openafs.git
19 years agoProperly AFS_GUNLOCK() when returning an error after having previously
Nickolai Zeldovich [Fri, 8 Feb 2002 07:05:34 +0000]
Properly AFS_GUNLOCK() when returning an error after having previously
AFS_GLOCK()'ed, in afs_mount() and afs_unmount().

19 years agoInclude linux/sysctl.h after afsconfig.h, since we need the
Nickolai Zeldovich [Thu, 7 Feb 2002 03:07:23 +0000]
Include linux/sysctl.h after afsconfig.h, since we need the
modversion stuff first.

19 years agolinux-provide-sysctl-interface-20020206
Kris Van Hees [Thu, 7 Feb 2002 00:06:35 +0000]
linux-provide-sysctl-interface-20020206

provide sysctl interface for toggling module state. provide hooks for
all currently known tweakable items

19 years agosolaris-p-crlock-not-safe-to-hold-across-call-to-crset-20020206
Nickolai Zeldovich [Wed, 6 Feb 2002 23:57:44 +0000]
solaris-p-crlock-not-safe-to-hold-across-call-to-crset-20020206

crset() expects p_crlock to not be held when it's called; deal accordingly

19 years agonewvolumebyname-analyze-the-vrequest-actually-used-20020205
Derrick Brashear [Tue, 5 Feb 2002 21:58:03 +0000]
newvolumebyname-analyze-the-vrequest-actually-used-20020205

it's ok to support a null request being passed in, but if a real one is
passed in we shouldn't analyze it when it's not the one we used when
making calls.

19 years agoAdd the necessary includes for AFS_ISGLOCK, AFS_GLOCK, AFS_GUNLOCK.
Nickolai Zeldovich [Sat, 2 Feb 2002 18:33:49 +0000]
Add the necessary includes for AFS_ISGLOCK, AFS_GLOCK, AFS_GUNLOCK.

19 years agoviced-callback-avoid-potential-looping-problem-20020201
Hartmut Reuter [Fri, 1 Feb 2002 22:12:51 +0000]
viced-callback-avoid-potential-looping-problem-20020201

avoid case where we never come out of loop attempting to clear host cvallbacks

19 years agoFirst pass at better signal handling:
Nickolai Zeldovich [Fri, 1 Feb 2002 20:29:59 +0000]
First pass at better signal handling:

  -- Add afs_osi_SleepSig, which returns EINTR when interrupted by signal.
  -- Similar CV_WAIT_SIG for Rx code.
  -- Implement afs_osi_SleepSig and CV_WAIT_SIG for Linux, Solaris.
  -- Use afs_osi_SleepSig in afs_{UFS,Mem}Read, allowing user to interrupt
     an AFS read if it's being handled by background daemon.
  -- Balance the number of open & close braces in CV_TIMEDWAIT macros,
     even though noone uses them.
  -- Untangle parts of the macro forest by moving some of the Rx CV code
     into rx_kmutex.c.

19 years agoBlock signals when sleeping in CV_WAIT() and CV_TIMEDWAIT(), similar
Nickolai Zeldovich [Wed, 30 Jan 2002 21:13:48 +0000]
Block signals when sleeping in CV_WAIT() and CV_TIMEDWAIT(), similar
to the linux-avoid-signal-wakeups-while-sleeping-20020129 change for
afs_osi_Sleep().

19 years agoAvoid deadlocking with afs_ResetVolumeInfo in InstallVolumeInfo and
Nickolai Zeldovich [Wed, 30 Jan 2002 16:16:23 +0000]
Avoid deadlocking with afs_ResetVolumeInfo in InstallVolumeInfo and
InstallNVolumeInfo, like we already do in InstallUVolumeInfo.

19 years agoreadd-t-uniq-bad-as-binary-20020129
Derrick Brashear [Tue, 29 Jan 2002 22:34:13 +0000]
readd-t-uniq-bad-as-binary-20020129

oops

19 years agoBlock all signals while sleeping; this prevents 100% cpu usage
Nickolai Zeldovich [Tue, 29 Jan 2002 17:36:20 +0000]
Block all signals while sleeping; this prevents 100% cpu usage
when a process waiting for an afs operation to complete receives
a signal.

19 years agoProperly release GLOCK when trying to mount AFS a second time.
Nickolai Zeldovich [Tue, 29 Jan 2002 17:26:26 +0000]
Properly release GLOCK when trying to mount AFS a second time.

19 years agoUse filemap_fdata{sync,wait} directly (available in 2.4.5 and above)
Nickolai Zeldovich [Tue, 29 Jan 2002 16:07:27 +0000]
Use filemap_fdata{sync,wait} directly (available in 2.4.5 and above)
rather than write_inode_now(), which steals afs inodes and puts them
on the global inode lists.

19 years agoconfigure-add-darwin-52-support-as-14-since-no-change-otherwise-20020128
Derrick Brashear [Mon, 28 Jan 2002 21:32:20 +0000]
configure-add-darwin-52-support-as-14-since-no-change-otherwise-20020128

it was previously agreed that we wouldn't add gratuitous new sysnames unless things
(not just what the vendor calls the operating system) changed. precedent is
sun4m_412 still being "current" for sun4m_413, etc.

19 years agorx-dont-include-arraysize-in-offsetof-20020125
Nickolai Zeldovich [Sat, 26 Jan 2002 03:40:30 +0000]
rx-dont-include-arraysize-in-offsetof-20020125

so this compiles on irix

19 years agolinux-afsd-pids-block-signals-so-we-dont-need-to-flush-20020124
Derrick Brashear [Thu, 24 Jan 2002 20:13:22 +0000]
linux-afsd-pids-block-signals-so-we-dont-need-to-flush-20020124

no point in keeping this around

19 years agolinux-conf-fix-type-which-crept-in-20020124
David Howells [Thu, 24 Jan 2002 18:09:28 +0000]
linux-conf-fix-type-which-crept-in-20020124

"oops"

19 years agosolaris-rx-shutdown-sleep-only-if-listener-pid-not-freed-20020124
Derrick Brashear [Thu, 24 Jan 2002 10:46:42 +0000]
solaris-rx-shutdown-sleep-only-if-listener-pid-not-freed-20020124

replace if with while in case the listener is freed before we call sleep

19 years agolinux-unmask-signals-in-rxk-listener-at-shutdown-20020124
Derrick Brashear [Thu, 24 Jan 2002 10:42:34 +0000]
linux-unmask-signals-in-rxk-listener-at-shutdown-20020124

since the rxk listener gets "kill"d at shutdown we need to unmask to wake
it up. this should probably be replaced but the logical way is evil,
involves simulating a packet to the listener so it wakes up and notices it's
time to die

19 years agolinux-cant-daemonize-due-to-exit-mm-call-oh-well-20020124
Derrick Brashear [Thu, 24 Jan 2002 10:09:13 +0000]
linux-cant-daemonize-due-to-exit-mm-call-oh-well-20020124

the call to daemonize() means we fail to shut down correctly on SMP machines;
this is due to the call to exit_mm() in daemonize()

19 years agolinux-rework-signal-blocking-for-afsdb-handler-and-clean-up-osi-invisible-before...
Derrick Brashear [Wed, 23 Jan 2002 19:27:39 +0000]
linux-rework-signal-blocking-for-afsdb-handler-and-clean-up-osi-invisible-before-im-completely-blind-20020123

based on suggestion from kolya@mit.edu

and osi_Invisible cleanup so ifdef logic would be more clear

19 years agolinux-use-while-loop-and-also-do-it-for-22-waiting-to-close-socket-20020123
Derrick Brashear [Wed, 23 Jan 2002 18:49:59 +0000]
linux-use-while-loop-and-also-do-it-for-22-waiting-to-close-socket-20020123

while loop idea from chas@cmf.nrl.navy.mil
reminder that we should worry about 2.2 from somkar@in.ibm.com

19 years agosolaris-wait-for-listenerproc-to-exit-20020123
Omkar Sathe [Wed, 23 Jan 2002 18:48:18 +0000]
solaris-wait-for-listenerproc-to-exit-20020123

like linux, actually wait for listenerproc to finish before freeing the socket

19 years agorework-linux-newvcache-clean-up-dentries-20020123
Derrick Brashear [Wed, 23 Jan 2002 09:21:54 +0000]
rework-linux-newvcache-clean-up-dentries-20020123

previous pass wasn't SMP-happy. rework closer to original code.

19 years agolinux-afsd-ignore-signals-20020123
Chaskiel M Grundman [Wed, 23 Jan 2002 07:55:08 +0000]
linux-afsd-ignore-signals-20020123

ignore signals sent to afsd on linux instead of running out of control

also daemonize ourselves

19 years agolinux24-add-configure-glue-for-new-inode-fields-20020108
Derrick Brashear [Wed, 23 Jan 2002 05:09:11 +0000]
linux24-add-configure-glue-for-new-inode-fields-20020108

sure would be nice if it compiled

19 years agolinux-use-results-of-all-inode-configure-tests-20020122
Derrick Brashear [Wed, 23 Jan 2002 00:43:35 +0000]
linux-use-results-of-all-inode-configure-tests-20020122

we tested for and then failed to use things before

19 years agohpux11-wants-system-ar-20020122
Nathan Neulinger [Tue, 22 Jan 2002 23:31:06 +0000]
hpux11-wants-system-ar-20020122

gnu ar hates empty .o files

19 years agolinux-fix-mmap-write-20020122
Nickolai Zeldovich [Tue, 22 Jan 2002 05:51:19 +0000]
linux-fix-mmap-write-20020122

a MAP_SHARED write wasn't being flushed back to the server

19 years agorx-cvsignore-rxperf-20020121
Love Hörnquist-Åstrand [Mon, 21 Jan 2002 19:34:18 +0000]
rx-cvsignore-rxperf-20020121

cvs shpuld ignore rxperf

19 years agorx-cleanup-deadlock-and-refcnt-leak-20020121
Nickolai Zeldovich [Mon, 21 Jan 2002 18:05:26 +0000]
rx-cleanup-deadlock-and-refcnt-leak-20020121

The first part of the patch fixes a likely deadlock in the
case when we're destroying a client connection with active
calls.

The second part fixes a minor connection refcount leak in
my recent modification to protect against half-reachable
clients.

19 years agosource-map-20020121
Matthew N. Andrews [Mon, 21 Jan 2002 10:24:12 +0000]
source-map-20020121

with modifications by shadow@dementia.org

a guide to what's where in the OpenAFS source

====================
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.
====================

with modifications by shadow@dementia.org

a guide to what's where in the OpenAFS source

this time the right file. a sure sign that it's bedtime.

19 years agoi-hate-autoconf-20020120
Derrick Brashear [Mon, 21 Jan 2002 09:14:21 +0000]
i-hate-autoconf-20020120

AC_PROG_CC not safe to call this early

19 years agorx-cleanup-gettimeofday-use-20020120
Love Hörnquist-Åstrand [Sun, 20 Jan 2002 10:35:21 +0000]
rx-cleanup-gettimeofday-use-20020120

don't blindly make assumptions about gettimeofday

====================
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.
====================

redo in a way that doesn't screw up with respect to libafsrpc and libafsrpc-
using suites

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

redo in a way that doesn't screw up with respect to libafsrpc and libafsrpc-
using suites

19 years agoalpha-netbsd-param-fixes-20020120
Love Hörnquist-Åstrand [Sun, 20 Jan 2002 08:04:44 +0000]
alpha-netbsd-param-fixes-20020120

make alpha little endian and remove duplicate define

19 years agorevert-rx-deal-with-missed-keepalives-by-accurately-tracking-call-number-20020120
Derrick Brashear [Sun, 20 Jan 2002 05:42:28 +0000]
revert-rx-deal-with-missed-keepalives-by-accurately-tracking-call-number-20020120

causes problems at minimum for linux

19 years agodes-support-alpha-bsd-and-fix-test-programs-20020119
Love Hörnquist-Åstrand [Sun, 20 Jan 2002 05:19:08 +0000]
des-support-alpha-bsd-and-fix-test-programs-20020119

All BSDs are not I386s.

Also include a patch that make sure that data are alligned on alpha. All
bsd's will fix it up in kernel when used, but you'll get a huge performace
penalty

Use afs_int32 in the des programs to make sure they work and make them
build again.

19 years agorxkad-make-stress-compile-again-20020119
Love Hörnquist-Åstrand [Sun, 20 Jan 2002 02:17:12 +0000]
rxkad-make-stress-compile-again-20020119

Make stress compile again.

19 years agobosserver-nofork-20020119
Love Hörnquist-Åstrand [Sat, 19 Jan 2002 16:31:25 +0000]
bosserver-nofork-20020119

Make it easier to debug bosserver by adding a -nofork flag for the !nt40
env's.

19 years agorx-protect-servers-from-half-reachable-clients-20020119
Nickolai Zeldovich [Sat, 19 Jan 2002 16:28:33 +0000]
rx-protect-servers-from-half-reachable-clients-20020119

This patch should protect Rx-based servers from half-reachable clients,
which issue requests but don't acknowledge the server's response, thereby
eventually tying up all of the server threads, and denying service to all
other clients.  Such clients can arise in case of uni-directional routing
failures, whereby all packets from the server to client are lost but not
the other way around.

The idea it to ping clients (using Rx ack ping) before attaching them to
a thread, if (a) we're running low on threads, and (b) the client hasn't
responsed to a ping recently.

19 years agoalpha-cleanup-20020119
Love Hörnquist-Åstrand [Sat, 19 Jan 2002 09:25:19 +0000]
alpha-cleanup-20020119

make alpha AFS_64BIT_ENV
Avoid warning by not passing data inside a pointer in the test program rw.

Provide compatibility symbols for those w/o preempt hack so its possible to
link test program `rw'.

When comparing to lwp_cpptr->statcheck, cast to int, since that is the type
of lwp_cpptr->statcheck. It may be the case that sizeof(int) !=
sizeof(afs_int32), but then other things will break too.

clean tests with -f

====================
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.
====================

and unbreak linux

19 years agodes-fix-make-p-table-on-alpha-20020119
Love Hörnquist-Åstrand [Sat, 19 Jan 2002 09:21:12 +0000]
des-fix-make-p-table-on-alpha-20020119

if printing with a long (l) unsigned (X), make sure to pass such a value as
  an argument (p_prime is of int type on alpha).

19 years agobozo-misc-cleanup-20020118
Love Hörnquist-Åstrand [Fri, 18 Jan 2002 21:03:32 +0000]
bozo-misc-cleanup-20020118

- remove bos_util and install it
- ignore it

19 years agonetbsd-alpha-more-patches-20020118
Love Hörnquist-Åstrand [Fri, 18 Jan 2002 20:59:26 +0000]
netbsd-alpha-more-patches-20020118

get closer to server working

19 years agolargefile-dump-support-20020118
Nickolai Zeldovich [Fri, 18 Jan 2002 18:47:00 +0000]
largefile-dump-support-20020118

64bit file support for vos dump

19 years agonetbsd-alpha-start-at-server-support-20020118
Love Hörnquist-Åstrand [Fri, 18 Jan 2002 18:45:27 +0000]
netbsd-alpha-start-at-server-support-20020118

not ready yet
adds the concept of a common OS param file

19 years agotest-suite-update-README-20020117
Derrick Brashear [Thu, 17 Jan 2002 10:49:35 +0000]
test-suite-update-README-20020117

reflect new test, ordering

19 years agotestsuite-arrange-for-bad-volume-dump-to-be-in-useful-location-20020117
Derrick Brashear [Thu, 17 Jan 2002 10:47:20 +0000]
testsuite-arrange-for-bad-volume-dump-to-be-in-useful-location-20020117

since we run the script not from the script dir we need to know where it lives, or do this.

19 years agotop-makefile-build-runtests-20020117
Derrick Brashear [Thu, 17 Jan 2002 10:43:28 +0000]
top-makefile-build-runtests-20020117

make top makefile/configure build run-tests

19 years agotest-suite-frontend-mkdir-afs-20020117
Derrick Brashear [Thu, 17 Jan 2002 10:10:37 +0000]
test-suite-frontend-mkdir-afs-20020117

mkdir /afs in case

19 years agotest-suite-create-cachedir-20020117
Derrick Brashear [Thu, 17 Jan 2002 09:33:06 +0000]
test-suite-create-cachedir-20020117

take care of case where no cache dir exists yet

19 years agotest-suite-check-for-salvager-correctly-getting-a-bad-uniquifier-20020117
Derrick Brashear [Thu, 17 Jan 2002 09:24:18 +0000]
test-suite-check-for-salvager-correctly-getting-a-bad-uniquifier-20020117

there are cases where a vnode has a uniq which is different than that in the parent. the salvager should clean it up.

19 years agotestsuite-frontend-restart-bosserver-20020117
Derrick Brashear [Thu, 17 Jan 2002 07:52:31 +0000]
testsuite-frontend-restart-bosserver-20020117

so bosserver rereads configs

19 years agosuite-be-less-verbose-20020117
Derrick Brashear [Thu, 17 Jan 2002 07:25:52 +0000]
suite-be-less-verbose-20020117

toss more output in the non-verbose case

19 years agoka-util-differently-useless-20020117
Derrick Brashear [Thu, 17 Jan 2002 06:47:28 +0000]
ka-util-differently-useless-20020117

drop state here

19 years agotest-script-substitute-fs-path-20020117
Derrick Brashear [Thu, 17 Jan 2002 05:08:45 +0000]
test-script-substitute-fs-path-20020117

don't assume /usr/local/bin

19 years agoupdate-suite-frontend-20020117
Derrick Brashear [Thu, 17 Jan 2002 05:01:57 +0000]
update-suite-frontend-20020117

fix to use the defined mechanisms to set up bos foo

19 years agotest-suite-pick-up-replica-of-root-afs-20020116
Derrick Brashear [Thu, 17 Jan 2002 02:26:16 +0000]
test-suite-pick-up-replica-of-root-afs-20020116

run in replicated root

19 years agorx-dont-challenge-forever-20020116
Nickolai Zeldovich [Wed, 16 Jan 2002 22:02:36 +0000]
rx-dont-challenge-forever-20020116

don't repeat challenges forever. give up after 50 tries

19 years agonfs-translator-call-lookup-with-correct-form-of-dir-20020116
Derrick Brashear [Wed, 16 Jan 2002 09:27:04 +0000]
nfs-translator-call-lookup-with-correct-form-of-dir-20020116

based on information from Kris Van Hees

still untested

19 years agointroduce-ka-util-20020116
Derrick Brashear [Wed, 16 Jan 2002 09:23:46 +0000]
introduce-ka-util-20020116

the start of a tool to write out a kaserver database. currently dumps but
cannot create a kaserver.DB0

19 years agovos-extra-info-but-disabled-20020115
Jeffrey Hutzelman [Wed, 16 Jan 2002 02:16:26 +0000]
vos-extra-info-but-disabled-20020115

extra info in vos examine, but disabled until we figure out a way
not to irritate things which parse vos examine output

19 years agolinux-dcache-inconsistency-fix-20020115
Chas Williams [Wed, 16 Jan 2002 02:08:43 +0000]
linux-dcache-inconsistency-fix-20020115

minor rework on a patch from Ted Anderson (ota@transarc.com) to
fit OpenAFS

19 years agolinux-newvcache-clean-up-dentries-20020115
Ted Anderson [Wed, 16 Jan 2002 02:03:49 +0000]
linux-newvcache-clean-up-dentries-20020115

With work and feedback from Omkar Sathe <somkar@in.ibm.com> and
Srikanth Vishwanathan <vsrikanth@in.ibm.com>

19 years agorxi-newcall-avoid-race-20020115
Nickolai Zeldovich [Wed, 16 Jan 2002 00:56:02 +0000]
rxi-newcall-avoid-race-20020115

avoid race when creating new calls

19 years agoopenafs-tools-cmd-20020115
Jeremy Stribling [Wed, 16 Jan 2002 00:49:29 +0000]
openafs-tools-cmd-20020115

Tools for configuring a new cell and for building RPMs of said tools

19 years agotest-suite-updates-20020115
Derrick Brashear [Tue, 15 Jan 2002 19:22:05 +0000]
test-suite-updates-20020115

cleanup all target, get missing headers

====================
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 perl modules in properly named subdir

19 years agodeal-with-linux-large-uids-20020115
Derrick Brashear [Tue, 15 Jan 2002 17:17:31 +0000]
deal-with-linux-large-uids-20020115

based on feedback from Hartmut Reuter <reuter@rzg.mpg.de>

2.2 headers define the right things for this to work

19 years agotest-suite-cvsignore-20020115
Derrick Brashear [Tue, 15 Jan 2002 16:33:08 +0000]
test-suite-cvsignore-20020115

we need one

19 years agotest-suite-update-20020115
Derrick Brashear [Tue, 15 Jan 2002 16:30:55 +0000]
test-suite-update-20020115

front-end script to tie together the test suite

====================
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 makefile

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

stop bos bnodes before deleting

19 years agotest-suite-pull-tools-directly-in-20020114
Derrick Brashear [Tue, 15 Jan 2002 04:14:51 +0000]
test-suite-pull-tools-directly-in-20020114

move dump tools directly into test suite as that is their intended use

canonical versions of these tools will be distributed otherwise
and may be updated but these provide the minimum functionality

19 years agofunctionality-test-suite-20020114
Derrick Brashear [Mon, 14 Jan 2002 07:49:15 +0000]
functionality-test-suite-20020114

Based on Arla test suite and AFSTools with some other additions;
intended to set up a standalone cell, perform operations

19 years agodumptools-20020113
Jeffrey Hutzelman [Sun, 13 Jan 2002 23:47:17 +0000]
dumptools-20020113

tools for manipulating afs volume dumps

19 years agoalpha-linux-build-fixes-20020113
Steven N . Hirsch [Sun, 13 Jan 2002 17:06:42 +0000]
alpha-linux-build-fixes-20020113

so alpha linux builds work again

19 years agobos-key-utility-20020112
Derrick Brashear [Sat, 12 Jan 2002 21:35:59 +0000]
bos-key-utility-20020112

modification of asetkey to allow manipulation of key files without
a running bosserver (asetkey requires krb4; this does not)

19 years agolinux-remove-hazardous-trace-events-20020111
Derrick Brashear [Fri, 11 Jan 2002 16:44:53 +0000]
linux-remove-hazardous-trace-events-20020111

use of osi_Sleep in tracing code makes this dangerous.

19 years agosolaris-obtain-locks-under-GLOCK-20020101
Nickolai Zeldovich [Thu, 10 Jan 2002 06:42:21 +0000]
solaris-obtain-locks-under-GLOCK-20020101

Fix the two places in Solaris afs_GetOnePage where locks
  are obtained/released without GLOCK.

====================
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 the two places in Solaris afs_GetOnePage where locks
  are obtained/released without GLOCK.

(Fix error in application of patch)

19 years agoREADME-NT-updates-20020108
Jeremy Stribling [Wed, 9 Jan 2002 03:08:16 +0000]
README-NT-updates-20020108

info on how to set up debugging environment and other updates

19 years agolock-tracing-20020108
Nickolai Zeldovich [Tue, 8 Jan 2002 20:54:12 +0000]
lock-tracing-20020108

patches to allow tracing with fstrace of afs locks

19 years agowindows-updates-20010108
James Peterson [Tue, 8 Jan 2002 20:44:31 +0000]
windows-updates-20010108

"1. The default Open AFS is set to normal security (doesn't generate random
user names).
If you are installing over a previous version (before 1.2.2b) it's default
is
high security; therefore, if you want the normal security, you should
uninstall the previous version (1.2.2a or earlier) and select to 'Not
Preserve previous settings'.

To manually change security you need to set the following registry keys:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\TransarcAFSDaemond\NetworkProvider
     LogonOptions = 1 - Integrated Logon
     LogonOptions = 2 - High Security options, Random User name generation
     LogonOptions = 3 - both

3. Windows 2000/NT, Win9x - First time installations will create necessary
directories when user decides to download CellServDB

4. Windows 2000/NT, Global Drive working.

5. Windows XP - Drive mapping via GUI working.

6. Rename pthread.dll to afspthread.dll"

19 years agolinux24-add-configure-glue-for-new-inode-fields-20020108
Derrick Brashear [Tue, 8 Jan 2002 20:41:11 +0000]
linux24-add-configure-glue-for-new-inode-fields-20020108

based on report from cg@cdegroot.com

19 years agoupdate-readme-devel-for-parallel-make-hints-20020101
Nathan Neulinger [Wed, 2 Jan 2002 04:12:20 +0000]
update-readme-devel-for-parallel-make-hints-20020101

suggestions which make parallel make easier to support

19 years agovlserver-remove-duplicate-targets-20020101
Nickolai Zeldovich [Wed, 2 Jan 2002 04:08:37 +0000]
vlserver-remove-duplicate-targets-20020101

"duplicates seem to confuse /usr/bin/make on MacOSX"

19 years agonetbsd-initial-support-20020101
Love Hörnquist-Åstrand [Tue, 1 Jan 2002 21:02:50 +0000]
netbsd-initial-support-20020101

update parameters so it will build

19 years agolinux-sparc-24-kernel-support-20020101
Mark W. Eichin [Tue, 1 Jan 2002 19:09:59 +0000]
linux-sparc-24-kernel-support-20020101

config for linux 2.4 kernels on sparc

19 years agoacinclude-update-for-new-darwin-version-strings-20020101
Ted McCabe [Tue, 1 Jan 2002 19:01:54 +0000]
acinclude-update-for-new-darwin-version-strings-20020101

so we can build for new darwin versions

19 years agonetbsd-server-support-20020101
Love Hörnquist-Åstrand [Tue, 1 Jan 2002 18:57:18 +0000]
netbsd-server-support-20020101

support for fileserver (namei) on netbsd

====================
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.
====================

like openbsd, stripping on netbsd strips libraries

19 years agolinux-osi-sleep-avoid-forgetting-events-20020101
Chas Williams [Tue, 1 Jan 2002 18:49:18 +0000]
linux-osi-sleep-avoid-forgetting-events-20020101

based on suggestion from Ted Anderson
"the changes make more sense
than the code as it currently exists.  the only think i am nervous
about is the dontSleep delete.  while it makes more sense to just
not wakeup sleepers if none exist, i suppose its possible that
some bit of afs code wants acausal (wake before sleep) events.
that does seem quite unlikely.  just looking at the sleep on
solaris, it checks the seq number to get the next event not
a previous event.

i imported the changes and make the fixup in osi_stoplistener().
i dropped some of the silly syntax changes that junked up the
diff -- this makes it a bit easier to see what was changed.
i just added an assert in afs_addevent for quality assurance
purposes."

====================
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 for osi_StopListener so it does the rigth thing

19 years agolinux22-also-alloc-avoid-potential-recursion-freeing-memory-and-schedule-when-vmalloc...
Nickolai Zeldovich [Sun, 30 Dec 2001 00:07:02 +0000]
linux22-also-alloc-avoid-potential-recursion-freeing-memory-and-schedule-when-vmalloc-fails-20011229

linux 2.2 lacks set_current_state. currently in 2.4 it's a definition and not a real function, so this will work

19 years agolinux-alloc-avoid-potential-recursion-freeing-memory-and-schedule-when-vmalloc-fails...
David Howells [Tue, 25 Dec 2001 18:19:20 +0000]
linux-alloc-avoid-potential-recursion-freeing-memory-and-schedule-when-vmalloc-fails-20011225

"The first is to change the gfp_mask passed to kmalloc(). Using GFP_KERNEL,
  it is possible that the VM will call back to the filesystem to free up
  memory to satisfy the kmalloc request. GFP_NOFS will prevent this possible
  recursion. I believe GFP_NOFS first appeared in the 2.4.6 kernel.

  The second change involves the call to schedule() when vmalloc() fails. This
  can also cause a hang. The schedule() call could be replaced with:

  set_current_state(TASK_INTERRUPTIBLE);
  schedule_timeout(HZ);"

19 years agorx-avoid-unnecessary-wakeups-20011225
Nickolai Zeldovich [Tue, 25 Dec 2001 18:13:20 +0000]
rx-avoid-unnecessary-wakeups-20011225

"This fixes a livelock condition introduced in my earlier
  resource starvation patch; apparently I had erred too far
  on the side of "wake up just in case".  The livelock bug
  is exhibited when running 10 fsstress processes at once;
  if many processes are waiting for a new Rx call, they get
  stuck in an uninterruptible kernel loop waking each other
  up."

19 years agofix-dcache-deadlock-20011225
Nickolai Zeldovich [Tue, 25 Dec 2001 18:10:39 +0000]
fix-dcache-deadlock-20011225

"This patch fixes a deadlock in the new dcache locking scheme.
  The underlying bug apparently existed before, but due to the
  absence of locking, it probably resulted in spuriously high
  refcounts rather than deadlock.

  The problem happens when there are zero-length dcache entries
  associated with a file; this is demonstrated by fsx, which
  hangs after running on AFS for a while.  The writeback loop
  never releases dcache entries unless they're stored back to
  the server as part of a sequential byte range."

19 years agofix-formatting-in-afs-h-20011225
Nickolai Zeldovich [Tue, 25 Dec 2001 18:09:44 +0000]
fix-formatting-in-afs-h-20011225

minor formatting cleanup

19 years agodcache-dont-erroneously-fail-to-fill-in-tsmall-20011225
Nickolai Zeldovich [Tue, 25 Dec 2001 18:04:52 +0000]
dcache-dont-erroneously-fail-to-fill-in-tsmall-20011225

"The particular problem seems to be, when size
  is computed to be zero, tsmall is not filled in with valid data,
  and ProcessFS is called with a zeroed out OutStatus.  This causes
  the file to magically turn into a directory (VDIR), among other
  things"

"The second part of the patch doesn't fix any bug that I've ran into
  thus far, but seemed like a good idea while I was reading the code
  to find the former bug."

19 years agodoc-heimdal-conversion-howto-20011224
Derrick Brashear [Mon, 24 Dec 2001 21:19:07 +0000]
doc-heimdal-conversion-howto-20011224

short explanation of how to convert from a kaserver to a heimdal kdc
including setting up iprop

20 years agomake-storedata-use-unsigned-values-so-quota-cant-get-corrupted-20011203
Nickolai Zeldovich [Sat, 8 Dec 2001 02:29:24 +0000]
make-storedata-use-unsigned-values-so-quota-cant-get-corrupted-20011203

Currently it's possible to give StoreData negative Pos/Length/FileLength
  arguments and thereby set the volume quota usage to arbitrary values.
  This patch makes these values unsigned, since negative file positions
  and lengths don't make sense anyway.

20 years agorx-add-performance-testing-client-20011206
Love Hörnquist-Åstrand [Thu, 6 Dec 2001 23:39:08 +0000]
rx-add-performance-testing-client-20011206

add rxperf with modified license which allows us to distribute it now
pending more useful OpenAFS documentation existing at some point in the
future.

20 years agobosserver-dont-enforce-world-readability-on-server-etcdir-20011126
Sam Hartman [Mon, 26 Nov 2001 21:29:16 +0000]
bosserver-dont-enforce-world-readability-on-server-etcdir-20011126

no reason server etcdir needs to be forced world readable; nothing need
default to those cellconfig files except in the localauth case and then
you need to be able to read the KeyFile anyway

20 years agowindows-afsdb-freelance-notes-20011120
Jeff Riegel [Wed, 21 Nov 2001 18:29:16 +0000]
windows-afsdb-freelance-notes-20011120

notes from Jeff about AFSDB and Freelance clients