If you are building the source and binary rpms from a released
tarball, instead of a real git repo, the .version file is required by
build-tools/git-version. With out this, the version defaults to
UNKNOWN, and although the source rpm will build, it won't compile.
Change-Id: I787b051d5150c65a0d71ac38b30f32e1c560c4b3
Reviewed-on: http://gerrit.openafs.org/2948
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
This reverts commit 96e433aebd.
It was not properly tested, fails to compile on some systems,
and destabilizes the client.
Change-Id: I80b08013c9f2d060ab7ed5a18d57311701aa8b88
Reviewed-on: http://gerrit.openafs.org/2951
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Oops -- when removing the -cell arguments to the bos commands, the
"admin" username argument to "bos adduser" was mistakenly removed as
well.
Change-Id: If5136869c7d2bff3340018a3110fd5408750e5a6
Reviewed-on: http://gerrit.openafs.org/2940
Reviewed-by: Russ Allbery <rra@stanford.edu>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Keeping dvp locked when vp is its parent can lead to deadlock.
Always unlock dvp, not just for 6 and earlier.
Change-Id: I26a60188d39ccd24b4db7479c57a525bb37618e9
Reviewed-on: http://gerrit.openafs.org/2942
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
The LRU queue is where volume objects are recycled from.
Therefore if they are removed then they must be put back.
Remove extraneous operations from cm_AdjustVolumeLRU().
LICENSE MIT
Change-Id: I1f6aadd0ffb7aeafa013087e4825b3c8d8252a2d
Reviewed-on: http://gerrit.openafs.org/2913
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
The names of the RHEL rpms have been updated to reflect what is
actually published on openafs.org, and the CLI syntax of the commands
run using -noauth have had the unnecessary -cell options removed.
Change-Id: I5c03d371c822d1d064660a3b00fb3d1e64bff141
Reviewed-on: http://gerrit.openafs.org/2914
Reviewed-by: Russ Allbery <rra@stanford.edu>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
When calling "vos partinfo" or "vos listpart" towards a server
running AIX with no partitions attached, it would return a
ENOMEM, because unlike on linux, malloc(0) returns NULL on AIX.
Thus, just don't do any malloc, when we have no partitions anyway.
Change-Id: Id1900e2ab11850ada8b2e91667288576d408014b
Reviewed-on: http://gerrit.openafs.org/2912
Reviewed-by: Andrew Deason <adeason@sinenomine.net>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
This patch reduces our dependence on call->lock, by allowing more
of the reader thread to run lock free. Doing so requires that
call->mode only be set by the reader thread. As a result, call->mode
can only be set to RX_CALL_ERROR by rxi_CallError(). The mode is
set to RX_CALL_ERROR by the reader thread immediately after regaining
the call->lock when it has been dropped.
Change-Id: Ie9541d8beac2d428526f8b2b4cc0004219e820be
Reviewed-on: http://gerrit.openafs.org/2880
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
The recent hcrypto pullup added a depedence on the errx() function,
which isn't present on Irix. Solve this by pulling in a load more
of libroken, in order to provide this function.
In the long term, libroken should get split out into its own
directory, and the ability to use a previously installed
libroken should be added. For now, this will hopefully get Irix on
its feet again.
Change-Id: I0642f80079e113403cbef6eca03ca652616ddc61
Reviewed-on: http://gerrit.openafs.org/2904
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Supress a few of our errors from the Irix compiler and linker, so its
output is a little less verbose.
This change suppresses the function declared but not used and
multiple declaration errors that we get due to our static_inline fudge
and the paramater declared but not used errors.
Other error suppression is possible - you just need the number
immediately after the 'cc-' in the build logs to say which number to
add to the -woff line.
Change-Id: I55485ff422feeecbb922e8fd63321605d6a8575c
Reviewed-on: http://gerrit.openafs.org/2908
Tested-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
This commit updates the code imported from the external
heimdal git repository to their revision
48ad3e1e6597f03cce5439831ef94b5ec11894e9
which is described as switch-from-svn-to-git-1593-g48ad3e1
Change-Id: I0fda4dcf7c7e06de70636969c2f3242b9de714be
Reviewed-on: http://gerrit.openafs.org/2907
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
All builds define AFS_LARGEFILES so kill the symbol and
discard the !AFS_LARGEFILES source code.
Change-Id: I36a2131e30b24d3d1a8f37f5629795bdd92c6b27
Reviewed-on: http://gerrit.openafs.org/2910
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
The AFS_64BIT_ENV was being pretty much universally defined. So,
remove the definition, and the conditional code that it controlled.
From this point on, all platforms are assumed to be capable of
handling 64bit values.
Change-Id: I3e4bde502af9f33f6998637c288e8fd0898ffa81
Reviewed-on: http://gerrit.openafs.org/2870
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
This reverts commit d552426b14.
util already has a base64.c, the Windows bits of this are half-baked,
and we need to find a config.h from somewhere so that this builds
cleanly.
Just back it out, until I can do it properly.
Change-Id: Iec3ef4579c8f83c7c00e03ad777489e287bbd2f3
Reviewed-on: http://gerrit.openafs.org/2909
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
This commit updates the code imported from the external
heimdal git repository to their revision
48ad3e1e6597f03cce5439831ef94b5ec11894e9
which is described as switch-from-svn-to-git-1593-g48ad3e1
This import was manually modified to preserve our current local
fix for sha512.c
Change-Id: Ie020beb7324469d33b85ac4c559cc473f2728d19
Reviewed-on: http://gerrit.openafs.org/2903
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Import still more files from Heimdal, so that we can try to make
Irix happy once more. Mutter, mutter, mutter.
Change-Id: Ic8f20885fd324ab3698109f75ba41cdf6d0975dd
Reviewed-on: http://gerrit.openafs.org/2902
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
The earlier inclusion of sha512 from Heimdal broke the build on
32bit platforms, because this file doesn't flag 64bit constants as
being such.
The correct place for this fix is in Heimdal, and it will be replaced
with a fix from Heimdal as soon as one is available. For now though,
this gets our build going again.
Change-Id: I0de822a933184078c491a16aafef45519fa2bbd2
Reviewed-on: http://gerrit.openafs.org/2901
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
If we see a /vicep*-like directory when we VAttachPartitions, and we
ignore it because it lacks an AlwaysAttach, log that we ignored it.
This may make things less confusing to admins that just try to create
a /vicep* directory and don't know why it doesn't work.
Feature suggested by Jason Edgecombe.
FIXES 128221
Change-Id: I0802f914658c7df413b22b3726a5085f1b646266
Reviewed-on: http://gerrit.openafs.org/2893
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Add the base64 code from libroken into libutil, using the src/external
framework.
Create a new, non-installed directory, to hold util headers, rather than
continuing to stuff everything into afs/
Change-Id: I6b743a3702c3bb07ac798392475e7e067765f238
Reviewed-on: http://gerrit.openafs.org/2900
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Fix the hcrypto build to deal with the changes introduced by the
latest import from Heimdal
Change-Id: I2effebffb6df2fdb351d070f8ea06ecf01aab46e
Reviewed-on: http://gerrit.openafs.org/2899
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
This commit updates the code imported from the external
heimdal git repository to their revision
48ad3e1e6597f03cce5439831ef94b5ec11894e9
which is described as switch-from-svn-to-git-1593-g48ad3e1
Change-Id: I90eb6752000d6794e830a704adcf367928b05a98
Reviewed-on: http://gerrit.openafs.org/2898
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Add the base64 implementation from libroken, and the n-fold.c file
from libkrb5 to the list of source files that we import from Heimdal.
It also pulls in the new sha512.c and validate.c files which are now
necessary to complete a userspace hcrypto build.
This patch will fail to build until the build fixes from the subsequent
change are applied.
Change-Id: I60a691d0ccebd6d6bfb823856f79b59670eb86ec
Reviewed-on: http://gerrit.openafs.org/2897
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
When we test for whether the C compiler can take the
-fno-strength-reduce flag, we add the flag to CFLAGS to do so.
However, we were not restoring the old value of this flag when we
completed the test, and so we were always setting -fno-strength-reduce
in the userspace compile.
Previously, this was harmless, as we always overwrote CFLAGS, but if
we're moving to a world where we honour the user's setting of CFLAGS,
we need to not leak changes in this way.
Change-Id: I029c9fe91132ced34867445afe673761a597ccbe
Reviewed-on: http://gerrit.openafs.org/2894
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
If the user hasn't specified CFLAGS on the command line to
./configure, then autoconf will set them to -g -O2 if the compiler
supports those options.
For compatibility with what OpenAFS has always done, and to let us
manually set optimisation and debugging flags later, disable this
behaviour.
Change-Id: Ic78d5f824433d94d76f16c107af3488132d57155
Reviewed-on: http://gerrit.openafs.org/2895
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
When releasing a volume, currently we perform an incremental dump on
RO_DONTUSE sites if the volume exists on the remote site. Since
RO_DONTUSE implies that we do not expect a site to exist there, delete
the extant volume if we find one and force a full dump.
If we perform an incremental dump, we run the risk of incrementally
dumping to a temporary volume that the administrator is not aware of,
and doesn't have any actual data but has a last update time late
enough that it may be missing some data after the incremental dump. So
to avoid that, force a full dump every time.
Change-Id: If751a66d6e1499b5d6f67457ae28277a6200c53a
Reviewed-on: http://gerrit.openafs.org/2731
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
sin_family isnt sent over the network and therefore doesnt need
htons(). sin_family is essential the same as domain, and no one
does socket(htons(AF_INET), ...)
Change-Id: Ie0e2396f78e0934d49aeeacfc9c5ffe98df211b6
Reviewed-on: http://gerrit.openafs.org/2891
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
The libuafs part of commit 4346144b was not needed, rx_stats was
already in the Makefile. Harmless, but generates warnings.
Revert the libuafs portion of that commit.
Change-Id: I14d64445c9690bc6d69881c6bd6c00c4670895f3
Reviewed-on: http://gerrit.openafs.org/2892
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
for platforms with no native atomics we use a mutex. well,
fine, but without defining kernel mutexes, that doesn't work so
well.
Change-Id: Ia85e24aad8684a3855511de9548acee4ab86e83e
Reviewed-on: http://gerrit.openafs.org/2890
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
There are many cases in the SMB server where an error from cm_NameI()
was either ignored or not acted upon until several other operations
are performed that could result in the same error being repeated.
This is a mistake which did not have negative side effects until
additional checks for callback status were added recently.
At present, if a CM_ERROR_ACCESS error is returned and ignored,
subsequent attempts to operate on the same cm_scache_t will result
in additional queries to the file server that will also end in an
abort response. This can trigger the file server to delay responses
to the client.
This patchset ensures that all cm_NameI() errors are acted upon
promptly.
LICENSE MIT
Change-Id: Ie334b624cc2b28f2c2a37787b5edef9d37cdb041
Reviewed-on: http://gerrit.openafs.org/2887
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
The parent path computation was leaving trailing slashes on the
path names which prevented the creation of mount points and symlinks
when UNC paths were used that contained mount points.
LICENSE MIT
Change-Id: I50cba9cb8a2b0ad45f84995d05f368052be683cc
Reviewed-on: http://gerrit.openafs.org/2886
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Rod Widdowson <rdw@steadingsoftware.com>
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Linux doesn't have <string.h>, it's in the linux subdirectory. But
it is already included by sysincludes.h, so it should not be needed
here.
Change-Id: I017c374e7b6d47e900ec73b40a36183a3f23f35c
Reviewed-on: http://gerrit.openafs.org/2888
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
The rx_atomic_add and rx_atomic_sub functions have arguments
reversed compared to the linux atomic_add and atomic_sub.
Adjust the macros to flip the arguments when calling them
Change-Id: I6d58c8f0ac8ccf09e62c9d3921b9f3ae50ca94c9
Reviewed-on: http://gerrit.openafs.org/2889
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
At 2ms it is possible for the packet we are sending to be resent
just about immediately as the retryTime computation occurs before
the send takes place and not afterwards. If the network send blocks,
the retryTime may have already expired.
We do not want rx_minPeerTimeout to be too large though because the
value will end up being used as the backoff time period when the
actual RTT for the connection is less than the rx_minPeerTimeout
value. Extensive testing shows 20ms to be an adequate compromise
that avoids the vast majority of unnecessary resends without
unnecessarily slowing down the connection if a packet is in fact
lost.
Change-Id: I87248ec1cff6a4d0862da0239d4c203024a783b6
Reviewed-on: http://gerrit.openafs.org/2881
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
When all of the packets for a call have been received,
immediately send an ack all packet to the peer. This permits
the peer to free the contents of the transmit queue and cancel
all pending resend events.
Change-Id: Ic06ccaca6f0c5e2f770c5c45c84f7fc155207bcf
Reviewed-on: http://gerrit.openafs.org/2871
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Add a new global mutex rx_refcnt_mutex to protect the conn->refCount
and call->refCount in place of relying upon the conn->conn_data_lock
and the call->lock.
This will relieve some lock contention with rx_ReceivePacket().
Change-Id: Iebc04a87149a9fe69a7e312e8968b08c2e94148d
Reviewed-on: http://gerrit.openafs.org/2837
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
actually enable atomics for the darwin kernel, whose API is
just slightly different
Change-Id: I566149428d253806c4546fa0c5a3f96e8098c4eb
Reviewed-on: http://gerrit.openafs.org/2884
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
rx_Readv() is a bit "faster" than rx_Read() and typically used by most
afs transaction.
server% rxperf server
client% rxperf client -c send -s server
SEND: threads 1, times 100, bytes 1048576: 2073 msec
server% rxperf server -v
client% rxperf client -c send -s server
SEND: threads 1, times 100, bytes 1048576: 983 msec
Change-Id: I27b06dbbb61f2bc34fa91bcda3c6e046c787ff62
Reviewed-on: http://gerrit.openafs.org/2877
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
RXAFS_GetVolumeStatus can return VNOVOL, VMOVED, etc. In order to
process them and update volume state a fid must be passed to cm_Analyze().
Use the volume root fid.
LICENSE MIT
Change-Id: Ia6115a17aae06144277271048e8287e5ad52ff2a
Reviewed-on: http://gerrit.openafs.org/2883
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Add some routines to maintain a pool of threads, for working through a
Vwork_queue.
This adds the afs_tp* series of functions. Originally written by
Tom Keiser.
Change-Id: I8735aa14ca6622ae0eca7a7589e69b0f3b3daf08
Reviewed-on: http://gerrit.openafs.org/1863
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Add some routines for specifying chunks of work to be done. The idea
is to be able to pass these to different threads, and specify
dependencies between them, wait on them completing, etc.
This adds the afs_wq* family of functions. Originally written by Tom
Keiser.
Change-Id: If556cf4da12de8c4be1e53376d85d791584ae177
Reviewed-on: http://gerrit.openafs.org/1862
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
if we set lastReceiveTime to 0 to hint that no net, honor that here as
"just time out"
Change-Id: Ifb06fad782669649027841e3930ff94600b6c900
Reviewed-on: http://gerrit.openafs.org/2875
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Only report detailed per-chain volume statistics on shutdown/SIGXCPU
if LogLevel is 125 (or 25 for smaller per-chain stats). If a
fileserver is configured with a large -vhashsize, printing out stats
for each chain can take awhile and use up a nontrivial amount of disk
space for logging, so only print out these stats if we're asked for
them.
Change-Id: Iceb38e29ab40958e05f3cf146687f679bd0f061c
Reviewed-on: http://gerrit.openafs.org/2759
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
sendtimes and recvtimes are actually the data byte counts to be
exchanged during an rpc test not an iteration count as implied
by their names
Change-Id: I3d567aacf1ab1d25fda20e48a6bbb68a63330230
Reviewed-on: http://gerrit.openafs.org/2876
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
the test for this build feature is reversed. by default, the value for
with_linux_kernel_packaging will not be defined which makes the existing
test pick MPS='SP' instead of LINUX_WHICH_MODULES. based on the configure
help messages, this would appear to be an opt-in not an opt-out.
...
Optional Packages:
...
--with-linux-kernel-packaging
use standard naming conventions to aid Linux kernel
build packaging (disables MPS, sets the kernel
module name to openafs.ko, and installs kernel
modules into the standard Linux location)
...
Change-Id: Ie16fba165e6c85b7ecbce887badb9ffc06a3a7a8
Reviewed-on: http://gerrit.openafs.org/2869
Reviewed-by: Simon Wilkinson <sxw@inf.ed.ac.uk>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: BuildBot <buildbot@rampaginggeek.com>
You can only specify a -vhashsize between 6 and 28 (inclusive).
Document that in the dafileserver man page.
Change-Id: I44d1c71f4ff303174e8aebf74b0b9075c07bc8b4
Reviewed-on: http://gerrit.openafs.org/2650
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
ucontext requires glibc version info to engage.
glibc info comes from features.h, which if we include afs_sysnames.h,
we get... so, let's swap these.
Change-Id: I26e5554f2c5c5f9f8a74fb591efdfdc1f6ae7257
Reviewed-on: http://gerrit.openafs.org/2868
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
