openafs.git
20 years agocachemgr-dont-leak-stat-info-to-clients-not-entitled-to-it-20010605
Jeff Riegel [Tue, 5 Jun 2001 15:16:10 +0000]
cachemgr-dont-leak-stat-info-to-clients-not-entitled-to-it-20010605

if you only have lookup don't leak stat information from cache, which normally requires read

20 years agousermode-cachemgr-open-mode-checking-fix-20010605
Jeff Riegel [Tue, 5 Jun 2001 15:11:01 +0000]
usermode-cachemgr-open-mode-checking-fix-20010605

default to read mode when opening

20 years agolinux24-use-flock64-instead-of-flock-20010604
Pavel Semerad [Mon, 4 Jun 2001 23:29:00 +0000]
linux24-use-flock64-instead-of-flock-20010604

"locking on linux 2.4.x does not work. 2.4.x uses internally 64bit offsets
(start and end of lock) and end is set to 64 bit int max. When it converts
to struct flock, len is set to 0xffffffff (cut off high 32 bits) and afs_lockctl
decides it is byte-range locking. So on 2.4.x kernel there should be used flock64
instead of flock (flock64 is typedef to flock on 64bit systems such as sparc64
and ia64)"

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

use flock64 for linux24

20 years agoviced-dotted-quad-logging-without-malloc-20010601
Nathan Neulinger [Mon, 4 Jun 2001 22:08:19 +0000]
viced-dotted-quad-logging-without-malloc-20010601

based on suggestion from rees@umich.edu

====================
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 for solaris cc joy

20 years agolibadmin-cleanup-and-afsconfigization-20010604
Nathan Neulinger [Mon, 4 Jun 2001 22:07:27 +0000]
libadmin-cleanup-and-afsconfigization-20010604

clean up warnings, get rid of unused variables, and set up for use of
afsconfig.h

20 years agolwp-iomgr-warning-suppression-20010602
Nathan Neulinger [Sun, 3 Jun 2001 18:20:15 +0000]
lwp-iomgr-warning-suppression-20010602

remove unused variable, add missing parens

20 years agofix-afsmonitor-patch-for-string-h-existance-check-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:52:05 +0000]
fix-afsmonitor-patch-for-string-h-existance-check-20010602

as not all platforms may have string.h

20 years agovol-volser-more-liberal-information-in-log-messages-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:34:34 +0000]
vol-volser-more-liberal-information-in-log-messages-20010602

hopefully no one is processing these logs in a way this breaks. make error
and volname information available as part of error message

20 years agosgiefs-support-is-gone-but-great-rock-lives-on-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:32:18 +0000]
sgiefs-support-is-gone-but-great-rock-lives-on-20010602

remove inclusion of missing sgiefs directory since we didn't get the binary
library needed for it to work anyhow

20 years agorxtrace-warning-cleanup-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:31:07 +0000]
rxtrace-warning-cleanup-20010602

more compiler warning squishing

20 years agosplit-deprecated-into-insecure-and-obsolete-in-top-makefile-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:29:52 +0000]
split-deprecated-into-insecure-and-obsolete-in-top-makefile-20010602

so it's obvious to people which mistakes are silly and which are dangerous

20 years agolibadmin-warning-cleanup-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:28:33 +0000]
libadmin-warning-cleanup-20010602

add missing parens, remove unused decls

20 years agokauth-const-warning-cleanup-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:27:03 +0000]
kauth-const-warning-cleanup-20010602

dir is not modified so this is safe

20 years agocomerr-get-rid-of-compile-warnings-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:26:15 +0000]
comerr-get-rid-of-compile-warnings-20010602

placate compiler

20 years agoaudit-make-int-functions-return-zero-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:23:18 +0000]
audit-make-int-functions-return-zero-20010602

rather than just having no code block, return 0

20 years agoafsmonitor-formatting-strings-update-20010602
Nathan Neulinger [Sun, 3 Jun 2001 17:22:10 +0000]
afsmonitor-formatting-strings-update-20010602

so error messages are properly formatted strings

20 years agonamei-ops-cleanup-20010601
Nathan Neulinger [Sun, 3 Jun 2001 17:19:15 +0000]
namei-ops-cleanup-20010601

get rid of unused/warnings from vol/namei_ops.c

20 years agolog-host-addrs-as-dotted-quad-not-hex-in-viced-20010601
Nathan Neulinger [Fri, 1 Jun 2001 20:14:58 +0000]
log-host-addrs-as-dotted-quad-not-hex-in-viced-20010601

so you get a dotted quad and byte-correct port instead of a hex ip and a potentially
byteswapped port number from viced log messages

20 years agoinitial-configure-work-take-two-20010601
Derrick Brashear [Fri, 1 Jun 2001 20:09:22 +0000]
initial-configure-work-take-two-20010601

this isn't useful yet but is a step in the right direction.

20 years agotop-level-makefile-changes-to-allow-partial-afs-compile-20010530
Chaskiel M Grundman [Wed, 30 May 2001 20:45:01 +0000]
top-level-makefile-changes-to-allow-partial-afs-compile-20010530

"The afs toplevel makefile theoretically allows you to build bits and
pieces of AFS without too much other stuff you don't need. Building the
kernel code this way is hard, since it imports stuff from so many modules,
and the dependencies aren't necessarily done right. This patch fixes many
of the special "for kernel use only" targets to be "for kernel use
only" and adds a few new ones, so that the only userlevel stuff built by
make TARGET=libafs is util, pinstall, rxgen, comerr (needed for
afs/Makefile), and cmd"

20 years agomake-xdr-int64-support-consistent-now-that-its-everywhere-20010530
Chaskiel M Grundman [Wed, 30 May 2001 20:39:27 +0000]
make-xdr-int64-support-consistent-now-that-its-everywhere-20010530

"an mr-afs related patch introduced xdr_afs_int64 into the code. This is a
good thing I suppose. The same patch also re-introduced the use of
xdr_long on some platforms, which isn't needed, and isn't really
appropriate. Said patch also only included the #defines for xdr_afs_int64
into rpc_main.c and not xdr.h. (it is my opinion that
rpc_main.c:write_int32_macros should only be used if kflag is set, but
transarc didn't do it that way, so I'm not going to change it)"

20 years agomake-rxgen-split-rpcs-not-declare-start-parameters-for-functions-for-data-which-is...
Chaskiel M Grundman [Wed, 30 May 2001 20:34:17 +0000]
make-rxgen-split-rpcs-not-declare-start-parameters-for-functions-for-data-which-is-copied-out-only-20010530

"fix for rxgen so that split rpc's don't erroneously declare parameters in
the "start" function that should only be declared in the "end":

example:

for the following declaration:
Store(IN struct _index_t *descr,
          IN string group<STAGE_STRINGSIZE>,
          OUT string spool<STAGE_STRINGSIZE>, OUT afs_uint32 *slot) split;

before patch:
int StartDMV_Store(z_call, descr, group)
        register struct rx_call *z_call;
        struct _index_t * descr;
        char * group, **spool;
{
[...]
after patch:
int StartDMV_Store(z_call, descr, group)
        register struct rx_call *z_call;
        struct _index_t * descr;
        char * group;
{
[...]
"

20 years agocatch-bad-afsconf-cehck-result-to-fix-fileserver-segv-20010530
Garry Zacheiss [Wed, 30 May 2001 20:31:21 +0000]
catch-bad-afsconf-cehck-result-to-fix-fileserver-segv-20010530

"afsconf_GetKey() in auth/cellconfig.c doesn't check the return
value of afsconf_Check(), which can lead to the fileserver SEGV'ing if
afsconf_Check() returns an error."

20 years agomake-backup-dumpi-print-volumes-over-2caret31-bytes-correctly-20010530
Mattias Amnefelt [Wed, 30 May 2001 20:26:09 +0000]
make-backup-dumpi-print-volumes-over-2caret31-bytes-correctly-20010530

makes backup dumpi print volumes > 2^31 bytes correctly

20 years agoupdate-pdf-documentation-to-include-fonts-20010530
Mattias Amnefelt [Wed, 30 May 2001 20:22:56 +0000]
update-pdf-documentation-to-include-fonts-20010530

"I noticed that the pdf documentation bundled with openafs doesn't include
the correct fonts to display properly (atleast with my acrobat). I've
generated afs-pdf:s with type1 fonts"

20 years agofix-msf-cat-warnings-in-compile-et-20010524
Nathan Neulinger [Thu, 24 May 2001 22:11:45 +0000]
fix-msf-cat-warnings-in-compile-et-20010524

so we don't get warning for msf stuff

20 years agosolaris-make-df-in-cwd-return-useful-output-20010524
Nickolai Zeldovich [Thu, 24 May 2001 22:09:50 +0000]
solaris-make-df-in-cwd-return-useful-output-20010524

"This patch makes "df ." work in AFS directories under Solaris 8.
(df looks in /etc/mnttab, which is obtained from the kernel, for
the device number and then does a pwd-style walk up the tree..
Currently the device number gets set to 0x100000 by default which
doesn't match the 0x1234 device number of AFS vnodes.)"

20 years agolinux-s390-kas-setpass-fix-20010515
Carsten Jacobi [Mon, 21 May 2001 21:24:04 +0000]
linux-s390-kas-setpass-fix-20010515

fix for s390, per Carsten:
(this behaviour occurs only
on 31-bit
S/390-Systems, the "s390x"-machines should change the AFS-password also
without this patch as far as I can imagine)

20 years agoafs-overview-documentation-20010519
Laura Stentz [Sat, 19 May 2001 18:45:43 +0000]
afs-overview-documentation-20010519

Contributed overview docs from IBM, this time binary mode.

20 years agoremove-bad-non-binary-commit-of-afs-docs-20010519
Derrick Brashear [Sat, 19 May 2001 18:44:43 +0000]
remove-bad-non-binary-commit-of-afs-docs-20010519

so i can readd with -kb

20 years agoadd-afs-overview-manuals-20010519
Laura Stentz [Sat, 19 May 2001 18:43:10 +0000]
add-afs-overview-manuals-20010519

Documentation contributed by IBM

20 years agohpux11-kernel-support-20010517
Derrick Brashear [Fri, 18 May 2001 18:15:20 +0000]
hpux11-kernel-support-20010517

put together from 1994(!) rx open source release available
at ftp://ftp.dementia.org/pub/rx, other architectures, rx_kcommon.c,
and some net research on hpux

has no chance of working, slim chance of compiling, and seriously needs
spinlock work to be useful

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

shell of this file. won't work

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

shell of osi_vfs.h

not useful yet

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

use macros in osi_machdep.h for doing spinlocks

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

include streams tli header

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

do a little cleanup after looking at other architectures

still won't work until we know more about sleep/wakeup and kernel
semaphore interface crap

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

try again. probably still not useful

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

get this stuff compiling

20 years agoupdate-makefiles-for-afsconfig-h-20010517
Nathan Neulinger [Fri, 18 May 2001 02:22:10 +0000]
update-makefiles-for-afsconfig-h-20010517

more updates so afsconfig.h can be included

20 years agolinux24-param-cleanup-20010516
Derrick Brashear [Wed, 16 May 2001 05:16:27 +0000]
linux24-param-cleanup-20010516

make usr version of file not define 64 bit env
make kernel version include modversions before afs_sysnames.h (which includes
stds.h)

20 years agoafsconfig-updates-20010515
Nathan Neulinger [Wed, 16 May 2001 00:28:09 +0000]
afsconfig-updates-20010515

updates to and to use afsconfig.h

20 years agosgi-fix-prototypes-20010515
Nathan Neulinger [Wed, 16 May 2001 00:16:58 +0000]
sgi-fix-prototypes-20010515

use strings.h on irix to get needed prototypes

20 years agofix-bad-logic-in-afsconfig-h-in-20010515
Nathan Neulinger [Wed, 16 May 2001 00:15:54 +0000]
fix-bad-logic-in-afsconfig-h-in-20010515

fix some iverted logic

20 years agoafs-superuser-foreign-realm-checks-20010514
Nathan Neulinger [Mon, 14 May 2001 23:56:33 +0000]
afs-superuser-foreign-realm-checks-20010514

This rewrite cleans up the code a bit, removes any athena specific
references (not needed anymore in this version), and adds support for
multi realm management of afs servers (you can now specify
"admin@OTHERREALM" in your userlist).

Code now checks as follows:

tname
tinst  - remote user info from conn
tcell
lcell - local cell
lrealm - local realm (defaults to lcell if not avail)

if no remote cell or instance
        allow localauth
if the cell of the remote connection matches local cell or local realm
        if not tinst
                allow if tname in UserList
        if tinst
                allow if tname.tinst in UserList
if cell doesn't match local cell or realm
        if not tinst
                allow if tname@cell in UserList
                allow if tname@CELL in UserList
        if tinst
                allow if tname.tinst@cell in UserList
                allow if tname.tinst@CELL in UserList

modified per openafs-devel discussion such that krb5 versions (/tinst
rather than .tinst) code path disabled for now

20 years agorestore-writelogbuffer-for-mrafs-20010514
Hartmut Reuter [Mon, 14 May 2001 23:54:01 +0000]
restore-writelogbuffer-for-mrafs-20010514

MR-AFs is making use of this to hook into logging to write multiline messages
into the log

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

move forward to current version of function

20 years agoaix42-pthread-self-return-needs-cast-20010514
Mark Dapoz [Mon, 14 May 2001 23:26:34 +0000]
aix42-pthread-self-return-needs-cast-20010514

To compile on AIX 4.2 on PowerPC

20 years agoinitial-afsconfig-h-file-20010514
Nathan Neulinger [Mon, 14 May 2001 23:23:12 +0000]
initial-afsconfig-h-file-20010514

start of config file work for autoconf

====================
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 Makefile use top-level location rather than config as Nathan intended

20 years agochange-effect-of-afs-quietfs-env-20010514
Nathan Neulinger [Mon, 14 May 2001 22:41:12 +0000]
change-effect-of-afs-quietfs-env-20010514

so ViceLog happens when file server starts

20 years agodarwin-clear-va-flags-in-copyoutattrs-20010514
Ted McCabe [Mon, 14 May 2001 22:30:35 +0000]
darwin-clear-va-flags-in-copyoutattrs-20010514

"I noticed that when I tried copying files out of AFS, I would often
not be able to modify the destination files/dirs due to the user flags
being non-zero (see man chflags).  Turns out that not all of the
fields of Darwin's struct vattr were being set in afs_CopyOutAttrs

This patch clears the va_flags entry. I wasn't sure how the
va_filerev and va_gen fields are precisely used, but since I wasn't
noticing any trouble I could attribute to them, I left them
unspecified.  I suspect someone with more knowledge can say for sure.
"

20 years agomake-vos-accept-force-switch-20010514
Nathan Neulinger [Mon, 14 May 2001 22:27:10 +0000]
make-vos-accept-force-switch-20010514

and let libcmd handling deal with -f

20 years agodes-warnings-cleanup-20010514
Nathan Neulinger [Mon, 14 May 2001 22:19:47 +0000]
des-warnings-cleanup-20010514

clean up some more warnings from des

20 years agolinux24-s390-initial-support-20010503
Carsten Jacobi [Thu, 3 May 2001 15:19:00 +0000]
linux24-s390-initial-support-20010503

base support for s390 linux 2.4

20 years agocompile-snprintf-for-solaris25-20010430
Mark Dapoz [Mon, 30 Apr 2001 23:15:50 +0000]
compile-snprintf-for-solaris25-20010430

so this will still build for solaris 2.5.1, build our snprintf

20 years agomake-des-test-buildable-20010430
Chas Williams [Mon, 30 Apr 2001 23:13:05 +0000]
make-des-test-buildable-20010430

remove bogus rcsid string lines with no strings

20 years agoafsdb-callout-and-userspace-implementation-20010430
Nickolai Zeldovich [Mon, 30 Apr 2001 23:08:39 +0000]
afsdb-callout-and-userspace-implementation-20010430

"This patch implements AFSDB support for both user-space programs
and for the kernel.. I've tested these on sun4x_57 (64-bit) and
on i386_linux22."

not currently enabled in any configuration

20 years agoremove-incomplete-dns-patch-20010430
Derrick Brashear [Mon, 30 Apr 2001 23:03:33 +0000]
remove-incomplete-dns-patch-20010430

this shouldn't have leaked in yet

20 years agofix-comment-for-linux-conf-20010430
Nathan Neulinger [Mon, 30 Apr 2001 16:18:39 +0000]
fix-comment-for-linux-conf-20010430

fix comment to reflect variable change

20 years agoremove-text-after-endif-20010430
Nathan Neulinger [Mon, 30 Apr 2001 16:09:41 +0000]
remove-text-after-endif-20010430

remove typo which crept in

20 years agowin95-initial-port-20010430
Jeff Riegel [Mon, 30 Apr 2001 08:03:55 +0000]
win95-initial-port-20010430

Windows 95/98 port from IBM Almaden

documentation and build support

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

Windows 95/98 port from IBM Almaden

changes to cache manager

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

Windows 95/98 port from IBM Almaden

makefile changes

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

Windows 95/98 port from IBM Almaden

osi changes to support 95/98

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

Windows 95/98 port from IBM Almaden

installer for 95/98 client

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

Windows 95/98 port from IBM Almaden

win95/98 panel app

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

Windows 95/98 port from IBM Almaden

win95/98 panel app help

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

Windows 95/98 port from IBM Almaden

makefile changes plus dns implementation

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

Windows 95/98 port from IBM Almaden

makefile changes plus changes for compat with dos line conventions

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

Windows 95/98 port from IBM Almaden

config for 95/98

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

Windows 95/98 port from IBM Almaden

makefile changes plus support for 95/98 env

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

Windows 95/98 port from IBM Almaden

ms compiler changes

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

Windows 95/98 port from IBM Almaden

makefile changes

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

Windows 95/98 port from IBM Almaden

lwp changes
to support 95/98 port

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

Windows 95/98 port from IBM Almaden

rx changes to support 95/98 port

20 years agolinux-ppc-linux22-base-config-20010429
Sam Hartman [Mon, 30 Apr 2001 00:42:33 +0000]
linux-ppc-linux22-base-config-20010429

base ppc linux 2.2 config

20 years agowinnt-redefine-snprintf-instead-of-building-one-20010429
Derrick Brashear [Mon, 30 Apr 2001 00:40:12 +0000]
winnt-redefine-snprintf-instead-of-building-one-20010429

WINNT has _snprintf, just use that

20 years agolinux-conf-use-options-not-afsd-options-20010429
Nathan Neulinger [Mon, 30 Apr 2001 00:36:36 +0000]
linux-conf-use-options-not-afsd-options-20010429

make distributed config consistent with rc file

20 years agofix-sense-of-ifdef-for-darwin-20010425
Jeffrey Hutzelman [Wed, 25 Apr 2001 06:44:34 +0000]
fix-sense-of-ifdef-for-darwin-20010425

reverse change from freebsd patch

20 years agolinux-make-bridge-script-more-respectful-of-build-environment-20010423
Derrick Brashear [Tue, 24 Apr 2001 03:48:32 +0000]
linux-make-bridge-script-more-respectful-of-build-environment-20010423

make it actually use the known path to the kernel source, and check for existance before doing anything

20 years agolinux-bridge-to-deal-with-redhat71-in-pre-autoconf-days-20010422
Derrick Brashear [Sun, 22 Apr 2001 20:37:33 +0000]
linux-bridge-to-deal-with-redhat71-in-pre-autoconf-days-20010422

Kinda crappy but it will suffice for now

20 years agolinux-kernel-extra-definitions-for-sp-versus-mp-20010422
Jeremy Katz [Sun, 22 Apr 2001 20:02:58 +0000]
linux-kernel-extra-definitions-for-sp-versus-mp-20010422

__SMP__, and __BOOT_KERNEL_SMP __BOOT_KERNEL_UP definitions as appropriate

20 years agolinux-revert-truncate-inode-pages-for-smush-20010420
Derrick Brashear [Sat, 21 Apr 2001 20:04:08 +0000]
linux-revert-truncate-inode-pages-for-smush-20010420

osi_VM_TryToSmush needs to be careful not to walk on in-use pages

20 years agocleanup-from-readdir-type-hint-guessing-20010420
Derrick Brashear [Sat, 21 Apr 2001 19:38:39 +0000]
cleanup-from-readdir-type-hint-guessing-20010420

Based on report by Chas Williams; FindVCache get getting a ref which we
need to release

20 years agoconditionals-for-linux-ac-kernels-20010420
Derrick Brashear [Sat, 21 Apr 2001 05:23:25 +0000]
conditionals-for-linux-ac-kernels-20010420

Adapted from katzj@linuxpower.org's work. Still needs makefile hackery and
glue scripts to do the defining

20 years agobe-tolerant-of-unbuilding-kdumps-20010420
Derrick Brashear [Sat, 21 Apr 2001 04:14:56 +0000]
be-tolerant-of-unbuilding-kdumps-20010420

so people who can't get kdump to build aren't so confused

20 years agotruncate-inode-pages-for-linux22-20010420
Derrick Brashear [Sat, 21 Apr 2001 02:35:29 +0000]
truncate-inode-pages-for-linux22-20010420

backport support for 2.2 kernerls

20 years agolinux-call-truncate-inode-pages-in-preference-to-invalidate-inode-pages-20010420
Chas Williams [Sat, 21 Apr 2001 02:23:04 +0000]
linux-call-truncate-inode-pages-in-preference-to-invalidate-inode-pages-20010420

To cure "Failed to invalidate all pages on inode 0xc9208720"

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

20 years agolinux-vfsops-updates-20010420
Jeremy Katz [Sat, 21 Apr 2001 02:06:52 +0000]
linux-vfsops-updates-20010420

set s_maxbytes if MAX_NON_LFS is defined

use &tvp->v nsttead of tvp

20 years agoafs-gunlock-when-umount-as-not-suser-20010420
Jeremy Katz [Sat, 21 Apr 2001 01:31:39 +0000]
afs-gunlock-when-umount-as-not-suser-20010420

bad to AFS_GLOCK then exit without AFS_GUNLOCK

20 years agolinux24-add-missing-truncate-inode-pages-call-20010420
Chas Williams [Sat, 21 Apr 2001 01:14:16 +0000]
linux24-add-missing-truncate-inode-pages-call-20010420

actually do the inode page truncation for linux 2.4

20 years agoundo-i386-linux-64bit-env-20010420
Derrick Brashear [Sat, 21 Apr 2001 00:58:24 +0000]
undo-i386-linux-64bit-env-20010420

It turns out defining AFS_64BIT_ENV ends up causing accesses to hang
if a server goes down; Should not be the case but other work will be needed
to figure out why or even if we need to be AFS_64BIT_ENV; in the meantime,
revert this part of patches done for xdr_int64; it's not ncessary now anyhow

20 years agovfs-dqrwlock-doesnt-exist-before-solaris7-20010420
Nickolai Zeldovich [Sat, 21 Apr 2001 00:34:59 +0000]
vfs-dqrwlock-doesnt-exist-before-solaris7-20010420

in order that the modified solaris code works on pre-solaris7 it's necessary
to not use the locks we didn't have then

20 years agolinux24-ppc-has-no-setgroups32-20010420
Chas Williams [Fri, 20 Apr 2001 23:42:11 +0000]
linux24-ppc-has-no-setgroups32-20010420

use generic ifdef to deal with ppc 2.4 not having setgroups32 yet

20 years agoinstall-updates-for-freebsd-20010420
Tom Maher [Fri, 20 Apr 2001 23:38:51 +0000]
install-updates-for-freebsd-20010420

make install build for freebsd

20 years agolinux-dentry-revalidation-always-revalidate-20010420
Derrick Brashear [Fri, 20 Apr 2001 23:34:03 +0000]
linux-dentry-revalidation-always-revalidate-20010420

Based on collaboration with Chas Williams and Chaskiel Grundman

Always force revalidation so if call which caused caching was in another
authentication "context" we don't inadvertantly leak information.

Cases where things matter involve bad caching for negative dentries
(something fails because you don't have permission, then you authenticate
and the negative dentry is not flushed) or when you have l only and
someone with rl caused the dentry to get cached, in which case stat()
information and nothing else is leaked

20 years agolinux-cleanup-after-check-bad-parent-20010416
Chas Williams [Mon, 16 Apr 2001 22:09:15 +0000]
linux-cleanup-after-check-bad-parent-20010416

check_bad_parent doing lookups is incrementing i_count and
nothing is cleaning up after it; this should make it happier

20 years agolinux-dont-d-drop-in-revalidate-20010416
Chaskiel M Grundman [Mon, 16 Apr 2001 19:11:04 +0000]
linux-dont-d-drop-in-revalidate-20010416

calling d_drop ends up precluding shrink_dcache_parent; don't call it

20 years agolinux-dentry-revalidation-rework-20010416
Chaskiel M Grundman [Mon, 16 Apr 2001 05:01:39 +0000]
linux-dentry-revalidation-rework-20010416

invalidate based on dataversion so negative dentries will actually go away

20 years agormdir(2) claims
Derrick Brashear [Sat, 14 Apr 2001 19:25:41 +0000]
rmdir(2) claims
       ENOTDIR
              pathname,  or  a  component  used as a directory in
              pathname, is not, in fact, a directory.

       ENOTEMPTY
              pathname contains entries other than . and .. .

meaning what we really want to return here is ENOTEMPTY

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

20 years agowarning-cleanup-20010414
Nathan Neulinger [Sat, 14 Apr 2001 18:49:51 +0000]
warning-cleanup-20010414

cleanup based on complaints for gcc -Wall

20 years agoallow-disabling-compile-of-obsolete-and-insecure-bits-20010414
Nathan Neulinger [Sat, 14 Apr 2001 18:45:16 +0000]
allow-disabling-compile-of-obsolete-and-insecure-bits-20010414

so that ntp, rcmds, inetd, mpp and package can be disabled easily

20 years agoupdate-linux-gcc-processor-optimization-flag-20010414
Nathan Neulinger [Sat, 14 Apr 2001 18:38:06 +0000]
update-linux-gcc-processor-optimization-flag-20010414

gcc now wants -march not -m(processor)

20 years agosolaris-ufs-call-updates-20010414
Nickolai Zeldovich [Sat, 14 Apr 2001 18:34:32 +0000]
solaris-ufs-call-updates-20010414

updates per message from Frank Batschulat <Frank.Batschulat@Sun.COM>
regarding problems with how afs hooked the solaris ufs code

20 years agoupdate-volumesets-for-deletevolentry-20010414
Mattias Amnefelt [Sat, 14 Apr 2001 18:31:49 +0000]
update-volumesets-for-deletevolentry-20010414

analog of patch for addvolentry

20 years agoinitial-freebsd-port-work-20010414
Tom Maher [Sat, 14 Apr 2001 18:27:12 +0000]
initial-freebsd-port-work-20010414

start of work on freebsd port

20 years agolinux-inode-and-dentry-validation-rework-20010413
Chas Williams [Fri, 13 Apr 2001 22:56:00 +0000]
linux-inode-and-dentry-validation-rework-20010413

gets us closer to the ideal in inode and dentry revalidation

20 years agomake-use-of-int-preblock-specific-to-s390-linux-20010412
Derrick Brashear [Thu, 12 Apr 2001 21:39:33 +0000]
make-use-of-int-preblock-specific-to-s390-linux-20010412

based on report Pavel Semerad <semerad@ss1000.ms.mff.cuni.cz>, but fixing the
root cause.

20 years agobetter-type-guessing-for-readdir-hint-20010411
Chaskiel M Grundman [Wed, 11 Apr 2001 20:17:51 +0000]
better-type-guessing-for-readdir-hint-20010411

instead of just saying unknown, if we know for sure, provide the info

20 years agowindows-compile-fixes-20010411
Volker Holfeld [Wed, 11 Apr 2001 19:57:23 +0000]
windows-compile-fixes-20010411

ignore serverLog on nt

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

correct typo for nt

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

update volinfo -filenames feature for nt

20 years agorx-keep-track-of-resent-packets-20010406
Nickolai Zeldovich [Fri, 6 Apr 2001 19:22:01 +0000]
rx-keep-track-of-resent-packets-20010406

previously the resent packet counter was never incremented

20 years agodont-fflush-fds-20010406
Nickolai Zeldovich [Fri, 6 Apr 2001 19:13:19 +0000]
dont-fflush-fds-20010406

bogus call to fflush in server log code, trying to fflush a file descriptor

20 years agopartition-include-headers-to-get-opendir-etc-20010406
Nickolai Zeldovich [Fri, 6 Apr 2001 19:11:44 +0000]
partition-include-headers-to-get-opendir-etc-20010406

so namei code comes closer to being useful in more cases

20 years agolinux24-new-afs-xsetgroups32-implementation-20010406
Chaskiel M Grundman [Fri, 6 Apr 2001 19:08:09 +0000]
linux24-new-afs-xsetgroups32-implementation-20010406

afs_xsetgroups32 is now a copy of afs_xsetgroups, with the call to
sys_setgroupsp changed to sys_setgroups32p

no change to sparc64 port stuff

20 years agodarwin-pagincred-will-notice-pags-20010406
Chaskiel M Grundman [Fri, 6 Apr 2001 19:03:46 +0000]
darwin-pagincred-will-notice-pags-20010406

make PagInCred notice pags on darwin

20 years agoremove-endif-warning-20010405
Nathan Neulinger [Thu, 5 Apr 2001 20:32:41 +0000]
remove-endif-warning-20010405

endif comment should be commentized

20 years agouse-consistent-data-typing-for-hosts-20010404
Nathan Neulinger [Wed, 4 Apr 2001 21:36:19 +0000]
use-consistent-data-typing-for-hosts-20010404

use afs_uint32 everywhere

20 years agoplacate-gmake-on-solaris-when-building-libafs-20010404
Nathan Neulinger [Wed, 4 Apr 2001 21:29:31 +0000]
placate-gmake-on-solaris-when-building-libafs-20010404

gmake tries to build something named libafs, make it not

20 years agovfsck-doesnt-build-on-hpux11-20010404
Nathan Neulinger [Wed, 4 Apr 2001 21:27:43 +0000]
vfsck-doesnt-build-on-hpux11-20010404

previous revisions caused it to be attempted to be built

20 years agomake-getfreetag-find-correct-free-tags-20010404
Hartmut Reuter [Wed, 4 Apr 2001 18:28:05 +0000]
make-getfreetag-find-correct-free-tags-20010404

"I have found a "nice" bug in namei_ops.c:

In GetFreeTag() the following code was used to find a free place in the
link table entry:

    /* Now find a free column in this row and claim it. */
    coldata = 0x7;
    for (col = 0; col<NAMEI_MAXVOLS; col++) {
        coldata <<= col * 3;
        if ((row & coldata) == 0)
            break;
    }

What happens?
With col == 0 we look for the bits 0x7. This is ok.
With col == 1 we look for the bits 0x38. This is also ok.
With col == 2 we look for the bits 0xe00 instead of 0x1c0 !!!
But later the allocation is done correctly with 0x40.
Thus the next time he will find again a free place for col == 2 !
For higher values of col the test is also incorrect, of course."

20 years agoremove-redundant-makefile-rules-20010404
Nathan Neulinger [Wed, 4 Apr 2001 18:24:02 +0000]
remove-redundant-makefile-rules-20010404

another set of redundant makefile rules removed

20 years agochanges-to-allow-identifying-threads-in-log-file-output-20010404
Hartmut Reuter [Wed, 4 Apr 2001 18:21:53 +0000]
changes-to-allow-identifying-threads-in-log-file-output-20010404

"This is an addition for MR-AFS log files where the thread
name is shown. rx switches the general server threads to listeners and
vice versa. To have reasonable names the listener must get back his old
name when converting himself to a server thread. To not bother other
programs with unresolved references I do this with two function pointers
which are initialized in the main() programs for MR-AFS."

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

use function pointers here instead of threadname