openafs.git
13 years agoLinux autoconf: fix name for struct cred test
Marc Dionne [Tue, 27 Apr 2010 22:36:46 +0000]
Linux autoconf: fix name for struct cred test

The new macro scheme changes the name of the tested define.
Update the code to match.

Change-Id: I0e9f04692d8fbb6e9452a9b169c191ce7148349b
Reviewed-on: http://gerrit.openafs.org/1868
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux autoconf typo
Marc Dionne [Tue, 27 Apr 2010 22:25:15 +0000]
Linux autoconf typo

The test for the new export ops has a typo which
makes the test fail where it should succeed: eops -> _eops

Change-Id: I97dd0707b911c51f6394ada426a98e06a83ad35e
Reviewed-on: http://gerrit.openafs.org/1867
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoAdd some directory package doxygen comments
Tom Keiser [Thu, 25 Mar 2010 17:08:51 +0000]
Add some directory package doxygen comments

Change-Id: I75c358f08c71b2de722de03bb65fb8696657d57f
Reviewed-on: http://gerrit.openafs.org/1861
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: autoconf fallout
Simon Wilkinson [Tue, 27 Apr 2010 21:16:24 +0000]
Linux: autoconf fallout

grap != grab - fix spelling of grab_cache_page_write_begin

Change-Id: I4032bff80b6d1d7568af8f9de59a56c0bd6a661a
Reviewed-on: http://gerrit.openafs.org/1859
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agonetbsd: remove old (non-original) NBSD port
Matt Benjmain [Tue, 27 Apr 2010 16:06:48 +0000]
netbsd: remove old (non-original) NBSD port

Prepare for rebase of NBSD port at the current OBSD port,
which is descended directly from the original NBSD port and
more closely resembles it.

Change-Id: I9ff67ebea04534ac5ddc11c0aa1d65028067806a
Reviewed-on: http://gerrit.openafs.org/1858
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Remove unnecessary crhold on incomplete kernel credentials
Marc Dionne [Sun, 25 Apr 2010 02:12:27 +0000]
Linux: Remove unnecessary crhold on incomplete kernel credentials

When credentials debugging is active in the kernel, sanity checks
in various credentials functions such as get_cred or put_cred
require that they are called with full fledged kernel credentials.
This can be a problem with recent kernels in osi_Init where we
build our own credentials and try to "crhold" them.
Getting a reference doesn't make much sense anyway since we're
using a statically allocated structure that we never try to free
or modify.

As a minimal fix, just don't call crhold when we're using
the kernel credentials structure.

Change-Id: If4f843ad33526c6a42f1cf67c57763593ffa50ac
Reviewed-on: http://gerrit.openafs.org/1828
Tested-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

13 years agoflush changes on LOCK_EX unlock
Derrick Brashear [Mon, 26 Apr 2010 21:43:16 +0000]
flush changes on LOCK_EX unlock

right now, flock with LOCK_EX, on unlock, triggers an async store.
make it sync, but also, make sure to ask to have any in-memory
data sent to us (e.g. VM_StoreAllSegments); unlike Solaris
VMSYNC_INVAL, we don't want to invalidate the pages, just get copies
written back.

LOCK_SH did not, does not and AFAICT should not trigger writes.

Change-Id: Id4a72f73b685b5566bb31f6f610f22d806899280
Reviewed-on: http://gerrit.openafs.org/1846
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoallow file length to be updated unless we're writing
Derrick Brashear [Mon, 26 Apr 2010 20:41:29 +0000]
allow file length to be updated unless we're writing

right now, merely having the file open O_RDWR, or mapped,
precludes a length update from being reflected.

Change-Id: I7f732b053c392aa6cc1c8ad6de8e0f4bd6468728
Reviewed-on: http://gerrit.openafs.org/1841
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Reorder the tests
Simon Wilkinson [Tue, 27 Apr 2010 18:31:55 +0000]
Linux: Reorder the tests

As a final change to the Linux autoconf stuff, reorder the list of
tests so that it's more easy to see what's being checked for, and to
prevent multiple copies of the same thing being added.

Change-Id: I84ca132e178b45ac25ebd48e71193bdd84965770
Reviewed-on: http://gerrit.openafs.org/1857
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Don't preprocess osi_vfs.h
Simon Wilkinson [Tue, 27 Apr 2010 17:57:30 +0000]
Linux: Don't preprocess osi_vfs.h

Before we did inodes properly, we needed to pre-process osi_vfs.h,
in order to merge our inode structure with the Linux one. Ever since
we moved to native inodes, that preprocessing has been disabled, and
we've just copied osi_vfs.hin to osf_vfs.h Skip this pointless step,
permanently rename osi_vfs.hin to osi_vfs.h, and remove the unused
support scripts.

Change-Id: I5e21eccf2242080cd8b994bd1654260bfb531420
Reviewed-on: http://gerrit.openafs.org/1856
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Don't waste autoconf checks on cpp defines
Simon Wilkinson [Tue, 27 Apr 2010 16:49:19 +0000]
Linux: Don't waste autoconf checks on cpp defines

If something is a #define, then there's no point in doing a test
compile to check for its presence. Instead, just do #if defined(X)

Remove the tests for for_each_process and prev_task.

Change-Id: Id8a9fddf2571cccd5d61982d4662bce2009c8f43
Reviewed-on: http://gerrit.openafs.org/1855
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Add general autoconf macro for Linux kernel
Simon Wilkinson [Tue, 27 Apr 2010 16:41:22 +0000]
Linux: Add general autoconf macro for Linux kernel

Add AC_CHECK_LINUX_BUILD() to do a standardised Linux build, which
takes a "checking" message, the autoconf variable to use to cache
the results, headers and code to run, preprocessor variable to
define, and a description of that variable.

Reimplement all of our existing check macros in terms of this one,
resolving many typos along the way.

Change-Id: I41988c83fcdbfbf8152f6dd0e7c4bd16c7a04240
Reviewed-on: http://gerrit.openafs.org/1854
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Add autoconf macro for structure checks
Simon Wilkinson [Tue, 27 Apr 2010 12:02:20 +0000]
Linux: Add autoconf macro for structure checks

Add a new autoconf macro for doing structure element checks, and
modify all simple structure checks to use it. This introduces a
standard name form - STRUCT_structure_HAS_element, so there are
some changes in the code to make use of this standard form.

Change-Id: Ife967322503ae6f428e76845000de04f26929e65
Reviewed-on: http://gerrit.openafs.org/1853
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Macroise kernel function autoconf tests
Simon Wilkinson [Mon, 26 Apr 2010 22:54:05 +0000]
Linux: Macroise kernel function autoconf tests

We spend a lot of time in autoconf checking to see if kernel
functions are available. Rather than copying and pasting the same
code everytime we do so, create a macro to do function tests, and
use that macro. This results in standardised naming, so change all
of the places that use autoconf results to use the same standardised
names.

Change-Id: I36212e6c28c4b8455f859da1bbf3e456a2aabc07
Reviewed-on: http://gerrit.openafs.org/1849
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Tidy up freezer code
Simon Wilkinson [Mon, 26 Apr 2010 21:13:39 +0000]
Linux: Tidy up freezer code

Linux now provides try_to_freeze(), which can be used to replace
all of our refrigerator code. This has been in the tree for a long
time, and actually predates many of the changes we've been modifying
our code to work with. So, use try_to_freeze wherever we can, and keep
a simpler compatibility function for kernels which are too old to have
it.

Change-Id: Iec50f83382f46d2b17b43da815a75755ea916bc6
Reviewed-on: http://gerrit.openafs.org/1845
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Simplify header file checks
Simon Wilkinson [Mon, 26 Apr 2010 20:34:17 +0000]
Linux: Simplify header file checks

Provide an autoconf macro to perform tests for the existence of
Linux kernel header files. Use this to standardise the naming of
header file presence #defines, and to simplify the linux tests.

Change-Id: I49629679db56c5f9a39487bd78fc2e59d5da0269
Reviewed-on: http://gerrit.openafs.org/1844
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoRename afs_nbsd_* functions in OBSD
Matt Benjamin [Tue, 27 Apr 2010 17:03:36 +0000]
Rename afs_nbsd_* functions in OBSD

The OBSD cm port prefixes function names with afs_nbsd_, an
artifact of its origin in the original NBSD cm port.  Make the
prefix afs_obsd_.

Change-Id: Id412fe0f752a147e2a9afda904db220ea6efd736
Reviewed-on: http://gerrit.openafs.org/1850
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoOverzealous .gitignore under rxkad
Simon Wilkinson [Mon, 26 Apr 2010 22:36:38 +0000]
Overzealous .gitignore under rxkad

The .gitignore in rxkad ignores sboxes.h and fcrypt.h,
since they used to get copied there from the domestic subdir.
Now that they actually live under rxkad, they souldn't be
ignored.

(with thanks to Marc for the commit message)

Change-Id: I1dd6766e75c15ca7f5604601bbe5b4e67f4fffa6
Reviewed-on: http://gerrit.openafs.org/1848
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoOverzealous clean target under rxkad
Marc Dionne [Mon, 26 Apr 2010 22:15:16 +0000]
Overzealous clean target under rxkad

The clean target under rxkad removes sboxes.h and fcrypt.h,
since they used to get copied there from the domestic subdir.
Now that they actually live under rxkad, they souldn't be
removed.

Change-Id: I867496dd0d3dd997d402ebfbfc6272447b1af98e
Reviewed-on: http://gerrit.openafs.org/1847
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agodcache dirty pages flag is solaris only
Derrick Brashear [Mon, 26 Apr 2010 19:55:57 +0000]
dcache dirty pages flag is solaris only

note that IFDirtyPages is Solaris-only, so there's no confusion. may be
worth visiting to see if and how the code should be minimized or adapted
for elsewhere

Change-Id: I7dfe4760eb4038d23a920c2e23f33d93e604bbca
Reviewed-on: http://gerrit.openafs.org/1840
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Don't hide memory management
Simon Wilkinson [Fri, 23 Apr 2010 16:07:40 +0000]
Linux: Don't hide memory management

Given that we now only switch from user space into kernel space in
one place, don't hide the mechanism we use to do so behind macros.
This makes it much easier to visually confirm the correctness of the
code.

Change-Id: Ie52e071c81a9178c792be1eaa7e36d8453ebb319
Reviewed-on: http://gerrit.openafs.org/1820
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Dan Hyde <drh@umich.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: The kernel module always runs in the kernel
Simon Wilkinson [Fri, 23 Apr 2010 16:10:15 +0000]
Linux: The kernel module always runs in the kernel

There's no situation where we'd be including osi_machdep.h and not
be building for the kernel, so just remove the case that disables all of
the locks if __KERNEL__ isn't defined.

Change-Id: I1141c37cf0cd517191dee7ca2404e109d4053c29
Reviewed-on: http://gerrit.openafs.org/1821
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agofix other oldtvix typo
Derrick Brashear [Mon, 26 Apr 2010 04:21:30 +0000]
fix other oldtvix typo

i failed to push this to commit 0dac8f93af5262472bc5bdf94dbb605a24da59d0,
and so, well, here's the real fix. sorry.

Change-Id: I6799ecb207f246ac3e4c48b1de55268d2fba4d23
Reviewed-on: http://gerrit.openafs.org/1837
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoInitialize oldvtix
Marc Dionne [Thu, 22 Apr 2010 21:22:37 +0000]
Initialize oldvtix

Initialize the variable to avoid compiler complaints that it
could be used uninitialized.  Use a sentinel value and issue
a warning if we try to use the variable and we haven't set
it to a useful value.

Change-Id: I037be2ad8e0b55d06b50aafdeca5012764640f16
Reviewed-on: http://gerrit.openafs.org/1810
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Use kernel network helper functions
Simon Wilkinson [Fri, 23 Apr 2010 17:03:49 +0000]
Linux: Use kernel network helper functions

Linux has various network helper functions available, which are
guaranteed to do the right thing in terms of setting address space
limits, not dereferencing NULL pointers, and the like. Use these, where
they're available, rather than rolling our own.

kernel_sendmsg and kernel_recvmsg appeared in 2.6.8.

Change-Id: I1cd91afd2182ad936756bbc4cf692262499c16e4
Reviewed-on: http://gerrit.openafs.org/1822
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Remove old code from RX kernel implementation
Simon Wilkinson [Fri, 23 Apr 2010 17:38:02 +0000]
Linux: Remove old code from RX kernel implementation

The RX implementation was split into LINUX and LINUX24 along with
the kernel, but the old 2.4 and 2.2 only code was never removed from
the LINUX/ directory. Do so now.

Change-Id: I80300a1390e297298f4af3c06086e25d0acf8c1a
Reviewed-on: http://gerrit.openafs.org/1819
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Don't pass f_pos down to the filesystem
Simon Wilkinson [Fri, 23 Apr 2010 15:54:39 +0000]
Linux: Don't pass f_pos down to the filesystem

In 2.6.8, Linux shifted from passing f_pos directly to the read
and write routines, and started passing a copy. This helps reduce,
but does not remove, the race issues with f_pos itself. Make this
change for us.

Take the opportunity to remove the uneccessary macros, and tidy up
some casting.

Change-Id: I3b4cdf1e6e8127cbe0055829605268953c4397a6
Reviewed-on: http://gerrit.openafs.org/1818
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Removed unused define
Simon Wilkinson [Fri, 23 Apr 2010 14:59:46 +0000]
Linux: Removed unused define

VFS_STATFS isn't used in the Linux 2.6 case, and is wrong anyway.
Remove it in an attempt to clarify some particularly muddy water.

Change-Id: I404429e3eeb87146bfdced5625b56857ba2ab9fd
Reviewed-on: http://gerrit.openafs.org/1817
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Don't roll our own llseek
Simon Wilkinson [Fri, 23 Apr 2010 14:07:58 +0000]
Linux: Don't roll our own llseek

llseek is a little more complex than just setting the pos field of
the file structure.

Where vfs_llseek is available to us (it has been since 2.6.8) use that
directly. If vfs_llseek isn't available, then use either the
filesystem's llseek method, or default_llseek (available since 2.6.0),
to ensure we get the proper locking.

Change-Id: I26bcbff984c7164be1724eee7a46cbe6e02bc510
Reviewed-on: http://gerrit.openafs.org/1816
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agocachemgr store file too big error exit via usual path
Derrick Brashear [Sat, 24 Apr 2010 21:05:47 +0000]
cachemgr store file too big error exit via usual path

have just one path for exiting on errors so we do everything

this means we will EndCall, instead of not, in the EFBIG case

Change-Id: I2df025eee4bd5173ed4f80a73f9a9a347e7257d2
Reviewed-on: http://gerrit.openafs.org/1827
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoMove rxkad/domestic/* into rxkad/ directory proper
Simon Wilkinson [Sun, 25 Apr 2010 21:07:04 +0000]
Move rxkad/domestic/* into rxkad/ directory proper

Historically, the fcrypt code lived in its own directory, named
'domestic', so that ACLs could be used to prevent nasty foreigners
from getting access to it. Now that those days are gone, having the
domestic directory just complicates builds. Remove it, and reunite
fcrypt with the rest of the tree.

Change-Id: I2d1f66463121bbb391260b613913d76c27931429
Reviewed-on: http://gerrit.openafs.org/1836
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoImprove shared library building support
Simon Wilkinson [Sun, 25 Apr 2010 18:11:06 +0000]
Improve shared library building support

This change removes the need to maintain two forms of symbol export
lists, instead generating the simpler lists required by AIX and Mac OS X
from the more complex Linux and Solaris mapfile. It can only handle
simple mapfiles at present, any more complex files (for example, those
which contain versioning information) will require changes to the
translation script. The now unused .exp files are removed, and a
dependency on the .map file added to the library build line.

This change adds support for export lists to Darwin for the first time.
Doing so identified a number of symbols in libafsrpc that are required
by libafsauthent, which were not being exported. Export these symbols,
and bump the minor revision of the shared library.

Change-Id: Ibd1d02bb89b85500dc2a010e6cde1f4b81efe050
Reviewed-on: http://gerrit.openafs.org/1831
Reviewed-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Remove prototype warning inhibition
Simon Wilkinson [Sun, 25 Apr 2010 10:31:36 +0000]
Linux: Remove prototype warning inhibition

The Linux kernel builds used to add -Wno-strict-prototypes to the
CFLAGS. Now that we've got prototypes for pretty much everything,
remove this line so we can see any real warnings that might occur.

Change-Id: I71b97bd3d19259f6ec6a55cfe095f5f4bf1ab78f
Reviewed-on: http://gerrit.openafs.org/1830
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoFix make clean and make distclean
Simon Wilkinson [Sun, 25 Apr 2010 09:55:23 +0000]
Fix make clean and make distclean

Fix these make targets so they actually clean the tree. Reorder the list
of file deletions in distclean so that it exactly mirrors the creation
order in configure.in, which should hopefully make it easier to keep
this all in sync in future. Add all of the missing files in packaging,
and the shared library build tools, to this list.

Make kauth and vol clean remove ka-forwarder and fssync-debug,
respectively.

Change-Id: Iae13419f1cb21058882c0b0e0c283e3c947e33c9
Reviewed-on: http://gerrit.openafs.org/1829
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoReally kill des_stub
Simon Wilkinson [Sun, 25 Apr 2010 20:52:20 +0000]
Really kill des_stub

Some bits of des_stub were lingering on. Get rid of it completely.

Change-Id: I853135621a757c78cdb50117500f056ade0e6ab5
Reviewed-on: http://gerrit.openafs.org/1835
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoFix objdir builds
Simon Wilkinson [Sat, 24 Apr 2010 16:34:02 +0000]
Fix objdir builds

Change f2db78a346112f5320efc6f0b6b9fe4ae0d893d3 included hard
references to compilation sources with no srcdir prefix, and so
broke objdir builds. Fix this so that we can build outside the
source tree again.

Change-Id: Iaae2d4a1047cc2ac0f5f4aae92cadc1f2a0d09e2
Reviewed-on: http://gerrit.openafs.org/1826
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agomacos growl agent warning parser fix
Derrick Brashear [Fri, 23 Apr 2010 18:40:45 +0000]
macos growl agent warning parser fix

warnings start at offset 5, not 6 (warn is shorter than fetch or store)

Change-Id: Ibaa1239b0f9797da149cb05b3f871611b13833c8
Reviewed-on: http://gerrit.openafs.org/1825
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoopenafs 1.5.74.1 openafs-devel-1_5_74_1
Derrick Brashear [Fri, 23 Apr 2010 15:56:08 +0000]
openafs 1.5.74.1

so it can be tagged and non-unversioned builds of the mac client changes can
be seeded for testing

Change-Id: Ibaa231d356bf667b13e271c9868c6ce0182bd2a6
Reviewed-on: http://gerrit.openafs.org/1814
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agomacos launchd be more patient
Derrick Brashear [Fri, 23 Apr 2010 15:02:32 +0000]
macos launchd be more patient

right now we would wait 20 seconds for shutdown (the default),
get impatient and be useless.

be less useless.

Change-Id: I8c6bc9b7546352cfa73d461b7cdddd2b5b5d7889
Reviewed-on: http://gerrit.openafs.org/1813
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoUnix: Modify disk cache versioning
Simon Wilkinson [Thu, 22 Apr 2010 21:24:11 +0000]
Unix: Modify disk cache versioning

This change increments the disk cache version number, and adds a
structure size record to the disk cache header. All old disk caches
will be replaced when the client is started.

With the various changes made to unify our file handles, and to
support large file handles on Linux, the size of the 'fcache'
structure was modified earlier in the 1.5 series. However, fcache
is also the building block of the CacheItems file, so these changes
inadvertently broke users upgrading from 1.4. In addition, as the
disk cache inode is now a union of many different structures, the
structure size is now potentially volatile across both kernel, and
OpenAFS revisions.

Up the version number so old disk caches are invalidated and won't
crash users who are upgrading. Also take the opportunity to add an
item to the header which stores the size of struct fcache used
by the disk cache. If the size on disk doesn't match that expected
by the kernel module, truncate the cache and start again.

Change-Id: I2ee8863d0bfaaaba34272c9e139638e17669a53e
Reviewed-on: http://gerrit.openafs.org/1811
Reviewed-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoWarning fix: cast to expected type
Marc Dionne [Thu, 22 Apr 2010 23:39:21 +0000]
Warning fix: cast to expected type

Cast argument to match the expected type from the format string.

Change-Id: Iddec912ff8a27f46d872fc3101677913adfb537a
Reviewed-on: http://gerrit.openafs.org/1812
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoUpdate nextVnodeUnique before checking inUse
Andrew Deason [Thu, 22 Apr 2010 16:54:06 +0000]
Update nextVnodeUnique before checking inUse

When attaching a volume, update the nextVnodeUnique field for the
volume, before we do any checks on the volume; for example, checking
inUse, which may result in a demand-salvage if we are running DAFS.

If we do not do this, we can schedule a demand-salvage without setting
nextVnodeUnique, and VUpdateVolume_r will update the volume header
uniquifier to nextVnodeUnique+200, when nextVnodeUnique is not set.
So, we will always set the uniquifier to 200. Fortunately, the salvage
should usually fix the uniquifer anyway.

So, set nextVnodeUnique before doing any of those checks, to avoid
screwing up the uniquifier when taking the volume offline.

Change-Id: Ib211bbf5e93efbebcb679259944e7abf211aa6e1
Reviewed-on: http://gerrit.openafs.org/1809
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

13 years agoPrefer EndCall errors in StoreMini
Andrew Deason [Thu, 22 Apr 2010 18:21:52 +0000]
Prefer EndCall errors in StoreMini

Partially revert b1eb6a7a3f80500f0187cc6a1dd2013e1a5e154a, so we do
not mask the rx_EndCall error with a EndRXAFS_StoreData error (for
example, if EndRXAFS_StoreData returns RXGEN_CC_UNMARSHAL, and
rx_EndCall returns VBUSY). We need to agree on how to do this
throughout the tree, but for now, just fix StoreMini.

Change-Id: I4913946089fd0857506d9186f85c5c8115a5b95d
Reviewed-on: http://gerrit.openafs.org/1808
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: Tidy up packaging
Simon Wilkinson [Thu, 22 Apr 2010 17:02:08 +0000]
Linux: Tidy up packaging

Remove the fuzz setting - we're no longer patching the build, so
we don't need it.

Change-Id: Ic56b4d9efed40aedb425d9852018907736ef957f
Reviewed-on: http://gerrit.openafs.org/1807
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoLinux: RedHat packaging updates for RHEL6
Simon Wilkinson [Thu, 22 Apr 2010 16:56:25 +0000]
Linux: RedHat packaging updates for RHEL6

Update our bundled spec file and related tools so they can be used
to build OpenAFS on the RHEL6 beta.

- Make kmodtool recognise el6 as having "modern" kernel naming
  conventions
- Replace %{PACKAGE_VERSION} (which seems to have disappeared)
  with the standard %{version} macro

Thanks to billings and phalenor on IRC for their testing efforts.

Change-Id: I6538af56f365952b98c2dcbafb1c043435449e63
Reviewed-on: http://gerrit.openafs.org/1806
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

13 years agoResolve symlinks for 'fs rmmount'
Andrew Deason [Thu, 22 Apr 2010 14:59:30 +0000]
Resolve symlinks for 'fs rmmount'

'fs rmmount' is the only 'fs' command that does not resolve symlinks
for its pioctls for some reason. Make it resolve symlinks.

Thanks to Arne Wiebalck.

Change-Id: Ifb6bdf0b381abd2e157d93d7eea4d853e1b3a689
Reviewed-on: http://gerrit.openafs.org/1805
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agobyte-range lock warning should include pid
Derrick Brashear [Wed, 17 Feb 2010 05:48:03 +0000]
byte-range lock warning should include pid

is is the same pid cmdebug would print. just include it in the logged
byte-range lock warning.

FIXES 126438

Change-Id: Idd83a4c4a56edf43ab257b3a7f08e1bbb774f04a
Reviewed-on: http://gerrit.openafs.org/1333
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdding osi_procname macro to different platforms
sanket [Sat, 17 Apr 2010 14:14:25 +0000]
Adding osi_procname macro to different platforms

osi_procname(procname, size) is added to different platforms to obtain
the processname in the string procname( preallocated ) with size size

Change-Id: Iab8fed3e99839cee7c07a2de6144af9b09af3a57
Change-Id: Ib4e45564ae942e74c331444969a462427f5e7cf3
Reviewed-on: http://gerrit.openafs.org/1771
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agomacos bulkstat again
Derrick Brashear [Wed, 21 Apr 2010 22:45:33 +0000]
macos bulkstat again

revamp this to avoid various issues with vnode references,
and to avoid potentially finding CVInit vcaches but not reclaiming

Change-Id: I6cb8d73b66c449a104799082259c97ede7e32e9b
Reviewed-on: http://gerrit.openafs.org/1802
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoFix UCONTEXT detection on ppc_linux26 via include order
Russ Allbery [Tue, 20 Apr 2010 05:01:27 +0000]
Fix UCONTEXT detection on ppc_linux26 via include order

param.linux26.h defines USE_UCONTEXT for all Linux platforms for
glibc 2.4 and higher, but it does this by testing __GLIBC__ and
__GLIBC_MINOR__.  These are defined by features.h, which is included
by any system header.  At least one system header must be included
before those are defined.  lwp/process.c was including <afsconfig.h>
and <afs/param.h> before any other headers, leading to those macros
being undefined.  Most of the Linux architectures either have their
own implementation or were explicitly defining USE_UCONTEXT in the
per-architecture param file, but ppc_linux26 was relying on the
default.

Fix this by reordering the includes to include the various system
headers before <afs/param.h> and add a comment explaining why.

This previously worked in earlier versions because the old
param.ppc_linux26.h file included <afs/afs_sysnames.h>, which
included "stds.h", which included a system header prior to the check
for ucontext.  The new generic param file reverses that order.

Change-Id: I214e54f684afdef07e6e2dec5b6e6e4d255ab4ec
Reviewed-on: http://gerrit.openafs.org/1791
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoRecover from afs_GetVolSlot errors
Andrew Deason [Wed, 21 Apr 2010 17:41:21 +0000]
Recover from afs_GetVolSlot errors

afs_GetVolSlot can panic in a few different ways, such as failing to
read from or write to VolumeInfo. Instead of panic'ing, return an
error to the application. Adjust callers to deal with getting a NULL
volume returned.

Based on a patch by Mike Meffie.

Change-Id: Ibb301fed795ee9ee5906b8e7973945a06218b8f1
Reviewed-on: http://gerrit.openafs.org/1801
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd RFC 5864 to the protocol documentation directory
Russ Allbery [Wed, 21 Apr 2010 03:46:32 +0000]
Add RFC 5864 to the protocol documentation directory

Add a copy of RFC 5864 (DNS SRV Resource Records for AFS) to the
protocol documentation directory for reference.  As permitted by
the IETF Trust License Policy section 3(e), I release this document
under the MIT/X Consortium license included in this copy of the
document.

LICENSE MIT

Change-Id: I8e22aac07b4cedbe18b8375213a7866cf98a1386
Reviewed-on: http://gerrit.openafs.org/1799
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agoInstall afsd.fuse and man page if built
Russ Allbery [Tue, 20 Apr 2010 04:18:57 +0000]
Install afsd.fuse and man page if built

If --enable-fuse-client is passed to configure and afsd.fuse is built,
install it into the same directory as afsd and install afsd.fuse.8 as a
symlink to the afsd.8 man page.  Add documentation of afsd.fuse to the
afsd man page.

Change-Id: I7d0cd3992a8466e626af2191c713e5623cc40d84
Reviewed-on: http://gerrit.openafs.org/1792
Tested-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd src/afsd/afsd.fuse and src/libuafs/afsd to .gitignore
Anders Kaseorg [Wed, 21 Apr 2010 08:09:03 +0000]
Add src/afsd/afsd.fuse and src/libuafs/afsd to .gitignore

Change-Id: I4b7e5261d92944ed871a1113b0f4a651357f9fcf
Signed-off-by: Anders Kaseorg <andersk@mit.edu>
Reviewed-on: http://gerrit.openafs.org/1800
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Tested-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd documentation for fs callback xstats
Andrew Deason [Sun, 18 Apr 2010 23:49:18 +0000]
Add documentation for fs callback xstats

Change I572ff682de4cc7ef27bb46dd028d3d797b873841 added the fileserver
callback xstats collection to afsmonitor. Provide some documentation
for these fields, along with the other fields displayed by afsmonitor.

Change-Id: I21618047519fbb28f6707ff9ba95a17fe27e0f3c
Reviewed-on: http://gerrit.openafs.org/1783
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agokauth admin tools ubik prototyping
Derrick Brashear [Tue, 20 Apr 2010 23:59:53 +0000]
kauth admin tools ubik prototyping

kill off ubik_Call (but not CallIter, yet)

Change-Id: I7af515b4e1c55c493c304a8ffd5c74cb96cd3b97
Reviewed-on: http://gerrit.openafs.org/1793
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Tested-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agoUpdated dkms.conf to remove MPS=$SMP from make
Jonathan Billings [Tue, 20 Apr 2010 19:53:21 +0000]
Updated dkms.conf to remove MPS=$SMP from make

* Removed the logic to set $SMP based on CONFIG_SMP from
  /boot/config-$kernelver
* When using --with-linux-kernel-packaging in the configure line, dkms
  no longer needs MPS=$SMP in the make line.

Change-Id: Ib65782050cae9f439ad5fed5c7bbbc403e556b4f
Reviewed-on: http://gerrit.openafs.org/1798
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdded explanatory comments for restorevol move
Jonathan Billings [Tue, 20 Apr 2010 16:13:11 +0000]
Added explanatory comments for restorevol move

Since restorevol is placed in %{_prefix}/afs/bin by make dest, we
need to fix it so it ends up in %{_bindir} to be used in the base
package.

Change-Id: I31f2602868230416525a84a9d716230f08cb34ab
Reviewed-on: http://gerrit.openafs.org/1796
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Tested-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agoFixed more locations to reflect recent changes
Jonathan Billings [Tue, 20 Apr 2010 16:00:21 +0000]
Fixed more locations to reflect recent changes

Moved restorevol man page to base package
Moved state_analyzer man page to server package

Change-Id: I714dd6eca86def48fefc0206c93038ae71415cc9
Reviewed-on: http://gerrit.openafs.org/1795
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Tested-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agoFixed RedHat SPEC file to reflect updates in code
Jonathan Billings [Tue, 20 Apr 2010 15:50:59 +0000]
Fixed RedHat SPEC file to reflect updates in code

Moved the restorevol binary from /usr/afs/bin to /usr/bin
Added the new state_analyzer.8.gz man page

Change-Id: Ibf4133034b1945d58843162cb45bdd485e8c34a1
Reviewed-on: http://gerrit.openafs.org/1794
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Tested-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agoRemove special-case call sequence for KAM_SetPassword on s390
Russ Allbery [Tue, 20 Apr 2010 04:36:09 +0000]
Remove special-case call sequence for KAM_SetPassword on s390

For Linux s390 (but not s390x), an additional argument was passed
to KAM_SetPassword between the kvno and the encryption key.  This
doesn't seem to match the rest of the code and is now, with stricter
prototyping, preventing the code from compiling.  Remove it and use
the same call sequence on s390 as everywhere else.

Change-Id: I750f64212e80d4bed4500e316ad7d354608866d7
Reviewed-on: http://gerrit.openafs.org/1790
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoUse AC_PREREQ
Andrew Deason [Mon, 19 Apr 2010 19:48:14 +0000]
Use AC_PREREQ

We use AC_USE_SYSTEM_EXTENSIONS, which was introduced in autoconf
2.60. To allow for less confusion and perhaps a more clear error
message, specify that we require using at least that version.

Change-Id: I42a3f4e889ad443d8d8cd3ba0c085c67606b8be3
Reviewed-on: http://gerrit.openafs.org/1784
Reviewed-by: Russ Allbery <rra@stanford.edu>
Tested-by: Russ Allbery <rra@stanford.edu>

14 years agoFixes and clarifications to vos setaddrs man page
Russ Allbery [Mon, 19 Apr 2010 21:00:31 +0000]
Fixes and clarifications to vos setaddrs man page

Add a caution explaining how the file server addresses are registered
and pointing users at NetInfo and NetRestrict plus restarting the file
server for the normal case.

Mention what version of OpenAFS introduced this command.  Drop the note
about the version of OpenAFS that added the -encrypt flag, since the
whole command is newer than that.

Reference vos listaddrs -printuuid specifically to get the UUID.

General formatting and wording cleanup: use terminology more consistently,
continue a long example line, wrap long lines, fix a spelling error, and
add cross-references to NetInfo and NetRestrict.

Change-Id: Idd6175339dc0feb1b777963bbb09731e42b83522
Reviewed-on: http://gerrit.openafs.org/1787
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoUpdate Debian packaging for 1.5.74
Russ Allbery [Mon, 19 Apr 2010 21:55:16 +0000]
Update Debian packaging for 1.5.74

Include the Debian changelog entry for 1.5.74-1 and install the new
vos setaddrs man page.

Change-Id: I4cc9e0590cb47d0ae885d7116f5bafd469f43cae
Reviewed-on: http://gerrit.openafs.org/1788
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agomacos disable bulkstat openafs-devel-1_5_74
Derrick Brashear [Mon, 19 Apr 2010 18:02:14 +0000]
macos disable bulkstat

there's still a bug

Change-Id: I23e02f9a05e001f0082b757c4a1dd78186fe6977
Reviewed-on: http://gerrit.openafs.org/1782
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agomacos bulkstat don't double reget refs
Derrick Brashear [Sun, 18 Apr 2010 17:45:51 +0000]
macos bulkstat don't double reget refs

if another code path already got back the refs from finalizevnode,
don't do it again

Change-Id: I7e06ceeb2c7e71b1be94fd0fccb6e28c37457b24
Reviewed-on: http://gerrit.openafs.org/1776
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

14 years agobulkstat restore bulkfetching state for reused vcaches
Derrick Brashear [Sun, 18 Apr 2010 17:35:09 +0000]
bulkstat restore bulkfetching state for reused vcaches

when i introduced NewBulkVCache I elided this case. Set it here.

Change-Id: Id88425b0ee662ecc23dd43cf7fa253ab61b02945
Reviewed-on: http://gerrit.openafs.org/1775
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoWindows: Preserve volume location info in case of comm fail
Jeffrey Altman [Sun, 18 Apr 2010 17:34:18 +0000]
Windows: Preserve volume location info in case of comm fail

The cache manager refreshes volume location information every
two hours. If during a refresh the communication with the
vldb server fails, the previously known volume location information
should continue to be used.

The previous behavior in which the volume location information
is discarded first and then the update is performed can result
in unnecessary client failures when a temporary disruption in
communication with the vldb server occurs.  Instead, wait until
we have a successful response from the vldb server before the
previous server list is discarded.

LICENSE MIT

Change-Id: I7c63c4f673a7a1360a74611c356329f31f9ceec3
Reviewed-on: http://gerrit.openafs.org/1769
Reviewed-by: Asanka Herath <asanka@secure-endpoints.com>
Tested-by: Asanka Herath <asanka@secure-endpoints.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: correct locking in cm_FreeServerList
Jeffrey Altman [Sun, 18 Apr 2010 17:20:52 +0000]
Windows: correct locking in cm_FreeServerList

The cm_serverLock must be held when checking the state
of the list head.

LICENSE MIT

Change-Id: I1134a6617d4c1e38b95512b1fdc41c5f5857ea7a
Reviewed-on: http://gerrit.openafs.org/1774
Reviewed-by: Asanka Herath <asanka@secure-endpoints.com>
Tested-by: Asanka Herath <asanka@secure-endpoints.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: prevent infinite idle error response
Jeffrey Altman [Sun, 18 Apr 2010 11:38:27 +0000]
Windows: prevent infinite idle error response

The idle error value (if any) is stored in the cm_req_t object.
Since the value is never cleared, the same value can be returned
for all requests that make use of the same cm_req_t object.

Change the behavior to only return an idle error once and then
clear it.

LICENSE MIT

Change-Id: Iac9ff8e56e856c2f0e440a34ac39ccd68de37c77
Reviewed-on: http://gerrit.openafs.org/1773
Reviewed-by: Asanka Herath <asanka@secure-endpoints.com>
Tested-by: Asanka Herath <asanka@secure-endpoints.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: Increase the memory map version
Jeffrey Altman [Sat, 17 Apr 2010 19:21:32 +0000]
Windows: Increase the memory map version

Since the cm_buf_t data structure changed, update
the memory map version number so that the cache will
not be reused.

LICENSE MIT

Change-Id: If01997ebd90c8f004d7cf4ba92f46fe4b9f570bf
Reviewed-on: http://gerrit.openafs.org/1772
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoopenafs 1.5.74
Derrick Brashear [Sat, 17 Apr 2010 14:14:50 +0000]
openafs 1.5.74

make 1.5.74 versions

Change-Id: I320cdbab2961dcf2a17bb71e8d4283aaf38941ce
Reviewed-on: http://gerrit.openafs.org/1768
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: 1.5.74 change log
Jeffrey Altman [Sat, 17 Apr 2010 15:13:21 +0000]
Windows: 1.5.74 change log

Important changes affecting the Windows distribution for 1.5.74

Change-Id: I980a66a2ab4a90c580249641fc22e3c7a91097dc
Reviewed-on: http://gerrit.openafs.org/1770
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoRx: make conn_call_lock and conn_data_lock usage consistent
Jeffrey Altman [Sat, 17 Apr 2010 02:42:34 +0000]
Rx: make conn_call_lock and conn_data_lock usage consistent

The rx_connection.flags field is protected by the conn_data_lock
but the conn_data_lock is not held everywhere the conn flags
field is altered.  This produces a race that can result in a
deadlock when waiter flags are inadvertently prevented from being
cleared.

The conn_call_lock usage in rx_EndCall which was removed in
Change e169708681eb1bbbb31951b95f68e861a4b01c7e must be restored.
If rx_EndCall never obtains the conn_call_lock it can't ensure
that the thread in rx_NewCall actively checking the calls will
not end up blocking when there is now a call channel that can
be reused.  This usage of conn_call_lock can be removed only
if a true producer/consumer model is implemented.

Change-Id: Id093fd6c4a42afb833c775411be0cd406abf4ef7
Reviewed-on: http://gerrit.openafs.org/1766
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoWindows: cm_UpdateCell must hold cell lock across server random
Jeffrey Altman [Sat, 17 Apr 2010 02:52:20 +0000]
Windows: cm_UpdateCell must hold cell lock across server random

cm_UpdateCell fails to hold the cell lock across the server
randomization.  As a result the vlserver list can be destroyed
while randomization is taking place.

LICENSE MIT

Change-Id: I7753e7e9f8a7d536ce0e2082aaca7b62ab745504
Reviewed-on: http://gerrit.openafs.org/1767
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: CM_SCACHESYNC_STOREDATA for non-files have no buffers
Jeffrey Altman [Fri, 16 Apr 2010 01:43:44 +0000]
Windows: CM_SCACHESYNC_STOREDATA for non-files have no buffers

Do not add QData objects with null cm_buf_t pointers to the
cm_scache_t bufWritep queue when synchronizing directory changes.
If a callback is required while the directory change is being
pushed it can result in a deadlock.

LICENSE MIT

Change-Id: Id66c52ed794ceec44011167659e41a0b61898625
Reviewed-on: http://gerrit.openafs.org/1761
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: define new event log messages for cm_Analyze VBUSY, VRESTARTING, etc.
Jeffrey Altman [Fri, 16 Apr 2010 04:31:58 +0000]
Windows: define new event log messages for cm_Analyze VBUSY, VRESTARTING, etc.

Add MSG_SERVER_REPORTS_VBUSY, MSG_SERVER_REPORTS_VRESTARTING,
MSG_ALL_SERVERS_BUSY, MSG_ALL_SERVERS_OFFLINE,
and MSG_ALL_SERVERS_DOWN.

Add event message throttling.  Only permit one copy of a message
to be generated every five seconds if the message will duplicate
the prior message.  This often occurs when a server or volume becomes
inaccessible and there were a large number of requests queued on it.

Integrate these new messages into cm_Analyze processing for VBUSY,
VRESTARTING, ALLDOWN, ALLOFFLINE, and ALLBUSY errors.

LICENSE MIT

Change-Id: Ib30595ded23724810e294133fa75873e20ffbf0b
Reviewed-on: http://gerrit.openafs.org/1762
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: wait for I/O on buffers to complete in cm_SetupStoreBIOD
Jeffrey Altman [Fri, 16 Apr 2010 04:09:25 +0000]
Windows: wait for I/O on buffers to complete in cm_SetupStoreBIOD

cm_SetupStoreBIOD constructs a list of dirty buffers for a file
that are to be written to the file server.  When constructing
the list, if when determining the first dirty buffer we come across
a buffer that is already actively involved in an I/O operation,
call buf_WaitIO() to wait until the buffer is no longer busy before
continuing.  This reduces lock contention and synchronization
conflicts.

LICENSE MIT

Change-Id: Icd79b9578b7411b138f14257bcb885249960eab0
Reviewed-on: http://gerrit.openafs.org/1760
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoWindows: split cm_buf_t.flags field to ensure proper locking
Jeffrey Altman [Fri, 16 Apr 2010 03:58:21 +0000]
Windows: split cm_buf_t.flags field to ensure proper locking

It turns out that for all these years the locks protecting
the cm_buf_t flags field have been racy.  Some of the flags
were protected by the cm_buf_t mutex and others by the
buf_globalLock.  This patchset splits the flags field so that
the appropriate lock will be used exclusively to protect a
common set of flags.

LICENSE MIT

Change-Id: I85c98c85244c987df30a811d405f7d897e407ba2
Reviewed-on: http://gerrit.openafs.org/1759
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoRemove unused variable in vos.c
Marc Dionne [Fri, 16 Apr 2010 23:34:35 +0000]
Remove unused variable in vos.c

Remove an unused variable introduced by a recent commit.

Change-Id: I04432d87a27b656807430f041f84d6ada2632434
Reviewed-on: http://gerrit.openafs.org/1765
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

14 years agowork around finder setting modes on symlinks
Derrick Brashear [Fri, 16 Apr 2010 15:49:43 +0000]
work around finder setting modes on symlinks

we could mask the mode setting on symlinks, however, it would be nice
to change the fileserver to allow mode setting on symlinks in some
(safe) cases. preserve our ability to do so.

Change-Id: Iba69965c607530bcf2210b508c7aa37403c47477
Reviewed-on: http://gerrit.openafs.org/1764
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd vos setaddrs command and man page
Harald Barth [Fri, 16 Apr 2010 05:45:35 +0000]
Add vos setaddrs command and man page

The vos setaddrs command sets the IP addresses for a server entry
in the Volume Location Database (VLDB). Specify one or serveral hosts.
All existing hosts in the VLDB entry are replaced with the new entries
on the command line.

Change-Id: I3c26e49c4a6e2aebae363017d074329ac265132a
Reviewed-on: http://gerrit.openafs.org/1744
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoConvert param.rs_aix61.h EOF from CR-LF to LF
Jeffrey Altman [Fri, 16 Apr 2010 13:32:42 +0000]
Convert param.rs_aix61.h EOF from CR-LF to LF

The src/config/param.rs_aix61.h source file was stored
in the repo with CR-LF end of line.  This is causing
problems for Windows Git which converts CR-LF to LF
for storage in the repo.

Change-Id: I8937322ac368ac275d3dbc6bcccf2808d3db061f
Reviewed-on: http://gerrit.openafs.org/1763
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

14 years agoTidy up UKERNEL includes
Simon Wilkinson [Thu, 15 Apr 2010 19:52:11 +0000]
Tidy up UKERNEL includes

UKERNEL is just another userspace build - there's no need to
maintain completely separate header file lists in each object file
for "userspace" and "ukernel". Tidy this up to improve the
readability of these sections of code.

Change-Id: I69f476a0b8aae1204cd4207c7c656ec7e07184df
Reviewed-on: http://gerrit.openafs.org/1758
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoRx: restore thread safety to rx_NewCall
Jeffrey Altman [Thu, 15 Apr 2010 03:21:37 +0000]
Rx: restore thread safety to rx_NewCall

Thread safety in rx_NewCall requires that only one thread be
actively allocating or recycling a call at a time.  Since we are
no longer holding the conn_call_lock across the entire transaction
we need to have another synchronization mechanism.  Add a new
rx_connection flag, RX_CONN_MAKECALL_ACTIVE, which when set indicates
that a thread is actively obtaining a call.  If any other threads
see this flag set, they will wait until being signalled that the
thread has completed its activity.

In addition, because the call->lock may be dropped when processing
rxi_ResetCall(), we must hold a reference to the call once we
begin using it.  Otherwise, the call may be garbage collected
behind our back.

Change-Id: Ie97757f812d7043203ffcaf399400789cda39da1
Reviewed-on: http://gerrit.openafs.org/1755
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agomacos and bsd readdir type guessing fix
Derrick Brashear [Wed, 14 Apr 2010 16:10:31 +0000]
macos and bsd readdir type guessing fix

right now we never return a link hint. we know here it's a link. say so.

Change-Id: Iab81b83fc8de714ad52c99c86153159aa6f641f3
Reviewed-on: http://gerrit.openafs.org/1753
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agodarwin bulk race needs an iocount
Derrick Brashear [Thu, 15 Apr 2010 04:13:43 +0000]
darwin bulk race needs an iocount

if another user raced with us, we need an iocount back since finalize
will have dropped one.

Change-Id: I93b2bb5b6004b39436684e78c4620f4d7d9d0c8e
Reviewed-on: http://gerrit.openafs.org/1756
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agosnowleopard fuse fallout
Derrick Brashear [Thu, 15 Apr 2010 05:29:31 +0000]
snowleopard fuse fallout

didn't notice in testing that 64 bit version of afsd got slaughtered.

put it back.

Change-Id: I3a6588df36a31ab014d79faf4ed5646339f6f84b
Reviewed-on: http://gerrit.openafs.org/1757
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoReplace unsafe use of gets()
Marc Dionne [Tue, 13 Apr 2010 23:35:39 +0000]
Replace unsafe use of gets()

Build tools compain that this is dangerous - replace gets() with a
bounded fgets().

Change-Id: I3bd1854a1dc4a11c801cabb987ab680fa978bd20
Reviewed-on: http://gerrit.openafs.org/1750
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agodrop afs_xserver lock during GetCapabilities
Derrick Brashear [Wed, 14 Apr 2010 22:33:47 +0000]
drop afs_xserver lock during GetCapabilities

new contact to a fileserver can trigger an InitCallBackStateN RPC
to us, which our agent will need afs_xserver to handle. don't hold it;
we only need it to fill in capabilities. racing here is ok.

Change-Id: Ie0aaea3ab462e421bd31ba3b703d8cd0cb0d61df
Reviewed-on: http://gerrit.openafs.org/1754
Reviewed-by: Marc Dionne <marc.c.dionne@gmail.com>
Tested-by: Marc Dionne <marc.c.dionne@gmail.com>
Reviewed-by: Andrew Deason <adeason@sinenomine.net>
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAutogenerate a Debian changelog for correct package versioning
Russ Allbery [Tue, 13 Apr 2010 23:32:49 +0000]
Autogenerate a Debian changelog for correct package versioning

The Debian packaging infrastructure takes the package version from the
most recent entry in the changelog file.  Change the changelog file to
a template and add an entry to the top that will be set to the current
version of OpenAFS, with a Debian revision that will sort before any
official package.

Change-Id: Ic7fb1d30be10210b7d032cdc49a963410997db82
Reviewed-on: http://gerrit.openafs.org/1749
Tested-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd a FUSE implementation for afsd
Andrew Deason [Thu, 8 Apr 2010 19:50:18 +0000]
Add a FUSE implementation for afsd

This adds afsd.fuse, which allows for mounting AFS via FUSE (via
libuafs), instead of via the OpenAFS kernel module.

Change-Id: Iaafe4a5f3034fed943e2e73f79ac95580946f9a8
Reviewed-on: http://gerrit.openafs.org/1725
Tested-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoUse afsd code in libuafs
Andrew Deason [Tue, 6 Apr 2010 22:07:33 +0000]
Use afsd code in libuafs

Share the same CM code for the kernel client as in libuafs, so we
don't duplicate code for initializing the cache and other things. In
order to do this:

 - Remove some libuafs global variables that share name and
   functionality with those in afsd, and declare some static

 - Remove uafs_Init(), and move the ukernel-specific code in it to
   osi_Init(); replace with uafs_Setup(), uafs_ParseArgs(), and
   uafs_Run(), which just call into afsd functions

 - Remove libuafs' cache initialization code (CreateCacheFile,
   SweepAFSCache, etc); instead just use afsd's

 - Add uafs_mount(), to perform the 'mount'ing step that takes place
   in the normal kernel CM

 - Add afsd_uafs.c for the glue between afsd and libuafs

Note that this now requires libcmd for libuafs.

Change-Id: I62306a18ad255680031494caf1720b29e22856d2
Reviewed-on: http://gerrit.openafs.org/1724
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoRename uvldbentry spare1 to matchindex
Jeffrey Altman [Mon, 12 Apr 2010 18:45:14 +0000]
Rename uvldbentry spare1 to matchindex

In the nvldbentry structure 'spare1' was renamed to 'matchindex'.
Do the same for uvldbentry as that is what the field is used for.

Change-Id: I2d1200b1044f6ae9b4523eb603a21b0484b046c7
Reviewed-on: http://gerrit.openafs.org/1741
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoFix new UKERNEL warnings on 64-bit
Marc Dionne [Tue, 13 Apr 2010 22:58:11 +0000]
Fix new UKERNEL warnings on 64-bit

Commit 830cb48c enabled new warnings when building UKERNEL, which
causes builds with --enable-checking to fail.  These are 64-bit
specific warnings from int to pointer conversions and one printf
warning.

Changes:
- cast printf argument to (int) in afs_usrops.c
- use (iparmtype)(uintptrsz) to convert 32-bit integers to
pointers
- move the definition of uintptrsz to src/afs/afs.h so its
available to other source files, and remove the original definition
in afs_syscall.c

Change-Id: Id0a5fd762cf10a741f89e3a2ed0a85556137df49
Reviewed-on: http://gerrit.openafs.org/1748
Reviewed-by: Andrew Deason <adeason@sinenomine.net>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoAdd support for pkg-config m4 macros
Andrew Deason [Tue, 13 Apr 2010 18:53:13 +0000]
Add support for pkg-config m4 macros

Include pkg.m4, so we can use pkg-config macros in autoconf even if
pkg-config is not installed.

Change-Id: Ie81307beb84499297e251f0f90781d544044a235
Reviewed-on: http://gerrit.openafs.org/1747
Reviewed-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoSquash afs_GenericStoreProc warning
Andrew Deason [Tue, 13 Apr 2010 17:59:30 +0000]
Squash afs_GenericStoreProc warning

code is potentially unitialized. Initialize it.

Change-Id: I9c8a787d9f7f05ea88752063f1b31e0be61e28ba
Reviewed-on: http://gerrit.openafs.org/1746
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agolibuafs: Add $(XCFLAGS) to CFLAGS on all platforms
Andrew Deason [Tue, 13 Apr 2010 17:41:31 +0000]
libuafs: Add $(XCFLAGS) to CFLAGS on all platforms

Only darwin had $(XCFLAGS) in the libuafs build for some reason. Add
it to all platforms, so they get e.g. -Wall if so configured.

Change-Id: Ifab39af3abf087932feb1b9085ccb26baaec3164
Reviewed-on: http://gerrit.openafs.org/1745
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoDAFS: avoid volume lock contention during initialization
Michael Meffie [Tue, 12 Jan 2010 02:16:06 +0000]
DAFS: avoid volume lock contention during initialization

Avoid the excessive volume lock contention during startup to
improve the time to pre-attach a very large number of volumes.
The parallel attach worker threads avoid the volume lock
while scanning the partitions for volumes and send batches of
volume ids to the main thread to be preattached under the
volume lock.

FIXES 124489

Change-Id: Ieb33a3bdd5b06349abd9c3dd994c620021cd6194
Reviewed-on: http://gerrit.openafs.org/1092
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

14 years agoFileserver capabilities support for the UNIX client
Felix Frank [Thu, 4 Mar 2010 03:41:15 +0000]
Fileserver capabilities support for the UNIX client

The attached patch has the client perform a GetCapabilities RPC
on fileservers it encounters.
It uses an additional server flag bit to keep track of the servers that
have been queried already.

In the case of afs_CeckServers(), GetTime RPCs are largely replaced by
GetCapabilities. GetTime is performed on a server if and only if
afs_setTime is nonzero and either
(a) no setTimeHost has yet been determined or
(b) the server in question has been designated as setTimeHost
The GetServers() function could thus be simplified even further wrt. the
setTime mechanism, but doing so would imply more rewriting (violating
the KISS principle; a followup patch should deal with that).

When a client is asked to reset callback states, it also resets the
"capabilites known" bit.

Thanks go to Simon Wilkinson, Jeffrey Altman and Jeffrey Hutzelman for
input regarding logic and implementation details.

FIXES 124972

Change-Id: I4e22db3c5ca71f00e8c77f337e38daa0fa5b9124
Reviewed-on: http://gerrit.openafs.org/1640
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>