Windows: Save Wix config at start if possible

Save the current configuration at the start of the install
process so the user can be presented with a dialog prior
to installation asking whether the existing or saved
configuration should be used or whether a new configuration
should be created.

Reviewed-on: http://gerrit.openafs.org/5281
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 020b415be8fa357cded6eb4c50454aaa5b0722bf)

Change-Id: I49abfa6d6e7b929e3880686f7dc7bad4175adcba
Reviewed-on: http://gerrit.openafs.org/5284
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

Windows: add assertions to cm_scache.c

Reviewed-on: http://gerrit.openafs.org/5280
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 1a0b99e654831fe28794f8e0cec2ab94f07ec965)

Change-Id: I5bcce8d961157ebb85049871bd15404d7a5ed94e
Reviewed-on: http://gerrit.openafs.org/5283
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

Windows: cm_ShutdownSCache corrections

Obtain cm_scache.dirlock, cm_scache.rw and cm_scacheLock
in the correct order.

Do not release cm_scache.rw when it is not held.

Since the cm_scacheLock is being dropped, preserve the value
of scp->allNextp prior to dropping the lock.

Reviewed-on: http://gerrit.openafs.org/5279
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit c58ae34f72d25434cd59228806ca59a6ff39a903)

Change-Id: I65f3c4be5436031cdd99d912b0277b54d4033bf6
Reviewed-on: http://gerrit.openafs.org/5282
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

Windows: More interlocked ops for cm_cell flags

Reviewed-on: http://gerrit.openafs.org/5274
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 73b91bf0c0a86e55d6879573aeadd99a137ebf3b)

Change-Id: I0d79599b22c50523f975d51de4fe2a9eb07c1fdb
Reviewed-on: http://gerrit.openafs.org/5276
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Interlocked ops for cm_user flags

Reviewed-on: http://gerrit.openafs.org/5273
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit f75214282795b23c433b2512b00d24b3e6166b76)

Change-Id: I092a0d74919b1d8ecf62fbb57c5b957f16184a05
Reviewed-on: http://gerrit.openafs.org/5275
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

CellServDB update 14 Aug 2011

Reviewed-on: http://gerrit.openafs.org/5270
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit a5d66d05fa0308d505de8bde59442e29be9d04f8)

Change-Id: I511bd54512c30dced28d8a0bc79424a5b82d91cb
Reviewed-on: http://gerrit.openafs.org/5271
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: cm_FindServerByUuid correct lock acquisition

Obtain cm_serverLock when 'locked' is FALSE instead
of when TRUE.

Reviewed-on: http://gerrit.openafs.org/5268
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit b97383c1e68a0b94baa8d05b2ab2531f96e7b63c)

Change-Id: I1fb188fa493554dce366cbc4b84a32528486829d
Reviewed-on: http://gerrit.openafs.org/5269
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: prevent cm_server races

Use interlocked operations to modified the flags field.

Close a race in cm_NewServer() which can result in multiple
cm_server objecs being created for the same addr/port/type
tuple.

Reviewed-on: http://gerrit.openafs.org/5266
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit c20010850e2b43a0b9a6d74357ea6111dbf3409e)

Change-Id: If13f23a921f598db097b391348513a3300d3b10c
Reviewed-on: http://gerrit.openafs.org/5267
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Insert Server Reference List changes

When inserting a new cm_serverRef_t object into a server list
perform the following operations:

1. take advantage of the fact that the cm_serverLock is held
   exclusively to purge the list of any deleted entries that
   could not be removed previously.

2. check to ensure that the item that is being added does not
   already exist in the list.  If it does, discard it.

Reviewed-on: http://gerrit.openafs.org/5258
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit 5b40c5f5294964fc09df2c8332ab63cd2d729264)

Change-Id: I94398476267dda82e82306a87aa0bba6aa41da00
Reviewed-on: http://gerrit.openafs.org/5264
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

Windows: Fix cm_serverRef ref counts

Use Interlocked operations consistently

Simplify cm_ServerInsertList().  It no longer increments the
refCount on the serverRef object.  Instead it leaves the refCount
as is.  Its the caller's responsibility to add a reference if
required.

Add reference counts and hold locks in places where the
volume server list was used unprotected.

Reviewed-on: http://gerrit.openafs.org/5248
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit 8f446c7463c9183d59a30343682e31ad9f85b307)

Change-Id: I0ed8ea1551527e0d62e57967da6816415a3b36b5
Reviewed-on: http://gerrit.openafs.org/5254
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

libafs: don't call afs_PutDCache(NULL) in afs_GetDownD()

It's possible for an eviction candidate to be omitted by the small
for loop (around line 670), leaving its reference in victimDCs
set to NULL.  In the big for loop that follows, don't call
afs_PutDCache() when we hit one.

Found-by: clang static analyzer with help from AFS_NONNULL
Reviewed-on: http://gerrit.openafs.org/5260
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit 10d27341808be41b29fbcc09b8bd7523c3b7a541)

Change-Id: I4ff3ab9a57c1c48be0f7533ab3adb1e4b6ababf9
Reviewed-on: http://gerrit.openafs.org/5263
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

Windows: remove unused variables in fs.c

Reviewed-on: http://gerrit.openafs.org/5247
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit 578db3bbecf218e3ab80f4896b7cde4d3975daaa)

Change-Id: I8c334239a888d88422ca65d3372228417d60657a
Reviewed-on: http://gerrit.openafs.org/5253
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

dir: Protect against circular hash chains

The dir package didn't protect against circular hash chains when
performing directory lookups. A corrupt directory could therefore
cause a client or a fileserver to go into an endless loop if that
directory contained a loop in its hash chain pointers.

Fix this by exiting the lookup if the hash chain has more elements
than the total number of entries in a directory. This maximum number
of entries is taken as being (number of entries per page) * (max
number of pages), which is considerably more than the real maximum
value.
(cherry picked from commit bb25bdfcb059fc54a57fd4733ce3184e231ca88d)

Change-Id: I7290dbaad66bccfe3b03a843184464f0681f9429
Reviewed-on: http://gerrit.openafs.org/5250
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

libafs/dir: Verify directory pathnames

Provide a new routine, GetVerifiedBlob() which will ensure that the
pathname contained within a directory blob is correctly terminated
before returning it to the caller. For the purposes of this function,
correct termination is defined as having a terminating \0 character
within the same directory page as the blob itself.
(cherry picked from commit d1946ffe9be0031a2daf907f5e96cf0ee7f5e15e)

Change-Id: I69b9465f02417babf9b1d5179197278fac64f192
Reviewed-on: http://gerrit.openafs.org/5249
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

aklog: attempt to warn about needed weak crypto switch for Lion

Lion's Kerberos is rather unfortunate. deal with the multitude of
missing functionalities by hardcoding this case here.

Reviewed-on: http://gerrit.openafs.org/5240
Tested-by: Derrick Brashear <shadow@dementix.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit 145a8490c761a0ebf7b3a1cc3017bbe8635f8014)

Change-Id: I0efe2ee18e36f0531de8261e5f032df7de3d7333
Reviewed-on: http://gerrit.openafs.org/5252
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

redhat: update dkms config

make the dkms config properly name the version. remove no-longer-needed
disconnected option for configure

FIXES 130170

Reviewed-on: http://gerrit.openafs.org/5246
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>
(cherry picked from commit ef3ef1992115e6b61a7859fa2295726af0fb9318)

Change-Id: If0bfb3fdf4d8499dfe14435739317f335a4c164e
Reviewed-on: http://gerrit.openafs.org/5251
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

libafs: Get rx conn ref with afs conn ref

When we get a reference to an afs_conn with afs_Conn and its variants,
we assume we can use the tc->id rx connection without holding any
locks. However, if tc->forceConnectFS gets set, the tc->id connection
can be destroyed and recreated out from under us. So, to avoid using a
possibly freed rx connection, grab a reference to the rx connection at
the same time as we grab a reference to the afs conn. And also put
back the same reference with afs_PutConn.

(cherry picked from commit 03f0c656c1734b9be4debdf19b8f10771ff4420a)
Reviewed-on: http://gerrit.openafs.org/4625
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Change-Id: If19a267e23941e3305f4f13cd91a0935d214f1b1
Reviewed-on: http://gerrit.openafs.org/5232
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

1.6.0 release notes

add release notes for 1.6.0

Change-Id: Ie1f7c025b45d333c5bd7d17981a7d36e2992107e
Reviewed-on: http://gerrit.openafs.org/5245
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

Windows: ChangeLog for 1.6.0 (final)

Reviewed-on: http://gerrit.openafs.org/5239
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 6f725c441529ccd1616adeda19c7c0b43dcfd9cb)

Change-Id: I40e3c8ccb6b490fe02b22995f1e2341c0e3048ba
Reviewed-on: http://gerrit.openafs.org/5244
Reviewed-by: Derrick Brashear <shadow@dementix.org>
Tested-by: Derrick Brashear <shadow@dementix.org>

libafs: Avoid duplicate afs_Analyze in bulk stat

In afs_DoBulkStat, we can call afs_Analyze multiple times for the same
set of connection objects. Since afs_Analyze puts its reference to the
given afs_conn and rx_connection structures, calling it more than once
can cause the reference counts on those objects to be lower than they
should be.

Instead of making another afs_Analyze call, just alter the error code
inside the normal do/while afs_Analyze loop, so the 'loop' afs_Analyze
call gets the appropriate error code from the first bulk stat'd entry.

(cherry picked from commit ef28bc08c18e750f6100535665d5258a317a0a2b)
Reviewed-on: http://gerrit.openafs.org/5086
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Change-Id: I8bb484aade39830a56cb967a3e3167e63777e333
Reviewed-on: http://gerrit.openafs.org/5127
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementix.org>

doc: add old release notes

pull in old release notes

Change-Id: I2865c1e7a6490bf11e77d63d2acd7f2bdf99b150
Reviewed-on: http://gerrit.openafs.org/5243
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

Windows: Interlocked operations for cm_buf

cm_buf flags and qFlags

Separate flags and qFlags in the cm_buf structure to improve
performance.

Reviewed-on: http://gerrit.openafs.org/5197
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 7ab34058120ebcc218e4061ea3ac3c8eeca6d83e)

Change-Id: I8cbc806833af630fb56c8ef388fe3a21df1f5478
Reviewed-on: http://gerrit.openafs.org/5218
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: implement InterlockedAnd/Or for X86 Debug

Reviewed-on: http://gerrit.openafs.org/5191
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit ec1295d11b5e75fa942fe8c3a0f16309c5486ed4)

Change-Id: Icd6b76abec2c9a677b5e344ef36587bfbe696a43
Reviewed-on: http://gerrit.openafs.org/5212
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: Interlocked for cm_buf cmFlags

Reviewed-on: http://gerrit.openafs.org/5196
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 4876a416d2e59e87f9da8834db1ea4770f2c618c)

Change-Id: I7416e0be7f0de82eddc366dbdb77e94ceabab797
Reviewed-on: http://gerrit.openafs.org/5217
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: fix tptserver director creation

Reviewed-on: http://gerrit.openafs.org/5195
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 8661c9b6710322195b00de6ae03b3172712104aa)

Change-Id: Ie762b896e68028053a8746181f2086af57e02d0d
Reviewed-on: http://gerrit.openafs.org/5216
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: Interlocked ops for cm_volume

Use Interlocked operations for protection of cm_volume flags and
qFlags as well as cm_vol_state flags.

Reviewed-on: http://gerrit.openafs.org/5194
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit cbd075a36000d0b54b64eb7d9736587b27a08e9e)

Change-Id: Ib2b11cbadcbdbc244866d6bf90ab4d55ef589819
Reviewed-on: http://gerrit.openafs.org/5215
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: Interlocked ops for cm_scache

Use Interlocked operations for cm_scache flags and mask field
changes.

Reviewed-on: http://gerrit.openafs.org/5193
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 75297d4439bb30ddd9968805aecf2cc2812bfdf0)

Change-Id: Ic10b726eb5791c7c8d3eb7c3ad529522f627f73c
Reviewed-on: http://gerrit.openafs.org/5214
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: Interlocked ops for cm_cell

Use Interlocked operations for cm_cell flag updates.

Reviewed-on: http://gerrit.openafs.org/5192
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit deb8c893c2bd78b2871ccf1ff733539534a2890e)

Change-Id: I3d0a31e9f809aa91009a8cb1256937ab9646818a
Reviewed-on: http://gerrit.openafs.org/5213
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: make osi_Log macro safe for if..else

wrap the osi_Log macro's internal if statement with
a do {...} while(0) block in order to ensure that
it is safe for use in if..else controls without bracing.

Reviewed-on: http://gerrit.openafs.org/5189
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 6691ff6daceb7960dc925983a2b9129877e67c9a)

Change-Id: I954ea5a567bbfd77b02d3e8634a64f877668280e
Reviewed-on: http://gerrit.openafs.org/5211
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: adjust scache LRU postion upon deletion

If the object represented by a scache object is deleted,
update the LRU position of the scache object to make it
the first object in the LRU queue to be recycled.  This
preserves the cached objects for those that might prove
useful in the future.

Reviewed-on: http://gerrit.openafs.org/5161
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit c5126838590eaf15e0df569c31b2a1bc12b3e0af)

Change-Id: Icf60ad3127c295e4fca0e99e338ba4c62cd9e392
Reviewed-on: http://gerrit.openafs.org/5210
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: LockOrderValidation memory usage optimization

Instead of using malloc() and free() to allocation lock reference
structures, cache allocated objects in a free list.  This reduces
memory fragmentation.

Reviewed-on: http://gerrit.openafs.org/5159
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 17e50a4b98de057a699681261ba126a5ac12a0cc)

Change-Id: Ifc8d7b6a92e32557575e24cec7513f890439ce67
Reviewed-on: http://gerrit.openafs.org/5209
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: after dir enum adjust dir scache LRU

During a directory enumeration the directory scache object
is reference counted so it can't be recycled.  However, if
there are more directory entries than the maximum number
of cached scache objects the directory scache object will
end up being the next object to be recycled after the refcount
is dropped.  Since the directory is clearly a hot object, before
dropping the reference, adjust the scache LRU position so that
it is the last object to be recycled.

Fix the variable name for the directory scache to be 'dscp'
for consistency.

Reviewed-on: http://gerrit.openafs.org/5158
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 2bce3b50ef4fd9b9bdd03daf6e8332710f541922)

Change-Id: Ie4400d769a7ac2d0bfed3ccebe02760619bb76b2
Reviewed-on: http://gerrit.openafs.org/5208
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: use %p to print cm_scache_t pointers

Reviewed-on: http://gerrit.openafs.org/5152
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 4111ee2374d030e1f3bb1fa3530f2877f2576d07)

Change-Id: If4ff7b219cce5c631666e0843123f6c7d8dbcec6
Reviewed-on: http://gerrit.openafs.org/5207
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

RedHat: Return status values from client init

The init script provided with OpenAFS always returns 0 when the status
subcommand is called, even if the service is not running.

For example:

$ sudo service afs status; echo $?
afsd is stopped
0

This change makes sure the init script exits with the value returned
by the status function from /etc/init.d/functions. With this patch,
the afs init script behaves as expected when used, for example, in a
Chef service resource:

$ sudo service afs status; echo $?
afsd is stopped
3

Reviewed-on: http://gerrit.openafs.org/5123
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 6b83ba621b520174dc4679c220888619dc0b44a0)

Change-Id: I4103875caba88d61c3edf480f3a7e0b24227a319
Reviewed-on: http://gerrit.openafs.org/5231
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

SOLARIS: Do not release NULL root vp on unmount

When we unmount, and afs_globalVp is NULL (e.g. because root.afs was
unavailable when the client was started), we will panic the machine if
we try to release it. So, if afs_globalVp is NULL when we hit our
unmount handler, don't touch it.

Reported by Andy Cobaugh.

Reviewed-on: http://gerrit.openafs.org/5117
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
(cherry picked from commit b621a2986099488426a026818532a9600db9aeba)

Change-Id: Iaa5db97e9fc87d1eb083d20a9fc6a49cbb5a066b
Reviewed-on: http://gerrit.openafs.org/5230
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

vos: spell "vldb" correctly

i'm not even going to ask.

Reviewed-on: http://gerrit.openafs.org/5125
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit d54c9b05d003bf861fd51e904b631e5425a079d6)

Change-Id: I0188a58363657a0ccf48afef79bf63ff4c4b4f9d
Reviewed-on: http://gerrit.openafs.org/5229
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

volser doesn't depend on tviced, but on vlserver

Nothing within the volser/ directory depends on tviced, so remove the
unecessary dependency. Add an explicit dependency on vlserver, so that
libvldb is available to us.

This is required to get rid of some potential circular loops when we
start including volser objects in libafsauthent

Reviewed-on: http://gerrit.openafs.org/5156
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 400c72561e7622716a46a38affd2f5a7842519eb)

Change-Id: I0ff6463cc63093f1c82ed74523256cbdb7fc75c8
Reviewed-on: http://gerrit.openafs.org/5228
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

vos: don't free stack garbage on error

If wantExtendedInfo is true, then pntr is used uninitialized.
In the other case, UV_ListVolumes will have set it to NULL
before doing anything (even if it returns an error), so this
free() is dead anyway.

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5164
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 619b420903c99e48618db0d2f12085111573f279)

Change-Id: Icd76d3ac160ccb04b05445552a9e6d04f053efbb
Reviewed-on: http://gerrit.openafs.org/5227
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

butc: avoid testing stack garbage; remove dead initializer

"code" is unconditionally set early in saveDbToTape() so there's
no need to initialize it.  On the other hand, dumpEntry.id is used
before dumpEntry is initialized, so set it to what appears to be
the expected value before any non-local exits could cause it to be
inspected.

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5166
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit c92f04af6094ce04b1541a1b23b254d3c1421290)

Change-Id: If99aaf327749550623ef5abc3dd077f5291bc899
Reviewed-on: http://gerrit.openafs.org/5226
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

rx: avoid nat ping until connection is attached

drop nat pings on connections we haven't talked on yet

Reviewed-on: http://gerrit.openafs.org/5130
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 496fb87372555f6acddd4fd88b03c94c85f48511)

Change-Id: Id64a771f9cb50191a665f5ccb98ec66d991f47bb
Reviewed-on: http://gerrit.openafs.org/5225
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

butc: avoid freeing uninitialized pointer in writeDbDump()

In error conditions, charList could be freed before it is initialized.
Move the initialization up to before the error checks.

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5167
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 43834bff1a2b1af348ff69d538a884bf1070b90c)

Change-Id: I05d1602b3f59b521d7daa3fd23f6609bf14c460e
Reviewed-on: http://gerrit.openafs.org/5224
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

afsd: look in the right place for -splitcache argument

The argument to -splitcache is in as->parms[34], not [30].

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5169
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 90ea68979c6740583747d0af500ed4a034eba651)

Change-Id: Iff8072cf8c7b5d9a5d7486fc9778f079fab1426c
Reviewed-on: http://gerrit.openafs.org/5223
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

bos: don't dereference a null pointer when printing an error message

The parameter we are interested in is at MRAFS_OFFSET + 17, not
MRAFS_OFFSET + 13.

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5178
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit bf4b2fd3e26dcf2a0db704815e05d77a558d38c6)

Change-Id: I92db9e5ec10054cef4421ab1b2fd0f60baaf07ca
Reviewed-on: http://gerrit.openafs.org/5222
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

tbudb depends on tubik ...

... so say so in the Makefile

(cherry picked from commit 88725587e97d717086801b2522e4625e5fdeb3a8)
Reviewed-on: http://gerrit.openafs.org/5155
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Change-Id: Ibe759403533cd4b2d538cf1cd905c6268a341f0f
Reviewed-on: http://gerrit.openafs.org/5235
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

macos: axe static vfs_fsentry

for whatever reason, lion 32 bit doesn't like it when this is static.
fine, so it's not static now.

Reviewed-on: http://gerrit.openafs.org/5206
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
(cherry picked from commit d18ff03b59cf26423795f735decdcaf13097c446)

Change-Id: I3949fcb69e14c00e0ade2826345be489426d7a04
Reviewed-on: http://gerrit.openafs.org/5220
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

viced: avoid aborting on host table exhaustion

if we exhaust the host table, instead of aborting, return VBUSY at
the client, to defer until hopefully hosts are freed.

Reviewed-on: http://gerrit.openafs.org/5181
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit a3b4cd3c7325c7878ec7b47008f0eec8bfcb1175)

Change-Id: Ic761e487ef0cfe402103bf06bd051958b07641b9
Reviewed-on: http://gerrit.openafs.org/5219
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

rpm: Set executable bits on installed libraries

Set the executable bits on the libraries installed in libdir.  This
change is important because it causes 'rpmbuild' to generate Provide
tag metadata for the libraries in the package, which is necessary now
that some binaries in other packages have generated Requires tags for
libraries packaged in the base package.  'rpmbuild' will not generate
the Provides tag if the libraries lack executable permission.

This change is part of 3f7d8ec219e1aa04b6c0417ecf5e730d40b4f149
on master (the rest of that change isn't applicable to the 1.6 branch)

Change-Id: Ie95d0d5e16745cfbf0d2a733a9421f94a89216ff
Reviewed-on: http://gerrit.openafs.org/5204
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

rpm: remove postinstall message from openafs-client

Printing out information on how to configure cacheinfo and ThisCell
is a bit noisy, and pam_afs.so is probably not what most people
want to use nowadays.

Reviewed-on: http://gerrit.openafs.org/5026
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
(cherry picked from commit 30cd8dafa73d90a943f00af05e4841699bc18534)

Change-Id: Ib6b7448b5e123a593d595b73f8102d82d1ae2535
Reviewed-on: http://gerrit.openafs.org/5203
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

rpms: Fix handling of x86 architectures

Once upon a time, our specfile would assume that if you were
building for i386 you were building userspace, and that i586 or i686
implied doing a kernel only build. This is no longer the case, and
now everything on modern Fedora is built for i686, so we should adapt
the spec file for this.

Reviewed-on: http://gerrit.openafs.org/4970
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 84a3c4714fce3c1722802963e1fcc4d7dad8cc55)

Change-Id: I16c2a4a6586e63be967b5fb019ede10a24c6cde4
Reviewed-on: http://gerrit.openafs.org/5202
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

rpms: Fix our %version handling

When we're doing version requirements in -devel package Requires:
lines, we should be using %{version}-%{release}, not just %{version}

FIXES 130137

Reviewed-on: http://gerrit.openafs.org/4969
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 15c073887b905adaef301863ed87c49f9d77b394)

Change-Id: I047f535b023a01468a3ad169d1742002abb59123
Reviewed-on: http://gerrit.openafs.org/5201
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

kdb: don't dereference a null pointer on corrupt database

When iterating through the database, kdb would dereference a
null pointer if it encountered an error retrieving the value
or if the value was not the right length, in code that was clearly
cut-and-pasted from the other branch of an "if" statement where a
specific entry was requested on the command line.  Print the name
of the entry with the problem as was apparently intended.

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5174
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit ae6a3929489035ddcd17785abab8900cebd22eb3)

Change-Id: I5c79189f82a636b923c074d3c69cb3b6fef732aa
Reviewed-on: http://gerrit.openafs.org/5188
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

FBSD: catch up with the disappearance of VOP_GETVOBJECT

The vnode operation VOP_GETVOBJECT disappeared in FreeBSD 6.0, an
embarrassingly long time ago.  Six years ago, a kluge was added
to emulate its behavior, but it did not correctly emulate the
return value of the old VOP implementation.  As a result,
osi_VM_StoreAllSegments() could never actually do anything.  Since we
don't support FreeBSD before 8.0, remove all references to VOP_GETVOBJECT
and examine vp->v_object directly instead.

This has the result that osi_VM_StoreAllSegments() will actually do
something now, which may not be desirable.  (Previously, if somehow
the vnode had no associated VM object, it would crash, and otherwise
it would do nothing at all.)

Found-by: clang static analyzer
Reviewed-on: http://gerrit.openafs.org/5183
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 7d0cd1393ff5c69cba2f14fc76aa8f7ca588ccc3)

Change-Id: I57198967ad022f50af073ac43643f3a55b0801ef
Reviewed-on: http://gerrit.openafs.org/5187
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

macos: reset next vcache pointer after reacquiring xvcache

dropping the xvcache lock means that things can change out from under
us. in case they do, reset the next vcache pointer before looping

Reviewed-on: http://gerrit.openafs.org/5184
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit e8607d8d250fb41fcf4e60a09917f971c4d6104a)

Change-Id: I2ad487e348f84cd7ced884c920230e3371fce902
Reviewed-on: http://gerrit.openafs.org/5186
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: add missing dafs man pages to wix installer

dafileserver.html
dasalvager.html
davolserver.html

Reviewed-on: http://gerrit.openafs.org/5149
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from fec4e6bc595014f34c38707c0015c1f76edd770e)

Change-Id: Ied99d695c59242903d39249c1d1bb25058d40a24
Reviewed-on: http://gerrit.openafs.org/5151
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: conditionalize mappings of error values

Visual Studio 10 adds a large number of additional POSIX C99
error values to errno.h.  Wrap each mapping with #ifndef to ensure
that we do not redefine the C runtime errno.h definition.

Reviewed-on: http://gerrit.openafs.org/5129
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from f631a11f5e9ab50e62b3aaebbff9ef200ac799fe)

Change-Id: I528547f95ebbe83390924844d21a9e850dc5967d
Reviewed-on: http://gerrit.openafs.org/5148
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: unified afs errors must use nt mapping

On Windows, error.h does not provide a complete list of POSIX
C99 error values.  OpenAFS fills in the gaps with a private
error mapping table afs/errmap_nt.h (src/util/errmap_nt.h).
If errmap_nt.h is not included prior to processing unified_afs.h,
values such as ELOOP will be mapped to EIO instead of the unique
value defined by errmap_nt.h.

Reviewed-on: http://gerrit.openafs.org/5128
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 71e64b6f21817872676e74a8c67c0f0bcfb23391)

Change-Id: If6582d7242418c15eb007471199a6dfcc6732e3b
Reviewed-on: http://gerrit.openafs.org/5147
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Do not execute tasks on deleted files

If a cm_BkgDaemon thread finds a queued request whose cm_scache_t
has the CM_SCACHEFLAG_DELETED flag set, do not execute the request
and fail it immediately with CM_ERROR_BADFD.  Any attempt to execute
the request will fail with VNOVNODE from the file server.

Reviewed-on: http://gerrit.openafs.org/5120
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from afc3ded56795d30e4e0f38d11f8302568943c49b)

Change-Id: I92610c215d7d44ceb1894bb3ab7784534a66e452
Reviewed-on: http://gerrit.openafs.org/5146
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: cm_BkgDaemon should not do cm_SyncOp's job

cm_SyncOp is designed to synchronize operations among multiple
threads.  The background daemon threads should not filter requests
based upon cm_SyncOp states.  Doing so is racy and does not produce
better performance.

Reviewed-on: http://gerrit.openafs.org/5119
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 7e3615208d9e1b03d8eba4fb802744b2b81f8868)

Change-Id: I2293643bf708c3e87093005d95c7d1203c6f680a
Reviewed-on: http://gerrit.openafs.org/5145
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Do not release locks on deleted files

If the cm_scache_t flags include CM_SCACHEFLAG_DELETED, do not
bother releasing an outstanding file lock to the file server.
The lock went away when the file was deleted.  Any attempt to
release will fail with VNOVNODE which is translated locally into
CM_ERROR_BADFD.

If a RXAFS_ReleaseLock RPC fails with VNOVNODE, treat it as
success.

Reviewed-on: http://gerrit.openafs.org/5118
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from fa11f71f040b2e6856047c53fdd902109e5c6e52)

Change-Id: I4b70abab200ea50b419e215f4b3f3b4d3d7ab484
Reviewed-on: http://gerrit.openafs.org/5144
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: permit perl command to be explicitly set

Add PERL variable to the build system.  If not specified
externally the variable will be set to 'perl'.  However,
ActiveState Perl should be used and not Cygwin Perl.  The build
environment should indicate that by specifying a PERL setting.

SET PERL=c:\perl64\bin\perl.exe

or similar.

Reviewed-on: http://gerrit.openafs.org/5115
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 4138a778fe821493c726826ffc4715f5f46222d9)

Change-Id: Ifd5ba10176ba1a7d6fda7506a0a4c250ad1945f6
Reviewed-on: http://gerrit.openafs.org/5143
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

man: add missing pod files to Windows makefile

Several of the demand attach pod files were not being processed
on Windows.

Reviewed-on: http://gerrit.openafs.org/5114
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 4a24a59d52266fcbbf92041ce3a74a455e84a25c)

Change-Id: I84663e5313d1845761ba7fe7043448ba000c9ec5
Reviewed-on: http://gerrit.openafs.org/5142
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

merge-pod changes for cygwin and MSWin32 perl

On Windows, the git repository is checked out as CR-LF.
Tell perl to open the pod file with cr-lf as the end of line.

On Windows, the input file names are of the form podX\foo.pod.in.
Cygwin perl cannot parse the directory for the file name unless
the path separator is converted from \ to /.

Reviewed-on: http://gerrit.openafs.org/5113
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 0b6247c27fcc0b8a2f307ccc545eea777a07f999)

Change-Id: Ifc4a649ee2b59115632f8d27a00b2bdb794822d1
Reviewed-on: http://gerrit.openafs.org/5141
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: add debugging to afskfw

More debugging output added when a debugger is present.

Reviewed-on: http://gerrit.openafs.org/5112
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from b61bd99bce68055dc0dd53b82d1d5318c25d9051)

Change-Id: If3aaa87f648f6fd1ae21d19d05522d48928dc4ff
Reviewed-on: http://gerrit.openafs.org/5140
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: afskfw remove TRUE conditional

Simply the code by removing an if(1) conditional.

Reviewed-on: http://gerrit.openafs.org/5111
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 7c59ec8f816120c738b1da83f7db9349d912f573)

Change-Id: I332b82b24c97810074e87521c8727b5827d7b54b
Reviewed-on: http://gerrit.openafs.org/5139
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: KFW_AFS_get_cred userrealm

The userrealm string in KFW_AFS_get_cred() should not include
the '@' symbol from the user principal.  Including the '@' produces
an invalid realm name.

Reviewed-on: http://gerrit.openafs.org/5110
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 214e9a6052b1dab2a0c4383632640936c3e518c6)

Change-Id: I24007935b35ac2a603486fc40a79f94b9b90fa8e
Reviewed-on: http://gerrit.openafs.org/5138
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: afslogon start service if not started

If the service is configured for auto start but has yet
to start, kick it off just in case.

Reviewed-on: http://gerrit.openafs.org/5109
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from a063ecdc78cf888df2459b32b0082d4767948563)

Change-Id: I8395230328385eb8d554cc85d2e659a3b7cacfb7
Reviewed-on: http://gerrit.openafs.org/5137
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: improve afskfw error message output

Add KTC and PT error messages to those that can be
translated within afskfw.lib.  This improves the error
logging for afslogon.dll, afscreds.exe, and afssrvadm.exe

Reviewed-on: http://gerrit.openafs.org/5108
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 549737f1e7d52256f053c86116b56c3211b084b6)

Change-Id: I4eb6f23e035b0774dffce8779d680b16ea607d2e
Reviewed-on: http://gerrit.openafs.org/5136
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: do not leak space allocation

smb_ReceiveNTTranCreate leaks a cm_GetSpace allocation on
error.  Don't do that.

Reviewed-on: http://gerrit.openafs.org/5062
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from c61f5208f553ed1674d1ae51d6e0f16fcdbba47b)

Change-Id: Id706189bd0aafb0b55c58f28fff80ba841a8ef48
Reviewed-on: http://gerrit.openafs.org/5135
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: smb_ReceiveNTTranCreate path not found

if the directory object cannot be found in the tree, return
CM_ERROR_PATHNOTFOUND instead of crashing.

Reviewed-on: http://gerrit.openafs.org/5061
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 9c2d3180aa760d2a406bb68dd84e73f7d1ec7019)

Change-Id: I3150f9d5ee1e1d548b29161086ebcea21b2351cb
Reviewed-on: http://gerrit.openafs.org/5134
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: cm_daemonCheckOfflineVol fix

When computing whether or not to perform an offline volume
check it is critical that the 'lastBusyVolCheck' variable
be assigned the current time instead of 'lastVolCheck'.
By setting the wrong variable a new offline volume check is
performed every 10 seconds which is undesireable.

Reviewed-on: http://gerrit.openafs.org/5054
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from 0783b8f3821f59004146b4c34a53189ada178d70)

Change-Id: I18617371d20798a923c5712188a229633ff43519
Reviewed-on: http://gerrit.openafs.org/5133
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Add shutdown event log message

Add an explicit message that the shutdown sequence is complete.
This is necessary because during a Windows OS shutdown, the service
is frequently killed prior to the memory mapped file is fully released.

Reviewed-on: http://gerrit.openafs.org/5053
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from e91f16ebb704a431f9731bb39bf9b5621b7f6ad0)

Change-Id: I80685fae2b581171c580b039ae029ca69bea467c
Reviewed-on: http://gerrit.openafs.org/5132
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Window: breakout CM error codes into separate header

Reviewed-on: http://gerrit.openafs.org/5052
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from a1af69182b12bedd8a413d38cef9a4691286da46)

Change-Id: I82adf60bef97575d47b65561fdc2b097f8ac610a
Reviewed-on: http://gerrit.openafs.org/5131
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

macos: fix vnode finalization

erroneously pushed a bad version of this. fix it now.

Reviewed-on: http://gerrit.openafs.org/5124
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 5ebef72c0299560716f3bc8d59122aedd61cc399)

Change-Id: I10aa962b158616d7d9e9207fd0307900abda0cee
Reviewed-on: http://gerrit.openafs.org/5126
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

macos: don't attempt finalize fixup on root vnode

because of how the root fid is created we can end up being dumb.
turns out we never want to bypass doing the full pass for root anyway
so just force fixup to not happen.

Reviewed-on: http://gerrit.openafs.org/5095
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit f8ff2591f73401e6d9b6def7630f428569e2f3a6)

Change-Id: I2196196d7373400f2d7f6d0654047273157df1ee
Reviewed-on: http://gerrit.openafs.org/5097
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

aklog: check ccache errors in get_user_realm

if there's an error finding the specified ccache, don't
bother continuing.

Reviewed-on: http://gerrit.openafs.org/5069
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 09d1d8575ce96ef945a4fd8a99a0347f4e258c1e)

Change-Id: I4b18cae14a9b620ea62d62eebc0b62f83f67a84a
Reviewed-on: http://gerrit.openafs.org/5072
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

macos: avoid KLRenewInitialTickets crash in Lion

the shimmed heimdal in Lion crashes on this call now.
the shim also exports diddly squat. fine, we pick over what
IS exported and use only calls available to us.

should be exactly as functional as before.

Reviewed-on: http://gerrit.openafs.org/5065
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 62cccc4547bd816dfffb2c1a4431c3c674d302f8)

Change-Id: I44ec9305c3f8e07cb68b2614b237b62e29a66516
Reviewed-on: http://gerrit.openafs.org/5071
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

macos: krb5_524 is uselessly stubbed

from MITKerberosShim-44:
dummy(krb5_524_conv_principal, 0);
dummy(krb5_524_convert_creds, 0);

this basically logs and returns success. not helpful.
so let's just not call them.

Reviewed-on: http://gerrit.openafs.org/5066
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit ee950e7ecf578bd225eec29b182f108450c5bfaf)

Change-Id: Idc2fa790d97b75971c29eec98bb96e799356dc4a
Reviewed-on: http://gerrit.openafs.org/5070
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

redhat: support epel yum configs in mockbuild

epel's mock is useful in terms of config, but we need to allow
use of it. do so here. add centos6 at the same time.

Reviewed-on: http://gerrit.openafs.org/5056
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 45f0cbf83aed2bd534e9b43822c033ef6b4a9b1b)

Change-Id: Ic46bcbcc8045773d8ce248fe2dea93b27d24ecf2
Reviewed-on: http://gerrit.openafs.org/5067
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Red Hat: update build script for newer Fedora versions

Remove old Fedora versions and add the currently-supported ones.

Reviewed-on: http://gerrit.openafs.org/4925
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 8bface302c14698a14b7b988e1a7714e84415c04)

Change-Id: I130c7289e06ede351adf8ace92fcdbac900b7819
Reviewed-on: http://gerrit.openafs.org/5068
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Add additional NEWS entries for 1.6.0pre5 through 1.6.0pre7

Taken from the release notes.  Do a bit of minor combining of related
issues since NEWS presents these all as changes in 1.6.0.

Reviewed-on: http://gerrit.openafs.org/5044
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit e06931172eec5a683490040f5564b294d263dac4)

Change-Id: I0fbd566258eec2f5d2664acf40289c4d6c544ee7
Reviewed-on: http://gerrit.openafs.org/5045
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Install afszcm.cat for i386_fbsd make dest

Because we missed it in gerrit/4815.

Reviewed-on: http://gerrit.openafs.org/5022
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit b6d05986f0c79bdf54169fc7b209b137c0e63a39)

Change-Id: Ia72180e49d1fb0a92911cb5158d8eb368d4c2a40
Reviewed-on: http://gerrit.openafs.org/5024
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

Revert "windows: include cmp_Cmd_Ref_3_en_US component"

This reverts commit 96f7d4d3e6ba75d40e7b47acb728a8dbef8c1a01.
Not until we have man3 in 1.6 do we need this.

Change-Id: If8490d6f66a220fe08995c011fee04d57d8e9adc
Reviewed-on: http://gerrit.openafs.org/5019
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

make 1.6.0pre7

update versions for 1.6.0pre7

Change-Id: I0efa0e544638de8f6f9d617d02072144f6fdd1ed
Reviewed-on: http://gerrit.openafs.org/5011
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: ChangeLog for 1.5.9907

Reviewed-on: http://gerrit.openafs.org/4954
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 30d68e6934810e3cfc67518d36cb8be26d6542b8)

Change-Id: I899f89d277d816bf429e90c09bf69f1703fd82e1
Reviewed-on: http://gerrit.openafs.org/5010
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Add make dist and make srpm targets

Add targets to generate distribution tarballs, and srpms, from a tree.
These will generate packages for whatever the current HEAD of the tree
is - if the HEAD is a release tag, then the packages will be named for
that release, if the HEAD is between releases, then git describe will
be used to create an appropriate version identifier.

The tarballs are generated from the current git repository contents,
anything not checked in will not be included.

Reviewed-on: http://gerrit.openafs.org/4984
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit f775c0fb6f9191dcf5d226bfddbf82f52cd1116c)

Change-Id: I03b4f02d979bbc851e043e5979a54d00f9c52439
Reviewed-on: http://gerrit.openafs.org/5009
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

rpms: Use git version information

Use the git version information when building RPMS, rather than hard
coding it into configure.ac.

(cherry picked from commit 6a27e228bac196abada96f34ca9cd57f32e31f5c)
Reviewed-on: http://gerrit.openafs.org/4983
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Change-Id: Ibfe38e924eb53e279aaeff77882957a32ebb279b
Reviewed-on: http://gerrit.openafs.org/5008
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

vol: Initialise list before error exit when cloning

The inode list wasn't being initialised before the first call into the
error handler. This makes it possible that we end up trying to discard
items from an uninitialised list, with all the chaos that would cause.
Fix things so that this list is correctly set up.

Reviewed-on: http://gerrit.openafs.org/5001
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit beda3b239a13903d4bcd9a5d91d5ff562962cb5a)

Change-Id: I3e3c4b8565e373f74bb95416e4c9b101d02fd78f
Reviewed-on: http://gerrit.openafs.org/5007
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

volser: remove pragma requiring ultranew gcc

this won't fly with the gcc in the field most places. move along.
(cherry picked from commit XXX)

Change-Id: I7ca987e05502c0ffc75f826b40200467bf0abfbb
Reviewed-on: http://gerrit.openafs.org/5015
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

volser: Actually return errors from ListOneVolume

The return code from GetVolInfo was being thrown away, and success
returned to the caller, regardless of the success of this function.
As GetVolInfo's exit codes aren't suitable for sending over the wire,
just return ENODEV if this function returns failure.

(cherry picked from commit 53cc2ebaea5e5488d5285f0d13ffa47069ee986f)
Reviewed-on: http://gerrit.openafs.org/5000
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Change-Id: I72e82d014ea2320a095bb4a3281449d250a52d3b
Reviewed-on: http://gerrit.openafs.org/5006
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: not safe to dereference before locking

Throughout cm_server.c, input parameters to functions that
are protected by cm_serverLock are dereferenced by assignment
during variable initialization prior to the cm_serverLock being
obtained.  As a result there is a race which can result in
either list corruption or dereferencing freed memory.

Reviewed-on: http://gerrit.openafs.org/4985
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 130155ff3c48f2da2433b359588346b4438d24a2)

Change-Id: I02e83faa889bb55b025253bbd1c51a389434eee4
Reviewed-on: http://gerrit.openafs.org/5014
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>

Revert "rx: prevent connection channel assignment race"

This reverts commit 3620b51779846175b51f82d677838c7cb4dc181d.

Change-Id: Idf8d95fe8fec0bdd94e9f9fc1c30695fb52604fa
Reviewed-on: http://gerrit.openafs.org/4986
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>

macos: note additional vfs features

currently we have inode numbers which do not change. note it.
additionally, we support large files. note it.

Reviewed-on: http://gerrit.openafs.org/4973
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 4470d849ec731c97dd2fb02dc6cbfe777949b510)

Change-Id: Iee356a8424efe3c79745bf1bf5a1553c988d9d12
Reviewed-on: http://gerrit.openafs.org/4982
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

rx: prevent connection channel assignment race

When rx was converted to use pthreads, the code that allocates
a call to a connection channel in rxi_ReceivePacket() was not
made thread safe.  The code prior to this patchset permitted a race
in the server connection case.  The rx_connection channel assignment
in rxi_ReceivePacket() and the call destruction in rxi_FreeCall()
and rxi_DestroyConnectionNoLock() did not consistently protect the
rx_connection channel array using the conn_call_lock.

This race could result in rxi_ReceivePacket() operating on a
rx_call which was disconnected from the previously assigned
rx_connection.

In addition, the code in rxi_ReceivePacket() that was intended
to protect the allocation of a call using rxi_NewCall() to the
connection channel array was racy with itself.

This patchset consistently applies the conn_call_lock to protect
the allocation / deallocation of calls to the connection channel
array and in the process simplifies the logic in rxi_ReceivePacket()
as it is no longer necessary to protect against a null call pointer
since the race can no longer be lost.

(cherry picked from commit 99b43273c0203881ea3d2d50f0abf000cdc0b03e)
Reviewed-on: http://gerrit.openafs.org/4963
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Change-Id: I90ad0dc1f67df2bcaca1da0bb459e730541876cb
Reviewed-on: http://gerrit.openafs.org/4981
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

OpenBSD: Add <sys/queue.h> header for <sys/lockf.h>

On OpenBSD, the <sys/lockf.h> header requires the TAILQ_* macros
which are defined in <sys/queue.h>. The latter is not automatically
included by <sys/lockf.h> . This patch makes sure that it is
available by putting it into the OpenBSD-specific param.h files
(so as not to impact any other OS).

Reviewed-on: http://gerrit.openafs.org/4956
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 3559e5c8edd90ffb17e8f3b722ff014654bae336)

Change-Id: I75103e0fb327fec91ae00647598450a30cc2cf23
Reviewed-on: http://gerrit.openafs.org/4980
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Red Hat: use repoquery to find kernels in build script

Remove the custom yum Python code in favor of repoquery.

Reviewed-on: http://gerrit.openafs.org/4926
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 1accef4e77c8cad518d728d15fbd194bebd49f33)

Change-Id: Ie68a006852d1564fa2da226a006a4e137d7cd5ed
Reviewed-on: http://gerrit.openafs.org/4979
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

Windows: always open dscp in smb_ReceiveNTTranCreate

There were two code paths in smb_ReceiveNTTranCreate that included
asserts in case the directory cm_scache_t object had not been
evaluated.  RT129299 contains a report that at least one of
them had been tripped in production.  There is no reason to avoid
evaluating the directory scp.  It must exist in the cache and
obtaining a reference in all cases simplifies the logic of this
overly complex function.

FIXES 129299

Reviewed-on: http://gerrit.openafs.org/4967
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 27af8b4b537236d8c678465d034721dd220641c8)

Change-Id: Id4abbc179fc1410b3d45bfc9c69814f683ebcb57
Reviewed-on: http://gerrit.openafs.org/4977
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

salvager: check namei linktable header magic

Recreate the namei linktable file if the header magic
is bad.

Reviewed-on: http://gerrit.openafs.org/4968
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit 82e772bed034ea278e7a7a1b4422d75cc14576dd)

Change-Id: I7c8b25da015e028f0a642bbfcf7d0d28bbcc16dd
Reviewed-on: http://gerrit.openafs.org/4976
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>

windows: include cmp_Cmd_Ref_3_en_US component

when building installers actually include the html 3 man pages

Reviewed-on: http://gerrit.openafs.org/4961
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 88b4e2e52799cfa5999cbc22873407ec8b77766a)

Change-Id: I40aacc81c43ec2d142886c9ce11a684c910c37b2
Reviewed-on: http://gerrit.openafs.org/4962
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: out of order lock smb v3 locking

Do not obtain the smb_rctLock after holding the cm_scache_t->rw
lock.

At most one hold is required in case of lock failure.  Obtain
it as the start of processing and drop it at the end if not required.

Reviewed-on: http://gerrit.openafs.org/4955
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit da6c97678f530ca76334613e229337a75f94e297)

Change-Id: I6b530366c74fd45ba585fd7af4615a1dc96ac055
Reviewed-on: http://gerrit.openafs.org/4960
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>

Windows: Improve logging for StoreMini and cm_BufWrite

Add trace logging to cm_StoreMini which never had it before.

Improve the logging of cm_BufWrite by adding the trucPos value
which is the new length of the file that is being reported to
the file server.

Remove cm_buf_t data references when cm_BufWrite is performing
operations using rx_Writev and iovec structures.

Reviewed-on: http://gerrit.openafs.org/4924
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
(cherry picked from commit 21854e03eb6fe70a70ea5fecde5c4ef8074e28c9)

Change-Id: I3bd5d7d834ddfa11044cf3f44af108f0554096b6
Reviewed-on: http://gerrit.openafs.org/4953
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>