place.

- redefine WORDSIZE definition so as not to need pvfs2-config.h
- remove installation of pvfs2-config.h
- remove configure checks for sizeof(long) [ no longer needed ]
- version bump to pvfs2-0.0.8pre2, because pre1 is old and boring now

pvfs2-types.h needs pvfs2-config.h, so add it to the headers we install

make example config files be automatically generated

Removed simple.conf and server.conf from source tree.  Now at compile time
some similar config files are generated (with good debugging defaults)
and placed in build dir if tree is not a distributed release.

another large file fix, this define gets us more functionality

added rudimentary font support, labeled max on y axis

finished integrating build rules for vis tools into top level makefile, some
other minor cleanups.  vis tools will now build by default if we find
the libraries that we need.

temporarily disable compilation of some of the animals in the flow protocol
zoo, to make it easier to push some api tweaks without updating all

started updating flowproto_multiqueue to operate on the client side as well
as the server side, now works correctly for simple memory -> bmi transfers

add a perl script to the taglist

add quiet-compile option for clean (good for slow links)

tags should look back in src dir for source files

let ctags parse .sm files too

stubbed in a little more of flow protocol, limited to be used exclusively
on the server side for now at least- will be compatible with existing
flowproto-bmi-trove on client end.

stubs for a new flow protocol

added a proof of concept vis tool (pvfs2-vis-bw-2d), very rough cut at drawing
real time bar graph of server throughput.  backend (pvfs2-vis.c) should
be good, though.  Also added some configure stuff to find SDL if it is present

verify and document our supported trove/job thread configurations

use gcc-specific flags when possible on both CC and MPICC

install new header

fix for broken builds.

Separate out gcc-specific cflags.  Use separate cflags for mpi builds.  Add rule to generate output from cpp.

also install pvfs2-client{,-core}

don't blow away the docs unless we do a distclean

install manpages

make dist bug- forgot to make sure trove autogen code is in place

disable generation of trove-autogen code in dist snapshots

added a mechanism to disable auto code generation by adding special
makefile included flags to the dist tree; still have one auto-generated
trove file

added some features to make dist, still needs a little more work to
prevent autogeneration of .c files in resulting package

Don't install the ADMINTOOLS_SERVER:  the server already does the job of
mkspace, and showcoll is too low-level for regular diagnostics.

ratzen fratzen /

fix install target since pvfs2-types.h is no longer generated,

mung things a bit so we can build GM stuff on systems like jazz

a 'make dist' start.  the script's in maint/make-dist.sh.  it might almost
work.  if not, please fix it.

do not autogenerate pvfs2-types.h

Sometimes the MPI implementation does not have MPI-IO support (hi neillm :> ).
Break out the mpi-io tests into a separate target that doesn't get built as
part of 'mpitests'.  This also breaks a potential dependency loop: ROMIO needs
PVFS2 to build, and PVFS2 needs ROMIO to build mpi-io tests.

first swing at a redhat style rc script for pvfs2, needs some polish (it
is generated at configure time to reflect --prefix option)

properly build mpi binaries from mpi objects

update Makefile for new MPI rules. generate makefile module for mpi-io-test

the new db tests will stuff the "right" db into @LIBS@

allow make V=1 to turn off quiet compile easily for testing

compile in support for bmi cache flowprotocol

clean up cscope output in make distclean target

make sure that the machine has a proper db development package installed
at configure time.  If not in standard path, --with-db=<dir> support was
added.  Also useful for testing against multiple installed versions of db
on a system.

add more distclean targets; allow CLEANFILES to be augmented in sub-make modules; add PRIVATEDEPENDS for depends generated by sub-make module

add target to build tags; simplify dependency avoidance

add bmi_ib to configure/make

cleaned up another loose .o file in make clean target

picked up some stray obj files in make clean target

add pvfs2-types.h.in to cscope.files

- added pvfs2-mkspace and pvfs2-showcoll for removal on make clean
- added versioning information to the configure setup so apps can know
  which version to print if they care (i.e. pvfs2-ls, pvfs2-ping, etc)
  NOTE: I've arbitrarily picked the version start of 0.0.1 -- if we
  don't like this version for the friendly release, be sure to update
  the version in configure.in
- declare the strnlen prototype in pvfs2-showcoll, as the warning seems
  to have re-appeared; fixed for good now

gr. 3rd time's the charm.  installing genconfig now works for out-of-tree buids

dur..actually finish typing the path to genconfig

install genconfig, too

made the kernel make file autogenerated if the --with-kernel arg is used with
conifgure and pointed to a proper 2.6.x kernel tree.  Made some aesthetic
changes to configure --help output.  Some other minor changes.

one more fix to make things out-of-tree-build friendly

since pvfs2-types is autogenerated, make the CFLAGS out-of-tree-build friendly

some configure/type updates:
- new PVFS type: PVFS_aint, which is an integer just large enough
  to also be used to hold a pointer
- configure test to detect typedef to use for above
- move pvfs2-types.h to pvfs2-types.h.in so that autconf can generate it
- added removal of config.cache to make distclean target

more pcache tweaks.  the io and getattr state machines should now be properly
using the pcache for entries they're dealing with.

also directory attrs are now being cached in the pcache as well.  net result
is subsecond vfs ls' on a moderately sized directory listing.

A system interface pcache replacement.  This will probably break everything,
but since we're breaking everything anyway, here goes.

catch lib/pvfs2-server.a in make clean

use tabs, not spaces

temporarily disabling flowproto_dump_offsets

put it a check for glibc 2.3.2 to determine whether or not we want to define
__PVFS2_TROVE_AIO_THREADED__.  This check should be improved eventually.

callback based aio completion for dbpf-bstream operations if you set
__PVFS2_TROVE_AIO_THREADED__

added some admin tools to make install target that I previously split out
into a seperate build rule

moved admin tool rules back down into top level makefile

somewhat dirty first cut at disabling trove for client side flow protocol

updated build rules for a couple of admin tools that require trove- not
very pretty right now, will fix a little later

server objects now get put into a library to link into server binary

install the new, more streamlined, header files

Added dbpf thread stubs and a trove_testcontext method that should eventually
replace trove_testsome to better fit with the job interface.

bugfix: dependencies weren't being generated for admin tools

flowproto_dump_offsets (used for debugging file and memory offsets within
flow execution without actually moving any data) is back in the saddle agian,
but untested.

add io-test to the tests we install

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

cscope target: add .sm files to the cscope database

if we install everyting in TESTS, we get name collision.  i'd like to thank
phil for ending my confusion by pointing out that i was using the server
'readdir' instead of the sysint 'readdir'.  now install a few useful tests.

install the tests as part of 'install_tests' instead of 'install'

initial "make install" target. could use some refining but gets the job done.

more random stuff :).

minor build fix.

Moved rule for building .c from .sm into root Makefile.

Correctly generates MPI program dependencies for out-of-tree builds.

update comments to remind me what libs are needed on solaris

replaced "find -iname" with "find -name" in the build process; the former
option isn't available on non gnu find

modified configure/makefiles so that you must specify --with-mpi= at configure
time in order to build mpitests.  Also made mpitests part of default target
if --with-mpi is specified.  Also added some helpful error messages in
makefile and configure, and added test to verify that mpicc works.

changed default BMI GM buffering mechanism

GM module now builds, missing a few API functions and features...

finished makefile changes to enable compilation of GM module if GM is
found; unfortunately the GM module doesn't build right now even if you
enable it, though...

added configure tests and beginnings of some makefile changes to support
building the BMI GM module if GM is found

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

Added pvfs2-mkspace utility, but still need to run server setattr for now.

Uses new pvfs2latexwrapper.pl to create doc files; helps with out-of-tree
builds.

Couple of minor changes to enable profiling.  Only sort-of works, but
shouldn't hurt anything.

Added flags parameters to some of the dspace calls; updated all other code too.
Added coverage target in makefile; not working yet.

Monster checkin:
- removed the trove "fs" files, which are no longer in use.
- added dbpf-collection.c, which keeps up with multiple collections in use
  at one time.
  - hooks in place in dbpf implementation to theoretically handle multiple
    collections; untested (works for one collection).
- added trove-create-stress.c, a stress test for object creation.
- added TROVE_SYNC flag as an option, which is handled in most calls.
- found and removed a number of problems in error handling in dbpf calls.
- created pvfs2codecheck to identify issues in the code to be cleaned up.
- eliminated a number of memory leaks in trove dbpf.
- moved flags into top of the op structure in trove.
- implemented (but did not test) dspace_verify().

use configure-derrived LIBS in addition to the libs we know we will need

added another quick comment

added MPI/BMI benchmark programs to the pvfs2 tree; required modifying
build system to understand how to compile MPI programs (see make
mpitests target).  Also added some scripts for running the benchmarks.

big change; lots of reorganization, cleanup, commenting, and some
minor bug fixes

added a special "statecomp" target that rebuilds the statecomp
tool without regenerating any dependencies

modified so that there is not explicit rule to rebuild _all_
dependencies if any SRCFILES change.  This doesn't appear to be
necessary anymore, and it clutters the output for people working
on generated code (statecomp and server)

added a bunch of documentation to cvs tree, and some not quite working
build rules for it

Enabled use of "malloc trace" via --with-mtrace to configure.  Currently enabled in the trove initialize; we should eventually get the initialize call moved into some higher level initialize call in the library and in the server.

fixed cflags passed to server so that the job library is built with
thread support

Added use of dspace db cache, made dspace create and dspace remove really nonblocking, renamed some handle functions so they didn't look like they were grabbing locks.

took -Wconversion out of the default cflags for the moment

shuffled some variables around in the makefiles to make rules a
little cleaner

added in core server code and rules to build it

fix to cscope rules.

added cscope build rules.

added in most of the server code

removed a bunch of target stuff for now because some build stuff was
accidentally checked in without the corresponding source

modified makefile to catch includes from generated header files in
src/common/statecomp

added statecomp tool into new tree and added rules to build it

turned on bmi flags to request protocol encoding test all the time for
now, removed print-struct.h from inclusion

undid some previous work on how to build documents; realized that it is
pretty broken but I don't want to fix right now

added first document to tree (developers-guidelines); added
infrastructure to makefile to build docs and clean them up

modified so that dependency files aren't included if make is executed
with the "clean" or "distclean" target specified on the command line;
modified distclean target to delete all of the files that configure
generates

added flags to turn on large file support by default

added in job test programs

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)

distclean now removes the top level makefile as well

moved deletion of trove-autogen.c from the clean target to the distclean
target ("make clean all" now works)