Love Hörnquist-Åstrand [Wed, 24 Nov 2004 18:57:58 +0000]
netbsd21-
20041124
netbsd 2.99.x (aka 3.0) support
(aka 2.1?)
Jeffrey Altman [Wed, 24 Nov 2004 18:42:31 +0000]
windows-updates-
20041124
update docs
====================
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.
====================
conditionalize the cleanup of language files on their existence
====================
reformat parts of afsd_init.c
add support for version number checking to afsd_service.exe
====================
Fix the afs_config.exe submount dialog operations: Edit Submount name and
Remove submount entry.
====================
Fix the version info data stored in the resource block to
use the same language identifier as is advertised.
Jeffrey Altman [Tue, 23 Nov 2004 04:19:42 +0000]
freelance-import-
20041122
Fix the afs_freelance.ini import code.
Check both the Client install directory and the %WINDIR% directory
for afs_freelance.ini
Jeffrey Altman [Tue, 23 Nov 2004 03:49:05 +0000]
windows-misc-
20041122
the VC++ 2003 Toolkit is missing some important libraries.
remove it from the README-NT file
====================
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.
====================
update docs
====================
simplify the freelance import from afs_freelance.ini code. don't generate
an new file if the old one does not exist.
begin conversion from old string functions to new strsafe functions.
this will need to be done for all of the afsd_service.exe source
modules before we can regularly use VS .NET 2005
Add support for VL_GetEntryByNameN. Still need to figure out what needs
to be done for VL_GetEntryByNameU. (multi-homed support)
====================
fix a deadlock situation if an Obtain Tokens dialog is produced
by an expiration event and the user chooses to cancel instead of
obtain new credentials.
Ken Hornstein [Fri, 19 Nov 2004 20:01:29 +0000]
aklog-intergration-
20041119
Import the aklog source from the MIT AFS-Kerberos 5 migration kit.
Derrick Brashear [Thu, 11 Nov 2004 07:12:42 +0000]
kdump-in-kernel-
20041111
first step at integrating kdump functionality into the kernel; all the rx stats.
Derrick Brashear [Thu, 11 Nov 2004 04:34:40 +0000]
proc-cellservdb-
20041110
put a cellservdb in /proc/fs/openafs
Nathan Neulinger [Thu, 11 Nov 2004 04:24:13 +0000]
update-objdir-
20041110
make objdir builds work (with fix from shadow@dementia.org in src/rxkad/...)
Derrick Brashear [Thu, 11 Nov 2004 04:20:45 +0000]
mask-loopback-address-allow-loopback-interfaces-to-be-advertised-
20041110
some software network adapters use instances of lo but are real network adapters
. allow them to be advertised. the modified functions are called only in the pro
cess of collecting a list to advertise.
at the same time, make sure we mask 127.0.0.1.
i wonder if that's a mistake.
Jeffrey Altman [Tue, 9 Nov 2004 17:07:28 +0000]
windows-version-bump-
20041109
1.3.7400
Jeffrey Altman [Tue, 9 Nov 2004 12:47:36 +0000]
winnotes-upd-
20041109
windows note updates for 1.3.74
Horst Birthelmer [Mon, 8 Nov 2004 19:59:57 +0000]
aix-vol-salvage-make-malloc-not-be-cast-
20041106
When you have a 64bit big endian machine with a 64bit off_t in the stat
the salvager won't work.
Malloc takes an "unsigned long" and a implicit cast will be almost
always zero :-).
Sam Hartman [Mon, 8 Nov 2004 06:30:37 +0000]
ppc-linux26-
20041108
FIXES 15770
support ppc linux 2.6
Jeffrey Hutzelman [Mon, 8 Nov 2004 06:26:01 +0000]
readdir-report-correct-inums-
20041108
FIXES 15962
We currently try fairly hard to make stat() on a volume root return the
same vnode number as is listed for the mount point by readdir(). This
behaviour is desirable; in fact, getcwd would not work otherwise.
However, we are _not_ careful about making readdir list correct inode
numbers for "." in a volume root or ".." in a directory whose parent is
a volume root. This means that applications which examine these entries
will still see inconsistent inode numbers. Clearly, it would be more
desirable to report consistent inode numbers in all cases, instead of
only in some cases.
The attached patch, written while I was tracking down some NFS
translator problems (which ultimately proved to be unrelated), makes
readdir return consistent inode numbers for volume roots. We are
running this on a few machines and have seen no problems, but it has not
been extensively tested.
Douglas Engert [Mon, 8 Nov 2004 06:20:54 +0000]
vfsck-sol9-breadwrite-args-
20041108
FIXES 15927
fix arg types for bread/bwrite for solaris 9
Jeffrey Hutzelman [Mon, 8 Nov 2004 06:07:16 +0000]
nfs3-no-readdirplus-
20041108
FIXES 15961
don't support readdirplus in nfsv3
Chaskiel M Grundman [Mon, 8 Nov 2004 06:04:12 +0000]
callback-rxcon-use-safely-
20041108
FIXES 15973
add get/put refs to all callback rxconn usages to preclude any possibility of race
Jeffrey Hutzelman [Mon, 8 Nov 2004 05:52:36 +0000]
nfs-root-mtime-
20041108
FIXES 15960
don't reset root mtime, solaris 9 wants it to be the same across multiple calls
Derrick Brashear [Mon, 8 Nov 2004 04:03:10 +0000]
any-way-damnit-
20041107
no comment.
Derrick Brashear [Sun, 7 Nov 2004 17:11:50 +0000]
mask-loopback-address-allow-loopback-interfaces-to-be-advertised-
20041106
some software network adapters use instances of lo but are real network adapters
. allow them to be advertised. the modified functions are called only in the pro
cess of collecting a list to advertise.
at the same time, make sure we mask 127.0.0.1.
i wonder if that's a mistake.
Chaskiel M Grundman [Sat, 6 Nov 2004 07:16:54 +0000]
client-dir-package-interface-update-
20041106
FIXES 15726
cleanup from dread-do-validation; change to using dcache and not just fcache
Kris Van Hees [Sat, 6 Nov 2004 07:02:44 +0000]
vos-timestamp-handling-
20041106
FIXES 15708
fix timestamp handling on create; display "never" for copy time
Chas Williams [Fri, 5 Nov 2004 04:21:29 +0000]
irix-largefile-update-
20041104
from peter tripician via chas. update so largefile support stands a chance of working
Jim Rees [Thu, 4 Nov 2004 15:49:58 +0000]
openbsd36-
20041104
auto-detect OpenBSD 3.6.
Jeffrey Altman [Tue, 2 Nov 2004 06:10:29 +0000]
dnlc-optimize-
20041102
* Optimize calls to the DNLC. Do not search if the FID is the Freelance
root.afs volume. Do not search if we are looking for the magic
_._AFS_IOCTL_._ name.
* Enable Buffer Trace logs in debug builds.
Jeffrey Altman [Mon, 1 Nov 2004 11:45:29 +0000]
readme-nt-
20041101
Remove last vestiges of Win9x comments
Update Wix 2.0 information
Provide download or ordering information on how to obtain all the
necessary tools required to build OpenAFS for FREE
Asanka Herath [Mon, 1 Nov 2004 10:45:48 +0000]
wix-compatibilityK-
20041101
Wix has been updated. to avoid warning/error messages some changes
to the scripts were required. there were no changes to functionality.
Jeffrey Altman [Sat, 30 Oct 2004 00:38:53 +0000]
windows-invalid-dir-handles-
20041029
* Define new error CM_ERROR_TOO_MANY_SYMLINKS
* Fix storage location for Freelance Mount Points broken in
previous patch
* Correct locking throughout the Directory Search code
which was resulting in invalid handle errors being
generated when objects were freed while they were still
in use by the CIFS client
Jeffrey Altman [Thu, 28 Oct 2004 22:20:10 +0000]
ioctl-dbg-windows-
20041028
Output error codes in hex
Derrick Brashear [Thu, 28 Oct 2004 21:46:42 +0000]
inlinebulk-put-ref-on-error-
20041028
if we got an error during InlineBulk, put back the reference
Derrick Brashear [Thu, 28 Oct 2004 19:08:06 +0000]
drop-h-lock-around-putconn-
20041027
the problem with rx_PutConnection is h_FreeConnection uses the host global lock
====================
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.
====================
the problem with rx_PutConnection is h_FreeConnection uses the host global lock
but it shouldn't matter (callback conn should be a client not server conn and
thus not h_FreeConnection on cleanup)
further, we should free the conns we GetConnection'd not the ones still in
the host struct at the end.
Jeffrey Altman [Thu, 28 Oct 2004 17:37:52 +0000]
more-freelance-symlinks-
20041028
* Fix Find Cell By Name pioctl call to return a name "Freelance.Local.Root"
for the fake root.afs volume. This allows the IsPathInAfs() type
functions to succeed for \\AFS\all
* Add missing dialog template for Explorer Shell Extension Symlink->Add
operation
* Increase the acceptable length of the link destination path in the
Add Symlink dialog box support code.
* Fix up some missing prototypes
Jeffrey Altman [Thu, 28 Oct 2004 15:33:05 +0000]
windows-unused-vars-
20041028
remove unused vars
Jeffrey Altman [Thu, 28 Oct 2004 02:22:25 +0000]
windows-docs-
20041027
Documentation updates
Jeffrey Altman [Thu, 28 Oct 2004 02:13:51 +0000]
windows-freelance-symlink-
20041027
* Remove the fallback to KRB4 in afskfw when obtaining tokens
* Add support for symlinks to Freelance root.afs volume
Stored at HKLM\SOFTWARE\OpenAFS\Client\Freelance\Symlinks
<number> = "<linkname>:<relative-path>."
Use symlink.exe to create, list, or remove
Jim Rees [Wed, 27 Oct 2004 20:46:54 +0000]
fbsd53-
20041027
support for FreeBSD 5.3
builds and runs as well as it did on 5.2
Jim Rees [Wed, 27 Oct 2004 16:01:03 +0000]
acinclude-fbsd-
20041027
auto-detect FreeBSD 5.3
Jeffrey Altman [Tue, 26 Oct 2004 02:55:32 +0000]
undo-patch-windows-
20041025
pull patch which breaks submount creation from gui
Jeffrey Altman [Tue, 26 Oct 2004 02:53:40 +0000]
windows-misc-
20041025
* fix loading of SysName from the registry
* reformat cm_scache; change flags to unsigned
====================
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 loading of SysName from the registry
* broken patch to tab_drives.cpp - this will be pulled
Jeffrey Altman [Wed, 20 Oct 2004 05:16:57 +0000]
hklm-reg-fix-
20041019
fix the logic to ensure that HKLM is checked if HKCU does not exist
for EnableKFW and Use524
Derrick Brashear [Tue, 19 Oct 2004 23:34:33 +0000]
provide-fPIC-method-for-amd64-
20040508
FIXES 3680
allow building fPIC on amd64
Derrick Brashear [Tue, 19 Oct 2004 07:07:53 +0000]
irix-cleanup-
20041019
clean up my messes.
Jeffrey Altman [Tue, 19 Oct 2004 01:51:48 +0000]
windows-cleanup-
20041018
cleanup AFS_component_version_number.h
Jeffrey Altman [Mon, 18 Oct 2004 23:31:56 +0000]
lang-clean-fix-
20041018
Fix the for loop used to clean the language directories
Jeffrey Altman [Mon, 18 Oct 2004 21:25:58 +0000]
windows-version-update-
20041018
1.3.7300
Derrick Brashear [Mon, 18 Oct 2004 19:39:16 +0000]
configure-linux-header-path-help-
20041018
thanks to Kris Van Hees for pointing this out
John Bucy [Mon, 18 Oct 2004 07:49:45 +0000]
setpag-print-throttled-
20041012
kernel printf when we start throttling for time
Derrick Brashear [Mon, 18 Oct 2004 07:12:49 +0000]
partly-revert-volser-restore-timestamp-cleanup-
20041018
messes up creation field on new volume.
Jeffrey Hutzelman [Mon, 18 Oct 2004 07:08:14 +0000]
amd64-ccobj-update-
20041018
update ccobj for amd64
Jeffrey Altman [Mon, 18 Oct 2004 05:22:18 +0000]
windows-version-
20041017
update version to 1.3.7200
Jeffrey Altman [Mon, 18 Oct 2004 03:44:14 +0000]
freelance-end-of-line-
20041017
modify the algorithm used for loading mount point entries to ignore
all data beyond the terminating dot.
Jeffrey Altman [Sun, 17 Oct 2004 22:19:13 +0000]
flushonhibernate-
20041017
Fix Asanka's patch.
Asanka Herath [Sun, 17 Oct 2004 22:02:01 +0000]
flushonhibernate-registry-
20041017
Add a new FlushOnHibernate registry option to allow users to disable
the Power Management Flush support
Jeffrey Altman [Sun, 17 Oct 2004 22:00:24 +0000]
winnotes-updates-
20041017
Updates for 1.3.72
New FlushOnHibernate registry option
Jeffrey Altman [Sun, 17 Oct 2004 21:24:19 +0000]
krb5-no-dot-
20041017
Generate error in the client if a krb5 principal contains a dot in the
first component.
Jeffrey Altman [Sun, 17 Oct 2004 16:18:24 +0000]
local-system-acct-
20041017
When testing for Administrator privileges always grant them to the
Local System account
Jeffrey Altman [Sun, 17 Oct 2004 03:46:56 +0000]
wix-clean-
20041016
Clean the intermediary build files before each build.
Jeffrey Altman [Sun, 17 Oct 2004 03:44:52 +0000]
windows-updates-
20041016
* Correct the handling of Over Quota errors
* Update all of the reference count fields to use unsigned long
instead of shorts or ints
* Reformat touched files
====================
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.
====================
* Update reference count fields from short to unsigned long
Jeffrey Altman [Sat, 16 Oct 2004 23:41:18 +0000]
des-fix-mustalign-
20041016
FIXES 15637
Forgot to fix the MUSTALIGN case
Jeffrey Altman [Sat, 16 Oct 2004 23:31:49 +0000]
des-fix-
20041016
FIXES 15637
Correct previous patch.
Derrick Brashear [Fri, 15 Oct 2004 07:01:35 +0000]
rx-peer-and-conn-refcounts-now-unsigned-int32-
20041015
FIXES 15459
in an extreme case, the peer object refcount could wrap
we should probably also log if wrapping is imminent
Jeffrey Altman [Thu, 14 Oct 2004 02:52:05 +0000]
pioctl-debug-
20041013
Add more debug statements
Peter Nelson [Wed, 13 Oct 2004 22:00:04 +0000]
makefileproto-linux26-
20041013
add "linux26" to the case where we symlink asm
Chaskiel M Grundman [Wed, 13 Oct 2004 17:29:36 +0000]
h-tossstuff-r-leak-
20041013
don't leak host hash objects in h_TossStuff_r
Peter Nelson [Wed, 13 Oct 2004 15:46:29 +0000]
amd64-use-flock-
20041013
so we stand a chance of compiling on amd64
Jeffrey Altman [Wed, 13 Oct 2004 15:23:25 +0000]
uninitialized-vars-
20041013
FIXES 15604
initialize flags in SetFields so that it has a default value
otherwise a runtime exception is generated on windows due to
lack of initialization before use.
Joe Buehler [Wed, 13 Oct 2004 15:07:53 +0000]
linux-pagelaunder-avoid-deadlock-
20041012
FIXES 15239
mmap'd files can deadlock on linux; fix by not letting kernel free pages out of our space predictively
====================
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.
====================
FIXES 15239
mmap'd files can deadlock on linux; fix by not letting kernel free pages out of
our space predictively
Jeffrey Altman [Wed, 13 Oct 2004 07:41:38 +0000]
expand-strings-
20041012
Stop blowing away the drive mappings of REG_EXPAND_SZ values due to
the use of the incorrect order of the parameters
Jeffrey Altman [Wed, 13 Oct 2004 03:48:24 +0000]
winnotes-
20041012
Update documentation files with latest changes
Jeffrey Altman [Wed, 13 Oct 2004 03:30:49 +0000]
rx-refcount-
20041012
FIXES 15585
* use new rx_GetConnection/rx_PutConnection functions to increment/
decrement the refCounts on the rx_connection objects so that the
objects are not invalidated during RXAFS_xxx calls when:
- the user tokens change
- the global cryptlevel changes
* import SysName from registry and parse along white space boundaries
Horst Birthelmer [Wed, 13 Oct 2004 02:28:25 +0000]
aix-des-cleanup-
20041012
incrementing pointers behind casts apparently pisses of vac 6
Jeffrey Hutzelman [Wed, 13 Oct 2004 02:16:40 +0000]
amd64-linux26-
20041012
FIXES 15617
support for amd64 linux26
Jim Rees [Wed, 13 Oct 2004 02:10:58 +0000]
build-fcrypt-from-domestic-
20041010
don't install rxkad/domestic/*.c, just compile them where they live
Eric Melski [Wed, 13 Oct 2004 02:04:55 +0000]
solaris-fix-ulimit-write-size-bug-
20041012
FIXES 15609
convert nonsensical comparison to one which makes sense in light of code and
comments below it
Chaskiel M Grundman [Wed, 13 Oct 2004 01:51:00 +0000]
dread-do-validation-
20041012
FIXES 15340
so we can do directory object validation in DRead()
Chaskiel M Grundman [Wed, 13 Oct 2004 01:36:59 +0000]
linux-26-vlru-cycle-
20041012
FIXES 15339
it was only a matter of time...
Chaskiel M Grundman [Wed, 13 Oct 2004 01:30:03 +0000]
asmlinkage-for-pioctl-
20041012
FIXES 15338
another place where we need to declare a function asmlinkage
Jeffrey Hutzelman [Tue, 12 Oct 2004 21:51:55 +0000]
cleanup-rx-peer-leak-
20041012
if the peer was the second in the chain we could throw away our reference to
the first, effectively leaking it forever.
Chaskiel M Grundman [Mon, 11 Oct 2004 23:31:26 +0000]
linux-setgroups-need-asmlinkage-
20041011
FIXES 15460
sys_setgroups pointers must be asmlinkage everywhere
Derrick Brashear [Mon, 11 Oct 2004 19:39:55 +0000]
viced-callback-lock-host-during-multi-breakcallback-
20041011
since it looks like CheckHost is using a gc'd client conn, this more or less has to be it.
====================
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.
====================
since it looks like CheckHost is using a gc'd client conn, this more or less has
to be it.
and don't deadlock.
====================
since it looks like CheckHost is using a gc'd client conn, this more or less has
to be it.
and don't deadlock.
====================
since it looks like CheckHost is using a gc'd client conn, this more or less has
to be it.
and don't deadlock.
Jeffrey Hutzelman [Mon, 11 Oct 2004 19:35:30 +0000]
rx-getputconnection-
20041011
FIXES 15584
add Get/PutConnection funcs to diddle conn refCounts.
Sam Hartman [Mon, 11 Oct 2004 19:19:35 +0000]
osi-timeval-can-be-timeval-
20041011
FIXES 15580
since the timeval is used only internally, let's just use struct timeval
Sam Hartman [Mon, 11 Oct 2004 17:21:28 +0000]
ppc64-26-support-
20041011
FIXES 15581
Derrick Brashear [Mon, 11 Oct 2004 17:01:58 +0000]
libafsrpc-cleanup-
20041011
FIXES 15586
remove unwanted files
Jeffrey Altman [Mon, 11 Oct 2004 00:50:37 +0000]
windows-combined-
20041010
* fix the Windows cache manager to prevent it from replacing the
rx_connection object associated with the cm_conn_t object on each
and every operation if "fs crypt" was set. This explains the
dramatic performance difference when crypt is used vs clear.
The problem: 'cryptall', a boolean flag indicating whether or not
"fs crypt" is set, was being compared to the rx_connection
cryptlevel which is either rxkad_clear:0 or rxkad_crypt:2.
1 != 2 and therefore the rx_connection was always destroyed
and replaced on each and every operation.
Lock the cm_conn_t object around every call to RXAFS_xxxx functions.
It is not safe for the cm_conn_t object to not be locked because
rx_DestroyConnection might be called from another thread if:
- the user's tokens have changed or expired
- the crypt mode has changed
* Modify cm_Lookup to evaluate a list of possible values for @sys
instead of just a single entry. Re-write cm_ExpandSysname and
add cm_LookupInternal.
* Reformat affected files
* Add cache file name to cache creation errors
Jim Rees [Sun, 10 Oct 2004 03:13:41 +0000]
crypt-take-voids2-
20041009
make various crypt/decrypt routines take void * args to eliminate more
compiler warnings
Derrick Brashear [Sun, 10 Oct 2004 03:03:11 +0000]
pr-initialize-dont-leak-ref-
20041009
we were leaking an afsconf_Open ref. let's not.
Derrick Brashear [Sun, 10 Oct 2004 02:53:32 +0000]
generic-client-init-function-
20041009
FIXES 15464
so we can have this code in just one place
Chaskiel M Grundman [Sat, 9 Oct 2004 23:34:45 +0000]
linux-disable-pmtu-discovery-
20041009
FIXES 15505
don't set DF bit; use setsockopt to turn of kernel "pmtu feature"
Jim Rees [Fri, 8 Oct 2004 20:30:37 +0000]
add-rocks-
20041008
add rocks to cmd args to match prototype
Kevin Coffman [Wed, 6 Oct 2004 18:35:19 +0000]
revert-fix-seclevel-
20041006
Ticket #15367
revert rev 1.12 fix-seclevel-
20030213:
Use requested security level instead of "clear" in pr_Initialize
Jim Rees [Tue, 5 Oct 2004 21:40:59 +0000]
crypt-take-voids-
20041005
make various crypt/decrypt routines take void * args to eliminate more
compiler warnings
Jeffrey Altman [Mon, 4 Oct 2004 08:35:13 +0000]
admin-group-nsis-
20041004
correct the test for -create and -remove so that the proper
action is taken for each
Jeffrey Altman [Sun, 3 Oct 2004 15:06:46 +0000]
pioctl-debug-
20041003
Add a new registry value HKLM/SOFTWARE/OpenAFS/Client "IoctlDebug" DWORD
which when set to a non-zero value will output error messages to stderr.
This is to assist in debugging ioctl failures when fs.exe, tokens.exe,
etc. fail because of an inability to open the ioctl file name.
Jeffrey Altman [Sun, 3 Oct 2004 14:53:35 +0000]
windows-symlink-access-
20041003
FIXES 15250
One more fix for the symlink problem. Access control checks cannot be
performed until after we have walked all of the symlinks. Otherwise,
we are checking access to the symlink itself and not to the file.
Jeffrey Altman [Sun, 3 Oct 2004 14:34:27 +0000]
win-power-mgmt-flush-test-
20041003
The windows power management code responds to a request to suspend or
hibernate by performing a "fs flushvol" as the logged in user on each
of the SMB/CIFS mounted shares. This can be very time consuming if
the cell servers cannot be reached.
This patch adds a test to ensure that there is at least one network
adapter in the machine which is not a loopback adapter.
While developing this patch other areas of concern have been raised.
The power management code waits a fixed period of time based upon
the hard dead timeout before allowing the suspend/hibernate to continue.
This allows the machine to shutdown even if there are active flush
operations being performed. This defeats the benefit of performing
the flush at all.
A better mechanism could be developed if the functions called via
cm_IoctlFlushVolume returned and checked error codes. Then it might
be possible to abandon the flush operation if a Server Not Reachable
state was obtained.
The power management flush operations will also not work on Terminal
Server. This would be important in the case where a terminal server
is shutting down due to a switch over to a UPS. The reason it does
not work on Terminal Server is that there is that it is not possible
for afsd_service.exe to enumerate the SMB/CIFS shares and impersonate
the individual logged in users.
It would be preferred for there to be a new cm_FlushAll() function
implemented which was not dependent on the use of the ioctl mechanism
for the purpose of identifying a volume ID or a user ID.
Derrick Brashear [Sat, 2 Oct 2004 16:54:28 +0000]
vol-fix-nuke-iteration-and-nuke-globals-
20040928
allInodes global dies; don't use ti as an iterator right after freeing it.
Jim Rees [Fri, 24 Sep 2004 16:42:32 +0000]
rx-nprocs-static-
20040924
make nProcs static so it will count up correctly
also change its scope to eliminate compiler warning
Nickolai Zeldovich [Fri, 24 Sep 2004 02:42:28 +0000]
rxi-flushwrite-fake-a-better-empty-packet-
20040923
Fake a better empty packet in rxi_FlushWrite(), with enough space for
an rxkad header. Otherwise rxi_PrepareSendPacket may panic.
Derrick Brashear [Thu, 23 Sep 2004 04:55:52 +0000]
threaded-volserver-and-protoization-update-
20040922
fix a deadlock between GCTrans and DeleteTrans
more protoization