openafs.git
22 years agosolaris-mtu-cleanup-20011005
Nickolai Zeldovich [Fri, 5 Oct 2001 21:44:35 +0000]
solaris-mtu-cleanup-20011005

This patch provides better MTU selection on Solaris, by
  actually going through the list of interfaces and picking
  the correct MTU, rather than assuming 1500.

  It also fixes a small bug in the server preference code,
  which wasn't checking for IPv6 interfaces, and if there
  were any IPv6 interface, it would believe all servers
  were on the local subnet.

22 years agocvsignore-updates-kolya-20011005
Nickolai Zeldovich [Fri, 5 Oct 2001 21:41:52 +0000]
cvsignore-updates-kolya-20011005

ignore more things since we missed them before

22 years agofix-prototype-warnings-and-eliminate-unused-variables-20011005
Nickolai Zeldovich [Fri, 5 Oct 2001 21:41:09 +0000]
fix-prototype-warnings-and-eliminate-unused-variables-20011005

minor cleanup in mkvers and kauth/admin_tools

22 years agolinux-compute-mtu-correctly-20011005
Nickolai Zeldovich [Fri, 5 Oct 2001 21:38:57 +0000]
linux-compute-mtu-correctly-20011005

This patch fixes two bugs:

  (1) pass more than 4 parameters to AFSCALL syscalls on Linux,
      allowing the ADVISEADDR op to pass the MTU array pointer.

  (2) properly convert MTU's to network byte order, since that's
      what the rest of the code expects them to be.

22 years agoafsd-afsdb-shutdown-cleanly-20011005
Nickolai Zeldovich [Fri, 5 Oct 2001 21:36:59 +0000]
afsd-afsdb-shutdown-cleanly-20011005

make userspace afsdb handler shut down cleanly

22 years agolinux-generate-struct-vnode-dyanmically-from-kernel-headers-instead-of-by-perverting...
Derek Atkins [Mon, 1 Oct 2001 23:58:59 +0000]
linux-generate-struct-vnode-dyanmically-from-kernel-headers-instead-of-by-perverting-configure-to-test-for-every-one-of-alan-coxs-additions-some-of-which-will-doubtless-never-make-it-into-mainline-kernels-20011001

hopefully this will be the end
(at least for a while) of problems where every linux version requires a new
openafs release. the struct vnode is now dynamically generated by parsing the
linux/fs.h file and pulling out the struct inode, then rewriting osi_vfs.h

22 years agoallow-configure-time-override-of-DEST-20011001
Derrick Brashear [Mon, 1 Oct 2001 20:20:04 +0000]
allow-configure-time-override-of-DEST-20011001

so DEST=foo ./configure will work

22 years agobe-more-careful-about-references-when-adding-and-deleting-callbacks-20011001
Derrick Brashear [Mon, 1 Oct 2001 18:51:41 +0000]
be-more-careful-about-references-when-adding-and-deleting-callbacks-20011001

perhaps too careful. to fix potential crash in TDel and TAdd

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

remove extra ) which snuck in

22 years agorequire-administer-permission-for-mount-point-creation-harder-20010926
Nickolai Zeldovich [Mon, 1 Oct 2001 18:35:50 +0000]
require-administer-permission-for-mount-point-creation-harder-20010926

make sure a mount point isn't being created if the user isn't allowed to
do that

22 years agolinux24-another-new-inode-field-sigh-20011001
Mark W. Eichin [Mon, 1 Oct 2001 18:31:49 +0000]
linux24-another-new-inode-field-sigh-20011001

with configure glue by shadow@dementia.org

one of these days we really need to implement a better solution

22 years agologin-makefile-needs-new-build-system-update-20011001
Nathan Neulinger [Mon, 1 Oct 2001 18:20:49 +0000]
login-makefile-needs-new-build-system-update-20011001

fixed missed case of afs/ in libdir

22 years agopam-build-on-hpux11-20011001
Nathan Neulinger [Mon, 1 Oct 2001 18:19:22 +0000]
pam-build-on-hpux11-20011001

so pam now builds on hpux 11

22 years agolinux-kdump-wrapper-20011001
Nathan Neulinger [Mon, 1 Oct 2001 18:17:20 +0000]
linux-kdump-wrapper-20011001

install a kdump wrapper for linux. (with mod to use $0 instead of kdump)

22 years agofix-destdir-support-for-fsconv-and-xfs-binaries-20011001
Jeffrey Hutzelman [Mon, 1 Oct 2001 18:11:03 +0000]
fix-destdir-support-for-fsconv-and-xfs-binaries-20011001

properly support DESTDIR case in vol makefile for certain files which were
broken before

22 years agorx-ignore-vab-harder-20010928
Derrick Brashear [Fri, 28 Sep 2001 14:09:21 +0000]
rx-ignore-vab-harder-20010928

if the other side suggests we use the vab security class, ignore them
harder than we did before

22 years agoafs-lookup-put-back-vcache-entry-on-failed-evalmountpoint-20010927
Derrick Brashear [Thu, 27 Sep 2001 18:37:49 +0000]
afs-lookup-put-back-vcache-entry-on-failed-evalmountpoint-20010927

based on observations of effects related to
afs-lookup-check-evalmountpoint-return-20010221 and affirmed based on
mail from Todd_DeSantis@transarc.com

this should reduce but perhaps not eliminate the VLRU cycle: increase
-stat parameter problem on Linux as it's not yet clear that this is the
only problem and circumstantially it is probably not.

22 years agoafs-client-inline-bulkstatus-rpc-support-20010926
Derrick Brashear [Thu, 27 Sep 2001 03:16:36 +0000]
afs-client-inline-bulkstatus-rpc-support-20010926

this cuts rpc traffic in half for directories on which you have
no permissions and is capable of doing more if a useful way can
be conceived of caching errors.

currently bulkstat works thus:
say a directory has 100 files. bulkstat will on the first hit stat
the first 30 files, integrating the information if successful,
and when you hit file 31, repeat. meaning if you can read that 100 files
is 4 bulkstatus rpcs (3*30+10). if you can't, you do bulkstatus, get an abort,
then do fetchstatus on the file and find out why you lost. you go on and
try bulkstatus again this time for files 2-31 instead of 1-30. same failure,
meaning you do 2N rpcs for the N files. inlinebulk doesn't abort but instead
fills in stat info for any of the up to 30 files it can, and errors for
any it can't.

currently the client will pull out any stat info filled in,
and then return the error from the first file, meaning if you can't
read the directory we do N inlinebulk rpcs instead of N bulkstatus +
N fetchstatus rpcs. if the errors for the other 29 could be meaningfully
cached (and you have to be careful not to pollute information
such that it's not useful for someone in a different auth context) then
we could get down to the same 4 rpcs we have for success, but that's an
exercise for later give how the cache manager works.

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

create audit event for inlinebulk

====================

rename spare2

22 years agofix-locks-for-openbsd-20010926
Jim Rees [Wed, 26 Sep 2001 22:49:33 +0000]
fix-locks-for-openbsd-20010926

don't confuse direct with dirent
use curproc instead of cur_proc()

22 years agobos-avoid-false-errors-when-salvaging-20010926
Derrick Brashear [Wed, 26 Sep 2001 16:28:22 +0000]
bos-avoid-false-errors-when-salvaging-20010926

there were some cases where we falsely reported the BZNOENT from salvage-tmp
being cleanup up as "bos: failed to start 'salvager' (no such entity)"..
despite having just reported:
bos: salvage completed

this should no longer happen.

22 years agoallow-namei-fileservers-to-use-directories-instead-of-partitions-for-vicep-20010924
Nickolai Zeldovich [Wed, 26 Sep 2001 01:02:46 +0000]
allow-namei-fileservers-to-use-directories-instead-of-partitions-for-vicep-20010924

 -- /vicepX/AFSIDat/README is created the first time the fileserver
    is started, rather than the second time.

 -- For the namei fileserver, /vicepa/Lock/vicepa is created as
    the lockfile.  Previously, on Solaris, the raw device was
    was being used for LockPartition(), and that breaks when
    there isn't a corresponding device.

 -- The VolPartitions and XVolPartitions volserver calls now return
    the list of attached (rather than mounted) partitions.  This is
    consistent with the NT fileserver and removes duplicate code.

The actual magic for allowing the fileserver to attach non-mounted
partitions is to create /vicepX/AlwaysAttach.

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

carefully avoid non vicep directories and/or partitions as it may get upset at
automounters or running cachemanagers!

====================

more carefully avoid non-vicepX partitions

22 years agocreate-inline-bulkstatus-rpc-and-server-stub-20010925
Derrick Brashear [Tue, 25 Sep 2001 16:44:39 +0000]
create-inline-bulkstatus-rpc-and-server-stub-20010925

this way it's possible to do a bulkstatus where some items succeed and
fill in that stat info, instead of just getting an error.

22 years agofileserver-remove-delay-on-stat-errors-20010925
Derrick Brashear [Mon, 24 Sep 2001 23:19:46 +0000]
fileserver-remove-delay-on-stat-errors-20010925

previously and since 3.5 rx enhancements fileserver would delay 3 seconds
on sending an rx abort when the abort was the 11th or more occurance of
the same error on the same call. the problem was fetchstatus/bulkstatus
on a directory on which you have "l" but not "r" would take forever after the
first 5 files (bulkstatus and then fetchstatus on each file, 2*5=10)

so in FetchStatus and BulkStatus stubs we suppress repeated errors from
Check_PermissionRights *only*, which has the effect of suppressing the
problem without removing the theoretical benefit of avoiding letting any
one client throw requests at as as fast as we can abort them unless those
requests happen to be fetch/bulkstat requests which fail a permission check,
but in order to avoid that we'd also need to cache what they last tried
to stat, and then things would get ugly.

22 years agocomerr-include-proper-old-compat-glue-20010924
Derrick Brashear [Mon, 24 Sep 2001 17:11:07 +0000]
comerr-include-proper-old-compat-glue-20010924

so old initialize function works correctly

22 years agoexport-vlops-in-vldbint-xg-file-20010924
Derrick Brashear [Mon, 24 Sep 2001 13:17:16 +0000]
export-vlops-in-vldbint-xg-file-20010924

based on report from lha@stacken.kth.se

include vlops in the "published" interface

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

based on report from lha@stacken.kth.se

remove from vlserver.p.h

22 years agoNEWS-updates-20010924
Derrick Brashear [Mon, 24 Sep 2001 11:51:05 +0000]
NEWS-updates-20010924

update NEWS for stuff in 1.2.1 and stuff to go into 1.2.2

22 years agoviced-deal-gracefully-with-afsconf-failure-20010924
Nickolai Zeldovich [Mon, 24 Sep 2001 11:48:14 +0000]
viced-deal-gracefully-with-afsconf-failure-20010924

-- Fileserver will complain if afsconf_Open() fails, rather than
    mysteriously segfault'ing later on.

22 years agorx-cleanup-20010924
Nickolai Zeldovich [Mon, 24 Sep 2001 11:44:32 +0000]
rx-cleanup-20010924

The rxi_ReceiveAckPacket() change makes sure we have two extra fields
(the max and recommended packet sizes) hanging off the packet, rather
than just one.

The PKTFLAG_FREE stuff in rx_packet.c just initialized flags in case
PKTFLAG_ACKED somehow didn't get cleared there.

The wirevec changes are off-by-1 errors: there's p->niovecs iovecs
in a packet, from 0 (header) to p->niovecs-1, so p->niovecs doesn't
refer to a meaningful iovec.

22 years agodux-vfsck-check-for-mounted-filesystem-20010919
Mattias Amnefelt [Sat, 22 Sep 2001 00:48:56 +0000]
dux-vfsck-check-for-mounted-filesystem-20010919

modified by shadow@dementia.org for dux 4.0a (ifndef UID_NO_CHANGE case)

22 years agodarwin-optimize-and-fix-dirpath-20010920
Chaskiel M Grundman [Thu, 20 Sep 2001 21:10:06 +0000]
darwin-optimize-and-fix-dirpath-20010920

change optimization, add alternate dirpath for dealing with darwin and macos
in same binaries

22 years agoundo-linux-call-afs-dopartialwrite-20010920
Derrick Brashear [Thu, 20 Sep 2001 06:08:14 +0000]
undo-linux-call-afs-dopartialwrite-20010920

scrap this approach

22 years agoubik-fix-typo-20010919
Derrick Brashear [Thu, 20 Sep 2001 04:58:23 +0000]
ubik-fix-typo-20010919

somehow this crept in

22 years agowinnt-dont-do-unneeded-version-check-20010918
James Peterson [Thu, 20 Sep 2001 00:18:54 +0000]
winnt-dont-do-unneeded-version-check-20010918

make afs config wizard not complain about version < 3.5

22 years agonews-include-windows-updates-20010918
Shyh-Wei Luan [Thu, 20 Sep 2001 00:16:19 +0000]
news-include-windows-updates-20010918

list windows client user-visible changes

22 years agoopenbsd-nits-20010919
Jim Rees [Wed, 19 Sep 2001 18:59:08 +0000]
openbsd-nits-20010919

With these small changes, "make" on OpenBSD actually completes!

22 years agodocument automake dependency
Jim Rees [Tue, 18 Sep 2001 21:44:02 +0000]
document automake dependency

22 years agocomerr-behave-like-the-rest-of-the-world-20010918
Derrick Brashear [Tue, 18 Sep 2001 05:26:47 +0000]
comerr-behave-like-the-rest-of-the-world-20010918

no other comerr changes the case of the error table name you specify in the
.et file. now neither does ours.

22 years agomisc-build-cleanup-20010917
Walter Wong [Mon, 17 Sep 2001 22:37:21 +0000]
misc-build-cleanup-20010917

1) Removal of unused variables
2) some NT ifdef cleanup
3) use of size_t rather than int to store the result of strlen()
4) some bugs that really only occured when errors occured.
5) some functions were implicitly declared as 'int' but doing a
   "return;" Some I declared as 'void' and prototyped them; some I
   just returned a value.

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

put symbol in an ifndef for winnt

====================

change erroneous void to int

====================

technically the cmdprocs return int

====================

update function to be void

22 years agont-makefile-clean-targets-20010917
Walter Wong [Mon, 17 Sep 2001 20:35:09 +0000]
nt-makefile-clean-targets-20010917

more work on providing a clean target for nt

22 years agoreadme-cvs-update-with-instructions-20010917
Walter Wong [Mon, 17 Sep 2001 20:26:24 +0000]
readme-cvs-update-with-instructions-20010917

providedetails about openafs cvs

22 years agoppc-linux-needs-install-path-20010917
Sam Hartman [Mon, 17 Sep 2001 19:33:50 +0000]
ppc-linux-needs-install-path-20010917

set path to install in pcc linux22 makefile fragment

22 years agolinux-call-afs-dopartialwrite-from-afs-memwrite-and-afs-ufswrite-not-afs-linux-write...
Derrick Brashear [Mon, 17 Sep 2001 19:12:20 +0000]
linux-call-afs-dopartialwrite-from-afs-memwrite-and-afs-ufswrite-not-afs-linux-write-20010828

don't erroneously call fakeopen

22 years agoInitial OpenBSD support. Most of user space builds. No kernel module yet.
Jim Rees [Fri, 14 Sep 2001 00:19:17 +0000]
Initial OpenBSD support. Most of user space builds. No kernel module yet.

22 years agodarwin-update-plist-building-20010912
Chaskiel M Grundman [Wed, 12 Sep 2001 06:00:46 +0000]
darwin-update-plist-building-20010912

update how plist files are constructed so subsconfigure for libafs can work

22 years agolinux-makefileproto-redhat-enterprise-support-20010912
Derrick Brashear [Wed, 12 Sep 2001 16:55:58 +0000]
linux-makefileproto-redhat-enterprise-support-20010912

based on work by thomas.mueller@hrz.tu-chemnitz.de

22 years agodarwin-plist-renaming-20010911
Derrick Brashear [Tue, 11 Sep 2001 21:30:21 +0000]
darwin-plist-renaming-20010911

allow configure processing of files

22 years agoafs-callback-kill-bcopy-again-20010911
Derrick Brashear [Tue, 11 Sep 2001 20:28:56 +0000]
afs-callback-kill-bcopy-again-20010911

this slipped in with the memory leak fix

22 years agoupdate-top-level-docs-for-new-install-targets-20010911 BP-openafs-stable-1_2_x
Derrick Brashear [Tue, 11 Sep 2001 07:56:23 +0000]
update-top-level-docs-for-new-install-targets-20010911

make top level files indicate current build instructions

22 years agoaix-makefileproto-cleanup-dest-building-20010911
Derrick Brashear [Tue, 11 Sep 2001 07:25:54 +0000]
aix-makefileproto-cleanup-dest-building-20010911

readd missing /es

22 years agomakefileproto-solaris-install-modules-in-modlad-dir-20010911
Derrick Brashear [Tue, 11 Sep 2001 05:46:37 +0000]
makefileproto-solaris-install-modules-in-modlad-dir-20010911

get rid of another bogosity

22 years agoconfigure-libafs-fix-paths-20010910
Derrick Brashear [Tue, 11 Sep 2001 05:00:29 +0000]
configure-libafs-fix-paths-20010910

update paths for building submodule

22 years agoirix-makefileproto-fix-dest-target-20010910
Derrick Brashear [Tue, 11 Sep 2001 04:30:49 +0000]
irix-makefileproto-fix-dest-target-20010910

rely on DEST not DESTDIR
also install instead of cp to implicitly create needed directory

22 years agolinux-configure-check-for-more-inode-struct-members-20010910
Derrick Brashear [Mon, 10 Sep 2001 21:52:30 +0000]
linux-configure-check-for-more-inode-struct-members-20010910

for redhat enterprise kernel

22 years agoupdate-libafs-configure-20010910
Derrick Brashear [Mon, 10 Sep 2001 21:50:43 +0000]
update-libafs-configure-20010910

so the kernel module subconfigure will work usefully

22 years agomakefile-cleanup-remove-extraneous-semicolons-and-other-unused-bits-20010910
Nathan Neulinger [Mon, 10 Sep 2001 21:21:01 +0000]
makefile-cleanup-remove-extraneous-semicolons-and-other-unused-bits-20010910

rationalize away some more crap

22 years agoadd-cvsignores-to-project-20010910
Nathan Neulinger [Mon, 10 Sep 2001 21:14:01 +0000]
add-cvsignores-to-project-20010910

start at cvsignores

22 years agowinnt-readme-update-20010910
Walter Wong [Mon, 10 Sep 2001 21:08:14 +0000]
winnt-readme-update-20010910

update winnt build instructions

22 years agodarwin-build-updates-20010910
Chaskiel M Grundman [Mon, 10 Sep 2001 21:07:32 +0000]
darwin-build-updates-20010910

separate plist on per-version basis

1.4 is not yet supported

22 years agowinnt-create-missing-clean-target-20010910
Walter Wong [Mon, 10 Sep 2001 20:54:20 +0000]
winnt-create-missing-clean-target-20010910

get winnt make clean closer to working

22 years agomake-install-support-for-transarc-and-prefix-style-universes-20010907
Jeffrey Hutzelman [Mon, 10 Sep 2001 20:15:57 +0000]
make-install-support-for-transarc-and-prefix-style-universes-20010907

according to jeff:
- Renames the top-level 'install' target to 'build'.  This should be
    transparent, since no one should be using that.

  - Improves on Sam's dirpath patches, by splitting out server binaries
    into separate bin, sbin, and libexec directories in GCS mode (these
    are all /usr/afs/bin in Transarc mode).

  - Updates the top-level 'all' target so that it builds the software
    but does not generate a dest tree.  Top-level 'lib' and 'include'
    directories are generated to hold the intermediate libraries and headers
    used during the build.

  - Adds a new top-level 'install' target, which installs things in the
    appropriate directories under ${DESTDIR}, based on configure's install
    directories plus the extras added by Sam's patch.

  - Adds a new top-level 'dest' target, which creates an old-style dest
    directory under ${DEST}.  The ${DEST} variable defaults to the
    traditional value of ${SRCDIR_PARENT}/${SYS_NAME}/dest.  Note that
    this variable used to be called ${DESTDIR}; it was renamed to avoid
    conflicts with the de facto standard usage of ${DESTDIR}.

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

fix missed makefile

====================

update another missed file

====================

eliminate bogus escaping

====================

get rid of another bogus escape

====================

remove unused include directory

====================

get rid of no longer used libdir

====================

remove unused incdir

====================

fix up some problems for make compatibility and missing trailing /

====================

put afs.exp in the right place

====================

remove bogus afssrvdir reference

====================

update dest version of output

====================

update ref to libexecdir

====================

kill refs to afssrvdir

====================

convert missed LIBDIR to TOP_LIBDIR

====================

remove explicit INSTALL and use makefile.@sys value instead

====================

except it's helpful to actually include makefile.@sys

====================

convert LIBDIR to TOP_LIBDIR

22 years agobucoord-cvs-bogosity-20010910
Derrick Brashear [Mon, 10 Sep 2001 17:37:32 +0000]
bucoord-cvs-bogosity-20010910

correct bogosity which crept into CVS (rtruct instead of struct)

22 years agowindows-install-updates-20010910
James Peterson [Mon, 10 Sep 2001 16:39:47 +0000]
windows-install-updates-20010910

more corrections for windows packaging

22 years agowinnt-makefile-fixes-20010908
Walter Wong [Sat, 8 Sep 2001 05:31:22 +0000]
winnt-makefile-fixes-20010908

clean up makefile problems which got introduced by other changes

22 years agorx-do-housekeeping-on-packets-with-new-flags-20010907
Nickolai Zeldovich [Fri, 7 Sep 2001 21:47:57 +0000]
rx-do-housekeeping-on-packets-with-new-flags-20010907

keep better track of packet state using flags in packet header indicating
which packets are in acked state and which are free

22 years agoafs-callback-plug-small-memory-leaks-20010907
Nickolai Zeldovich [Fri, 7 Sep 2001 21:12:13 +0000]
afs-callback-plug-small-memory-leaks-20010907

2 RPC stubs have small memory leaks since the actual string length is being
freed and not the whole allocated chunk

22 years agodarwin-make-afsd-script-exit-for-failure-20010907
Chaskiel M Grundman [Fri, 7 Sep 2001 19:59:03 +0000]
darwin-make-afsd-script-exit-for-failure-20010907

script failed to exit for one of the failure cases before

22 years agopam-afs-new-features-20010907
Carsten Jacobi [Fri, 7 Sep 2001 05:36:41 +0000]
pam-afs-new-features-20010907

add ignore_uid (like ignore_root) plus set_token (set token in auth step instead of setcred), refresh_token (no new pag), use_klog (fork a klog child), no_unlog, remainlifetime (sleep before deleting creds at logout)

22 years agomacosx-dont-interfere-with-cd-audio-20010906
Chaskiel M Grundman [Fri, 7 Sep 2001 05:00:05 +0000]
macosx-dont-interfere-with-cd-audio-20010906

don't use constant VT_AFS for our vfs type

22 years agoparam-remove-nested-commenting-20010906
Derrick Brashear [Fri, 7 Sep 2001 04:49:16 +0000]
param-remove-nested-commenting-20010906

based on comment from rjs3@andrew.cmu.edu

22 years agowinnt-more-makefile-fixes-for-cvs-head-20010906
Walter Wong [Fri, 7 Sep 2001 04:38:19 +0000]
winnt-more-makefile-fixes-for-cvs-head-20010906

take care of some places where we got out of sync between unix and windows,
add media target in afs_setup_utils

22 years agowinnt-initialize-variables-and-memory-before-use-20010906
Walter Wong [Fri, 7 Sep 2001 04:32:56 +0000]
winnt-initialize-variables-and-memory-before-use-20010906

initialize a variable false when assumed; also some memset(...,0

22 years agodirpath-support-configure-path-options-and-decanonicalization-20010906
Sam Hartman [Fri, 7 Sep 2001 04:18:59 +0000]
dirpath-support-configure-path-options-and-decanonicalization-20010906

Sam's notes:
"Here is a patch that supports directory paths, documentation for the
  above and decanonicalization of of supplied paths.  Here are tests I
  have run:

  * Build with  both transarc and non-transarc paths and examined
    dirpath.o
  * Build with roughly FHS-style paths and tested on Debian
  * confirmed that bos salvage  works even when salvager not in
    /usr/afs/bin
  (it gets the log correctly too)
  * confirms that bos getlog can get logs from /usr/afs/logs even when
    /ur/afs/logs
  is /var/lib/openafs/logs
  * confirmed that bos getlog can get /etc/motd
"

22 years agoafs-sysnames-add-sunx86-57-and-58-20010906
Derrick Brashear [Fri, 7 Sep 2001 04:05:43 +0000]
afs-sysnames-add-sunx86-57-and-58-20010906

allocate ids for sunx86 57 and 58

22 years agowindows-build-environment-cleanup-20010906
James Peterson [Fri, 7 Sep 2001 03:54:54 +0000]
windows-build-environment-cleanup-20010906

clean up previous problems in the win2k build env

22 years agolinux22-redhat-enterprise-kernels-add-inode-fields-20010906
Derrick Brashear [Fri, 7 Sep 2001 02:12:29 +0000]
linux22-redhat-enterprise-kernels-add-inode-fields-20010906

struct vm_area_struct   *i_mmap_shared;
added in a patch to the redhat enterprise kernel

22 years agolinux24-fix-path-leak-20010906
Chaskiel M Grundman [Thu, 6 Sep 2001 20:07:12 +0000]
linux24-fix-path-leak-20010906

based on suggestion from Srikanth and Chas Williams

22 years agoupdate-top-readme-20010904
Sam Hartman [Tue, 4 Sep 2001 23:28:20 +0000]
update-top-readme-20010904

add ppc linux stuff and update alpha linux

22 years agolibafs-linux-makefileproto-shouldnt-default-to-kgcc-20010830
Derrick Brashear [Thu, 30 Aug 2001 07:14:37 +0000]
libafs-linux-makefileproto-shouldnt-default-to-kgcc-20010830

nuke kgcc crock *again*

22 years agoconfig-makefile-further-cleanup-20010830
Derrick Brashear [Thu, 30 Aug 2001 07:07:16 +0000]
config-makefile-further-cleanup-20010830

ordering of dependancies caused ${DESTDIR}/include/afs to not exist in
time, masked if you built into an existing destdir. so let's make
the makefile simpler and more logically paraallel.

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

get missing backquote

22 years agomore-makefiles-cleanup-20010829
Nathan Neulinger [Thu, 30 Aug 2001 00:22:17 +0000]
more-makefiles-cleanup-20010829

get rid of cases where we'd end up with two / because of DESTDIR pasting

macroize calls to rm

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

get rid of another case of bogus trailing quoting

22 years agolibafs-make-subtree-during-build-for-later-use-20010829
Jeffrey Hutzelman [Wed, 29 Aug 2001 23:10:45 +0000]
libafs-make-subtree-during-build-for-later-use-20010829

get rid of bogus ..'s

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

make the subdir buildable

22 years agolibafs-irix-makefile-cleanup-20010829
Derrick Brashear [Wed, 29 Aug 2001 22:31:38 +0000]
libafs-irix-makefile-cleanup-20010829

clean up typo so we build correct module

22 years agolibafs-make-subtree-during-build-for-later-use-20010829
Jeffrey Hutzelman [Wed, 29 Aug 2001 20:17:19 +0000]
libafs-make-subtree-during-build-for-later-use-20010829

at build time, create a subtree for use later which can be used to
build *only* kernel modules, but only for the afs sysname you were building
for in the first place currently.

22 years agolinux-support-kernel-patches-used-by-suse-20010829
Derrick Brashear [Wed, 29 Aug 2001 20:04:00 +0000]
linux-support-kernel-patches-used-by-suse-20010829

check for added fields in struct inode

22 years agocvs-needs-readme-20010829
Derrick Brashear [Wed, 29 Aug 2001 19:45:37 +0000]
cvs-needs-readme-20010829

warn of the dangers of cvs snapshots
explain how to use one

22 years agovenus-makefile-cleanup-20010829
Nathan Neulinger [Wed, 29 Aug 2001 18:40:51 +0000]
venus-makefile-cleanup-20010829

cleanup and remove now unused for/do constructs

22 years agoafsd-makefile-cleanup-20010829
Nathan Neulinger [Wed, 29 Aug 2001 18:37:32 +0000]
afsd-makefile-cleanup-20010829

cleanup for afsd makefile

22 years agolinux24-make-kdump-build-20010829
Nathan Neulinger [Wed, 29 Aug 2001 18:36:27 +0000]
linux24-make-kdump-build-20010829

make kdump at least buildable, not particularly pretty

22 years agodux-documentation-update-20010829
Derrick Brashear [Wed, 29 Aug 2001 18:12:35 +0000]
dux-documentation-update-20010829

based on information from joda@pdc.kth.se

22 years agodux-create-loadable-kernel-module-20010829
Johan Danielsson [Wed, 29 Aug 2001 17:55:56 +0000]
dux-create-loadable-kernel-module-20010829

with minor correction by shadow@dementia.org to install the module, but
we now create a loadable kernel module for dux

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

actually build afs.mod

====================

fix typo

22 years agoxdr-int64-dont-build-on-dux-ever-20010829
Derrick Brashear [Wed, 29 Aug 2001 17:14:49 +0000]
xdr-int64-dont-build-on-dux-ever-20010829

some form of xdr_int64 is always available on dux, so don't build it.

22 years agompp-needs-unistd-for-protos-on-irix-20010828
Derrick Brashear [Wed, 29 Aug 2001 04:59:38 +0000]
mpp-needs-unistd-for-protos-on-irix-20010828

so we don't get the implicit int return type and upset the irix compiler

22 years agomakefile-updates-20010828
Nathan Neulinger [Wed, 29 Aug 2001 02:19:55 +0000]
makefile-updates-20010828

clean up all the makefiles to remove bogus targets, eliminate trailing /
requirement from DESTDIR, avoid needing to pass things like DESTDIR around
between makefiles

====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================

remove bogus quoting

22 years agolinux-call-afs-dopartialwrite-from-afs-memwrite-and-afs-ufswrite-not-afs-linux-write...
Derrick Brashear [Wed, 29 Aug 2001 01:46:57 +0000]
linux-call-afs-dopartialwrite-from-afs-memwrite-and-afs-ufswrite-not-afs-linux-write-20010828

attempt to deal with case where data not being flushed from cache

22 years agolinux-attempt-to-support-forcible-umount-20010828
Derrick Brashear [Tue, 28 Aug 2001 19:03:49 +0000]
linux-attempt-to-support-forcible-umount-20010828

not well tested, likely neither complete nor 100% correct.

22 years agosolaris-kernel-lacks-mem-functions-20010828
Derrick Brashear [Tue, 28 Aug 2001 15:05:34 +0000]
solaris-kernel-lacks-mem-functions-20010828

this is a band-aid, since it maps all memsets to bzero; a replacement
memset is actually needed for the non zero case

22 years agoafs-evalmountpoint-return-code-resync-20010826
Nickolai Zeldovich [Mon, 27 Aug 2001 04:35:06 +0000]
afs-evalmountpoint-return-code-resync-20010826

resync error codes so they work like they did before afs-lookup-check-evalmountpoint-return-20010221

22 years agowindows-authentication-recognize-time-out-of-bounds-error-20010823
Derrick Brashear [Thu, 23 Aug 2001 18:46:32 +0000]
windows-authentication-recognize-time-out-of-bounds-error-20010823

so people don't get "unknown" error 37

22 years agolinux-kernel-headers-have-conflicting-min-max-defs-20010823
Derrick Brashear [Thu, 23 Aug 2001 18:26:43 +0000]
linux-kernel-headers-have-conflicting-min-max-defs-20010823

so we cheat and call our macros something else.
based on report by Owen LeBlanc <LeBlanc@mcc.ac.uk>

22 years agowindows-installer-updates-20010823
Chas Williams [Thu, 23 Aug 2001 18:17:40 +0000]
windows-installer-updates-20010823

change references to be to openafs and not afs

22 years agohpux10-may-be-hppa-2-20010823
Nathan Neulinger [Thu, 23 Aug 2001 18:07:24 +0000]
hpux10-may-be-hppa-2-20010823

deal with more uname output