Age | Commit message (Collapse) | Author | Files | Lines |
|
(This used to be commit f5468d2c9abad41cb1e9d9b9ec6a1b56564961a4)
|
|
(This used to be commit 934964c90f39d588d2399f10e3738ab1b4e01290)
|
|
Jeremy.
(This used to be commit 46f3b04ddea4c63c6b37c7038d4bbd2ee1c9ca0c)
|
|
it may be a UNIX large writeX (which wraps len in that case).
Stevef this should fix your 128k write bug.
Jeremy.
(This used to be commit de2ebffa3c99ed28a3868fd956ef45629ca855b6)
|
|
I have a plan for dealing with the remaining..... Watch
this space.
Jeremy.
(This used to be commit 963fc7685212689f02b3adcc05b4273ee5c382d4)
|
|
callers.
Jeremy.
(This used to be commit 35aaa36f82c70964cee5d0778eb04547b226dd3f)
|
|
Jeremy
(This used to be commit 62b5ca3334598aec3304d21118f67702afc3854a)
|
|
Jeremy.
(This used to be commit 95d01279a5def709d0a5d5ae7224d6286006d120)
|
|
statics. Part of my library cleanups.
Jeremy.
(This used to be commit e848506c858bd16706c1d7f6b4b032005512b8ac)
|
|
it with accessor functions. "One global or pstring a day...." :-).
Jeremy.
(This used to be commit d50d14c300abc83b7015718ec48acc8b3227a273)
|
|
Jeremy.
(This used to be commit ac4f897f0dec0e33e79c316340b47c2803104635)
|
|
These fixes are needed for a working getaddrinfo etc. replacement.
Fixes from Wayne Davison <wayned@samba.org> from rsync.
Jeremy.
(This used to be commit 494bf6293bedbda4b10aa2eae452377b8130cd01)
|
|
Solaris.
Jeremy.
(This used to be commit 638579d75a2aa00836dc4c0c772381b775944b16)
|
|
Jeremy.
(This used to be commit 6b7ccd3efb05291f3b8779845a3842a09bb9aa07)
|
|
the incoming buffer in the non-signed case. Speeds
up writes by over 10% or so. Complete the server
recvfile implementation.
Jeremy.
(This used to be commit 81ca5853b2475f123faab3b550f0a7b24ae3c208)
|
|
v3-2-test
(This used to be commit 319dfbaf1b8d08171731d4575127f4bb5255a229)
|
|
Jeremy.
(This used to be commit 619e33248d12e764e902e4fdb5fb223cb9cfea3b)
|
|
(This used to be commit 95de80218c10a72c7b28541c3c2e475e083b68f1)
|
|
Ensure that CODEPAGEDIR, STATEDIR, and CACHEDIR have values.
(This used to be commit 10af752391be51c7381a4eaca27f29e1f2a677b3)
|
|
shouldn't either :-).
Jeremy.
(This used to be commit c3de44b6b063e126095b30536fdcb643c70e395e)
|
|
in any getnameinfo calls.
Jeremy
(This used to be commit 4d7badb0c44f287034f58d9a412e662c0fbecdc9)
|
|
Jeremy.
(This used to be commit 8c73e19f51d6e3f520cf44dd22f9b9584d4b460f)
|
|
unread bytes in the socket buffer.
Jeremy
(This used to be commit 84d22f7747126608b9460f9591bb5967d871b82d)
|
|
writeX calls.
Jeremy.
(This used to be commit 2d3ff9c502105f92720131355b41e48be8d656c2)
|
|
More changes needed to make the UNIX_LARGE_WRITEX_CAP
writes work (I'll add these tomorrow).
Jeremy.
(This used to be commit 1c71546b6152d2930b98f766311bbd161ee0ee4e)
|
|
(This used to be commit 779353b86d1699324149896f1ffa237c6ebe46ec)
|
|
Jeremy.
(This used to be commit 47dbddcb5361caa30ee60cf4e15bb50d557d1191)
|
|
This is a different fix than the bug reporter (Evgeniy Dushistov
<dushistov at mail.ru>, thanks!) created, but it lives without the boolean
status variable. Untested so far, but I can not add attachments to bugs right
now. But to me this looks really obvious.
(This used to be commit b481abf5914dcafe5642c4d9394d02603e905bbb)
|
|
The point is doing the following associations:
- non discardable state data (all TDB files that may need to be backed
up) go to statedir
- shared data (codepage stuff) go to codepagedir
The patch *does not change* the default location for these
directories. So, there is no behaviour change when applying it.
The main change is for samba developers who have to think when dealing
with files that previously pertained to libdir whether they:
- go in statedir
- go in codepagedir
- stay in libdir
(This used to be commit d6cdbfd875bb2653e831d314726c3240beb0a96b)
|
|
Hi!
Attached find a patch that I've wanted to check in for ages.
The whole area probably needs a major rewrite, but this is a
minimal patch that on a 32-bit box saves 1.5k per smbd per
defined share, twice as much on a 64-bit box.
Volker
From ebb80e664ecc49eb597a45cb57e1067fbae49e62 Mon Sep 17 00:00:00 2001
From: Volker Lendecke <vl@sernet.de>
Date: Wed, 31 Oct 2007 15:04:34 +0100
Subject: [PATCH] Change global->copymap from bool* to a bitmap
We right now have 401 parameters, so with bool being represented as a 64-bit
integer this saves about 3k of memory per smbd per share that is defined in
smb.conf.
(This used to be commit 94f2c35a683eace7f9f3dad9748aaec93f7c534f)
|
|
Jeremy.
(This used to be commit 5cf2811e8e1d9e6a1114bbdff89c333d5b374282)
|
|
Patch 1 of 3:
- Patch 1 adds the new variables
- Patch 2 makes uses of them for files belonging to the "state" path
and the "code pages" path
This patch seemed more easily acceptable, which explains why we
separated it from patch 3
- Patch 3 reassigns files to the "cache" path. Indeed all "debatable"
changes have been moved to that one
The point is adding:
- a path for non discardable state data: basically all TDB files
that may need to be backed up
- a path for shared data: mostly codepage stuff
- a path for cache data to host files such as
browse.dat, printers.tbd, <printer>.tdb
All these are currently mixed in "libdir" (${prefix}/lib/samba by default).
The patch keeps these new paths to point to ${prefix}/lib/samba by default
and does therefore not change the software behaviour. Used alone, it just
adds unused variables...so it can safely be used in sources without any
behaviour change and no impact on Samba developers work.
(This used to be commit b7d2fadbef044a89920da613b1aafc74a3d94e24)
|
|
"--with-shared-modules=vfs_zfsacl" on the configure command line.
Michael
(This used to be commit 69135360648e395eda29b15625b20a877e38bdd8)
|
|
message accordingly. Apart from not supporting create time we
now pass the S4 RAW-NOTIFY torture.
Jeremy.
(This used to be commit 8a77f520fa03afa60eac2aaeab4babe7dd8db4f0)
|
|
False instead of NULL. Fix more of the notifications to
be correct for Samba4 RAW-NOTIFY torture (we had missed
one when calling set_ea_dos_attribute().
Jeremy.
(This used to be commit 39d265375cf55eedddef2c4faa65398df73d5ed2)
|
|
not smbd/server.c
Jeremy
(This used to be commit 8fbefe18a2dc23adb0ebe488cfb37ab4a382207d)
|
|
Jeremy.
(This used to be commit 52f13d84955224ebbaead53d8428baade6a22fe0)
|
|
capability for large UNIX write if not signing and
recvfile set. Cope with large UNIX write length on
incoming processing. Stevef - we can now test 1-16Mb
writes from CIFFS.
Jeremy.
(This used to be commit 8cf78776b0a44bd026cef3d74eb11cfb415f8303)
|
|
Jeremy.
(This used to be commit b7088bb9c2a00d4717b9a7efa4bddc0c005f4efb)
|
|
Jeremy.
(This used to be commit 247675695dfbff503b5792de62f8bb1567ea2327)
|
|
to zero). If non-zero, writeX calls greater than this
value will be left in the socket buffer for later handling
with recvfile (or userspace equivalent). Definition of
recvfile for your system is left as an exercise for
the reader (I'm working on getting splice working :-).
Jeremy.
(This used to be commit 11c03b75ddbcb6e36b231bb40a1773d1c550621c)
|
|
v3-2-test
(This used to be commit 765b37c693516757686118fffd3c80cf4d5e7344)
|
|
sometimes uses a 12-word write and doesn't include a pad
byte (as Windows does). Fix this so that we are identical
to Windows clients. This will make recvfile processing
much easier to detect (as we can just read a standard
writeX header length to decide).
Jeremy.
(This used to be commit 3d3d1b806aef3617abaac46daf230ed32076e2ce)
|
|
v3-2-test
(This used to be commit c833c336435cffd20579903d22ccb910ab467e53)
|
|
at debug level 0.
metze
(This used to be commit 95f76ae7a52c6b22db22d03fed6b0848d2a61bee)
|
|
Jeremy.
(This used to be commit d5886218e53910b7015e6f13bc634e1bf338565a)
|
|
implementation. Needed for the zero-copy write code.
Jeremy.
(This used to be commit bfbdb6324c5d13bfde8b742e9c5a0e0c9092bd86)
|
|
This is the last obvious change I can see. At
this point we can start claiming IPv6 support
(Hurrah !:-).
Jeremy.
(This used to be commit bda8c0bf571c994b524a9d67eebc422033d17094)
|
|
look like a nightmare to use, so ignore them for now.
Jeremy.
(This used to be commit 814daded90781dc5a5bdd522ea8cfe5d47e6d7a7)
|
|
Right now I'm testing a vfs object. I can't right now in make test, because
"vfs objects" assumes the .so files to be in $libdir/vfs. This patch parses the
module name out of the object name in case it starts with "/". The module name
is assumed to be the last path component's basename.
(This used to be commit 95cc019af775a6ab28ea602ad767fa54d7c86197)
|