April 2004 - Wine-Devel - list.winehq.org

i810 audio fixes 1/2
by Roderick Colenbrander April 21, 2004

April 21, 2004

Hi, This is the first of two patches that will fix some problems on soundcards using the i810 audio codec. The main probem with this soundcard is that it only supports a frequency of 48kHz in 16bit stereo mode. This first patch to dsound queries if the card supports 16bit and if so it uses it, else it uses 8bit. Further it also detects if the card supports stereo or not. The patch was inspired by one from transgaming to rewind but that patch did a bit the opposite (defaulting to 8bit ..). Roderick Colenbrander

8 18

Re: Widl: Conditionals, Dotted Versions, Handles
by Robert Shearman April 21, 2004

April 21, 2004

>>> Ove Kaaven <ovek(a)arcticnet.no> 04/19/04 22:50 PM >>> man, 19.04.2004 kl. 17.03 skrev Robert Shearman: > > > Doesn't the "version" rule already handle this? > > > > It would appear not. The 'version' type is declared > as a number further up > > It's declared that the rule *returns* a number, you mean. The actual > rule is defined as > > version: > aNUM { $$ = MAKELONG($1, 0); } > | aNUM '.' aNUM { $$ = MAKELONG($1, $3); } > ; > > Now if this somehow returns 0, then you need to fix that rule, not add > another rule which makes a version of e.g. "1.2.3.4" valid syntax (but > still return a wrong result). Ok, I missed that before. I suspect it doesn't work properly because it correctly matches aNUM and attempts to convert it to a number and succeeds (but returning 0). We should re-arrange that rule so that the dotted version matches first and falls back to the second version if there is no '.'. However, I doubt I will have the time to resend that patch for a while. Rob

1 0

winetest reporting http://test.winehq.org/data/200404192305/
by Jakob Eriksson April 21, 2004

April 21, 2004

First of all, the report on http://test.winehq.org/data/200404192305/ is VERY pretty. Almost get misty eyed here. ;-) Things are coming along very nicely. Anyway, I am curious what happens if I run winetest.exe in Wine. Will it get reported accurately as Wine, or will it get misrepresented in for instance the XP column? regards, Jakob

4 7

Serial and pread
by Izak Burger April 21, 2004

April 21, 2004

Hello, I'm currently debugging a problem with serial comms over wine, and I see that when a serial device is opened (in server/serial.c) the serial_get_info function returns a type of FD_TYPE_DEFAULT, which in turn causes the serial device to be read from and written to with pread and pwrite in dlls/ntdll/file.c. Will this even work? I mean, how can you write or read to a serial port while using an offset? In any case, I changed the code to return FD_TYPE_SOCKET instead ( I thought a serial port is probably a lot more like a socket than it is a file) and now I get some activity on the serial port (before there was nothing), but it still does not work as it should. Not to worry, I just wanted to ask if all the trouble people were having with async serial comms might not be related to this? Any ideas greatly appreciated. regards, Izak

3 7

20040408 on freebsd4.9
by eternal April 21, 2004

April 21, 2004

hello all... been stumped on this one for about a week now... been on the wine-users list trying to get this worked out, and mr. pfeifer was even nice enough to lend some advice that yeilded some build progress, but it (wine) seems to hate the idea of running on my puter... here's the info...: tyan tiger 100 dual p3-850 512MB ECC ati radeon 8500LE FreeBSD4.9-RELEASE and below is the output from env (initial login as vertigo, startx, open console, su root, env): USER=vertigo MAIL=/var/mail/vertigo OLDPWD=/home/vertigo HOME=/root GTK_RC_FILES=/usr/X11R6/etc/gtk/gtkrc:/home/vertigo/.gtkrc-1.2-gnome2 PAGER=more ENV=/home/vertigo/.shrc COLORTERM=gnome-terminal LOGNAME=vertigo WINDOWID=25165870 TERM=xterm BLOCKSIZE=K GNOME_DESKTOP_SESSION_ID=Default SESSION_MANAGER=local/eTernaL.null.org:/tmp/.ICE-unix/620 PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/games:/usr/local/sbin:/usr/local/bin:/usr/X11R6/bin:/root/bin DISPLAY=:0.0 DESKTOP_STARTUP_ID= XAUTHORITY=/home/vertigo/.Xauthority SHELL=/bin/csh PWD=/home/vertigo/winecvs EDITOR=vi HOSTTYPE=FreeBSD VENDOR=intel OSTYPE=FreeBSD MACHTYPE=i386 SHLVL=1 GROUP=wheel CC=/usr/local/bin/gcc33 LD_LIBRARY_PATH=/usr/local/lib/gcc-lib/i386-portbld-freebsd4.9/3.3.1 first off, i get all the recent wine packages from cvs, i save it at /home/vertigo/winecvs, and again at /home/vertigo/winebackup. cd winecvs as vertigo, ./tools/wineinstall returns "not found". ls -la tools shows that is there... su root, check env (hence it's inclusion above), ./configure && make depend run fine. from past experience, i do the following make > winemake.txt after about 2700 lines of output, make stops with: rm -f winepath && ln -s ../../tools/winewrapper winepath /usr/local/bin/gcc33 -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o gui.o gui.c /usr/local/bin/gcc33 -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o main.o main.c /usr/local/bin/gcc33 -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o send.o send.c /usr/local/bin/gcc33 -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o util.o util.c ../../tools/winebuild/winebuild -o winetest.exe.dbg.c --debug -C. gui.c main.c send.c util.c /usr/local/bin/gcc33 -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o winetest.exe.dbg.o winetest.exe.dbg.c ../../tools/bin2res -f -o wine.ico ./gui.rc ./maketest ../.. advapi32_test.exe$(DLLEXT) comctl32_test.exe$(DLLEXT) gdi32_test.exe$(DLLEXT) kernel32_test.exe$(DLLEXT) msvcrt_test.exe$(DLLEXT) netapi32_test.exe$(DLLEXT) ntdll_test.exe$(DLLEXT) oleaut32_test.exe$(DLLEXT) rpcrt4_test.exe$(DLLEXT) shell32_test.exe$(DLLEXT) shlwapi_test.exe$(DLLEXT) urlmon_test.exe$(DLLEXT) user32_test.exe$(DLLEXT) wininet_test.exe$(DLLEXT) winmm_test.exe$(DLLEXT) winspool.drv_test.exe$(DLLEXT) ws2_32_test.exe$(DLLEXT) > winetest.rc || ( rm -f winetest.rc && exit 1 ) cp ../../dlls/advapi32/tests/advapi32_test.exe.so advapi32_test.exe.so && strip advapi32_test.exe.so cp ../../dlls/comctl32/tests/comctl32_test.exe.so comctl32_test.exe.so && strip comctl32_test.exe.so cp ../../dlls/gdi/tests/gdi32_test.exe.so gdi32_test.exe.so && strip gdi32_test.exe.so cp ../../dlls/kernel/tests/kernel32_test.exe.so kernel32_test.exe.so && strip kernel32_test.exe.so cp ../../dlls/msvcrt/tests/msvcrt_test.exe.so msvcrt_test.exe.so && strip msvcrt_test.exe.so cp ../../dlls/netapi32/tests/netapi32_test.exe.so netapi32_test.exe.so && strip netapi32_test.exe.so cp ../../dlls/ntdll/tests/ntdll_test.exe.so ntdll_test.exe.so && strip ntdll_test.exe.so cp ../../dlls/oleaut32/tests/oleaut32_test.exe.so oleaut32_test.exe.so && strip oleaut32_test.exe.so cp ../../dlls/rpcrt4/tests/rpcrt4_test.exe.so rpcrt4_test.exe.so && strip rpcrt4_test.exe.so cp ../../dlls/shell32/tests/shell32_test.exe.so shell32_test.exe.so && strip shell32_test.exe.so cp ../../dlls/shlwapi/tests/shlwapi_test.exe.so shlwapi_test.exe.so && strip shlwapi_test.exe.so cp ../../dlls/urlmon/tests/urlmon_test.exe.so urlmon_test.exe.so && strip urlmon_test.exe.so cp ../../dlls/user/tests/user32_test.exe.so user32_test.exe.so && strip user32_test.exe.so cp ../../dlls/wininet/tests/wininet_test.exe.so wininet_test.exe.so && strip wininet_test.exe.so cp ../../dlls/winmm/tests/winmm_test.exe.so winmm_test.exe.so && strip winmm_test.exe.so cp ../../dlls/winspool/tests/winspool.drv_test.exe.so winspool.drv_test.exe.so && strip winspool.drv_test.exe.so cp ../../dlls/winsock/tests/ws2_32_test.exe.so ws2_32_test.exe.so && strip ws2_32_test.exe.so LD_LIBRARY_PATH="../../libs/unicode:$LD_LIBRARY_PATH" ../../tools/wrc/wrc --nostdinc -I. -I. -I../../include -I../../include -fogui.res gui.rc *** Error code 1 Stop in /usr/home/vertigo/winecvs/programs/winetest. *** Error code 1 Stop in /usr/home/vertigo/winecvs/programs. *** Error code 1 Stop in /usr/home/vertigo/winecvs. -------------------------------------------------- at this point, i'm at a loss... i've made sure that the kernel had all the options that wine wants, because i did them when i set smp and apic_io... if anyone out there can help me out with this, it would be much appreciated. thanks in advance. shaun

1 0

EM_UNDO message handling for single line edit controls
by Santosh Siddheshwar April 21, 2004

April 21, 2004

Hi, I think there is a defect in the handling of EM_UNDO message for single line edit controls. If you go by what MSDN says this message should always return TRUE. But in EDIT_EM_Undo method (controls/edit.c) we are checking whether the style of the edit control is Read Only and returning FALSE if it is Read Only. My question is why should a read only edit control not be allowed to handle WM_PASTE, EM_UNDO and similar messages? Regards Santosh Siddheshwar ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

Re: winetest: timeout & no consoles
by Dimitrie O. Paun April 20, 2004

April 20, 2004

On April 19, 2004 6:27 pm, Ferenc Wagner wrote: > - force a 2-minute timeout on individual tests. Nice. I hope it's not too short. Does dissect & gather correctly recognize (and reflect on the HTML page) these timeouts? I'd be nice (and important :)) to know about them. -- Dimi.

2 2

Strnage fff-by-one error
by Dimitrie O. Paun April 20, 2004

April 20, 2004

Folks, This is weird one. The Winelib build of AbiWord exposes an off-by-one error. However, the Win32 build of AbiWord works perfectly (without this very problem) under Wine/ReactOS/Windows. AbiWord builds flawlessly under Winelib, not even a warning, so my guess one of our constants is off by one. The bug must appear in their code while building it, since our code to execute the Winelib app and the Win32 app is byte for byte identical. I've attached a few screenshots to demonstrate what happens when you open a menu/dialog/etc and then you close it. Namely, the invalidation (or rather painting) is off by one at right & bottom of the invalidated rectangle. -- Dimi.

4 4

Re: RFH: winetest testing
by Mehmet YASAR April 20, 2004

April 20, 2004

> 4. Programme stops on "user32:input" (the progressbar is at about 90%) Sorry, I made a mistake. I thought tests stopped on "user32:input" whereas there were running. After waiting I see a timeout error and a dialog box asking to send data to server.

1 0

Re: Replace winedefault.reg with wine.inf in README, README.it and PACKAGING
by Lionel Ulmer April 20, 2004

April 20, 2004

On Tue, Apr 20, 2004 at 09:14:28PM +0200, Ivan Leo Murray-Smith wrote: > As in subject. > License: LGPL * An initial set of registry entries. The current Wine standard is to use the regedit tool against - the 'winedefault.reg' file to generate a default registry. + the 'wine.ing' file to generate a default registry. The current preferred method of configuring/installing It's not 'use the regedit tool against...' but the 'rundll32 tool'. Lionel -- Lionel Ulmer - http://www.bbrox.org/

1 0

DateTimeCtrl doesn't select a new date
by Felipe W Damasio (Elipse Software) April 20, 2004

April 20, 2004

Greetings, Whe using CDateTimeCtrl and selecting a new date on that calendar thing, the date isn't selected and the widnow isn't closed. In fact, there's no way I can close that window, except closing the program. Attached is a little app Cesar Mello <cesar(a)elipse.com.br> wrote to show this problem. Also attached is the Windows' binary for testing purposes. Tested with the latest wine. Please let me know if I can provide any more info to help this fixed. Thanks in advance, Felipe

3 4

Re: wine/programs/winetest main.c Makefile.in
by Ferenc Wagner April 20, 2004

April 20, 2004

Alexandre Julliard <julliard(a)winehq.org> writes: > ChangeSet ID: 12069 > CVSROOT: /opt/cvs-commit > Module name: wine > Changes by: julliard(a)wine.codeweavers.com 2004/04/19 23:00:07 > > Log message: > Ferenc Wagner <wferi(a)afavant.elte.hu> > - Change to console mode so that winetest runs under WinRash. > - Move from spawnvp() to CreateProcess(). > - Force a 2-minute timeout on individual tests. > > Patch: http://cvs.winehq.org/patch.py?id=12069 Hi Alexandre, Is there a forcing reason to have oldstdout initialized to -1 besides playing safe? Not that I object it, just want to avoid grave errors in the future. -- Feri.

2 1

winetest: what next?
by Dimitrie O. Paun April 20, 2004

April 20, 2004

OK, So assume we get the show rolling, tests will run on various boxes, results will be send back to WineHQ, nice colorful pages will be generated, and a beautiful index page will put all these together, being linked from the Status (http://www.winehq.org/site/status) page. Are we done? I think not. This is a very important piece of infrastructure, and it's buried rather deep into the bowels of WineHQ. Moreover, it's not easy to monitor the changes in testing status. As such, I would like to have a "Weekly testing update" posted to in the News section, with stuff like: how many tests were added, how many failed/successed, etc. since last week. Maybe a list of DLLs where changes were observed, and so on. To get these statistics (and possibly other interesing bits) in the future, I'm thinking at storing all these test results in the database. NB: I'm not proposing that we change the current processing of dissect & gather, it is very nice, and it's a big bonus that it works without a database. All I'm saying is that we can augment dissect to also save stuff to the database. For the database, we need two simple tables: wine_test_releases, and wine_test_results. wine_test_releases: build_id VARCHAR(20) NOT NULL cvs_ts VARCHAR(20) cvs_tag VARCHAR(20) url VARCHAR(256) NOT NULL publish_ts TIMESTAMP NOT NULL This table will be populated by Brian's script, when a new release is publised. One of cvs_ts or cvs_tag will be NOT NULL, and will specify how to get to the exact source used for the build. Nightly builds will use cvs_ts (cvs -D <cvs_ts>), while official releases done by Alexander will use cvs_tag (cvs -r <cvs_tag>). Additional metadata may be required here, suggestions welcome. wine_test_results: build_id VARCHAR(20) NOT NULL tester_id VARCHAR(20) test_name VARCHAR(20) NOT NULL subtest_name VARCHAR(40) NOT NULL file_name VARCHAR(128) NOT NULL line_number INTEGER NOT NULL result INTEGER // 0 = success, 1 = timeout, etc windows_version VARCHAR (15) // 4.3.1233 I've probably missed something, so feel free to add to this. Comments, flames? -- Dimi.

3 3

Winetest results uploaded... where?
by Jakob Eriksson April 20, 2004

April 20, 2004

winetest.exe from http://www.astro.gla.ac.uk/users/paulm/WRT/CrossBuilt/ works fine for me in Windows XP, but where is the stuff uploaded? "test.winehq.org", ok fine, but where is the info going? regards, Jakob

2 1

Re: List other downloads on winehq
by Ivan Leo Murray-Smith April 20, 2004

April 20, 2004

> But since you have not agreed to any license, you are free to personally > use your download however you wish. Thanks to the new EU laws on copyright (Not the patent one, this is another thing), this will soon change. Ivan.

2 1

Appdb - Add submit time to appQueue
by Chris Morgan April 20, 2004

April 20, 2004

* appsubmit.php, admin/adminAppQueue.php, tables/appdb_tables.sql Chris Morgan <cmorgan(a)alum.wpi.edu> It would be nice to know when entries to the appdb were submitted. Jeremy, you'll want to run this alter command to add the column to the existing table. 'alter table appQueue add column submitTime timestamp;' Comments? Chris

2 1

A Few Newbie questions
by Lonnie Cumberland April 20, 2004

April 20, 2004

Hello All, I have been watching Wine grow over a period of time and it looks like things are getting better each time that I visit the site. Let me first say "GREAT JOB" to all of the developers on the project. I am wondering a few things and hope someone could take a stab at answering these for me. 1.) Is wine thread save and can I run many instances without problems? Also, can multiple users be running multiple instances without major degrades in system performance for each instance? 2.) What is the main differences between the WineX and Wine developments? Which is more stable? Thanks in advance, Lonnie.

3 2

Wine Directory Mapping sub-system
by Lonnie Cumberland April 20, 2004

April 20, 2004

Hello All, Is there some where that I can read docs that talk about the directory mapping sub-system? In particular, I am interested in the way that you can "map" user directories in the configuration file in a users HOME directory. What I am thinking about is how one might use this same idea to "Jail" Linux user directories so that users cannot navigate out of their home directory as a security feature. Just playing with ideas, but I thought that the Wine sub-system might offer a great feature for this in that applications could not do anything even if they were to "Break" Out" as they would be going through a translation layer just like the Window applications but onlu still being Linux applications. Any ideas on this? Thanks, Lonnie

2 1

Re: Cross compile fix
by Alexandre Julliard April 20, 2004

April 20, 2004

Hans Leidekker <hans(a)it.vu.nl> writes: > -.res.res.cross.o: > +.rc.res.cross.o: > $(CROSSWINDRES) -i $< -o $@ Why do you need that? windres should work just fine with a .res file as input. What error do you get? -- Alexandre Julliard julliard(a)winehq.org

3 4

Re: Re: [D3D8] Some vertex shaders fixes
by Christian Costa April 20, 2004

April 20, 2004

> Message du 20/04/04 02:54 > De : Alexandre Julliard > A : Christian Costa > Copie à : wine-devel(a)winehq.org > Objet : Re: [D3D8] Some vertex shaders fixes > Christian Costa writes: > > > + if ( !RegOpenKeyA( HKEY_LOCAL_MACHINE, "Software\\Wine\\d3d", &hkey) && > > !RegQueryValueExA( hkey, "vs_mode", 0, NULL, buffer, &size) ) > > I changed the key name to "Software\Wine\Direct3D" and the value to > "VertexShaderMode", in an attempt to make it a bit less obscure for > users who stumble upon it in regedit. Good idea! :-) > Feel free to suggest other names > if you don't like these. > It's fine. :-) Christian

1 0

Re: Re Winebuild : Fix spec32 file generation for Darwin
by Alexandre Julliard April 19, 2004

April 19, 2004

emmanuel maillard <e.rsz(a)libertysurf.fr> writes: > Sorry my previous mail include a broken patch > This one is the good one. > > Changelog : > Fix spec32 file generation for Darwin Why are you removing the init code and the SizeOfImage thing? This doesn't look right at all. -- Alexandre Julliard julliard(a)winehq.org

2 3

Re: [D3D8] Some vertex shaders fixes
by Alexandre Julliard April 19, 2004

April 19, 2004

Christian Costa <titan.costa(a)wanadoo.fr> writes: > + if ( !RegOpenKeyA( HKEY_LOCAL_MACHINE, "Software\\Wine\\d3d", &hkey) && > !RegQueryValueExA( hkey, "vs_mode", 0, NULL, buffer, &size) ) I changed the key name to "Software\Wine\Direct3D" and the value to "VertexShaderMode", in an attempt to make it a bit less obscure for users who stumble upon it in regedit. Feel free to suggest other names if you don't like these. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: d3d8 CVS brokenness?
by Ivan Leo Murray-Smith April 19, 2004

April 19, 2004

CVS current (Read right now) works for me. Ivan.

4 4

Re: List other downloads on winehq
by Ivan Leo Murray-Smith April 19, 2004

April 19, 2004

> In some jurisdictions... OK, in all jurisdictions that recognise the rule of law and the right of copyright holders. I'm sure you could use it in Somalia or Iraq without anybody sueing you. Ivan.

3 2

Re: appdb/. appview.php
by hatky April 19, 2004

April 19, 2004

WTF? I think this patch needed to have the file bug.php, it is not there, there is no such file, now the link no longer work, plz fix ASAP, either revert the patch or send the file............. On Monday 19 April 2004 23:48, Jeremy Newman wrote: > ChangeSet ID: 12051 > CVSROOT: /opt/cvs-commit > Module name: appdb > Changes by: jnewman(a)wine.codeweavers.com 2004/04/19 15:48:46 > > Modified files: > . : appview.php > > Log message: > Tony Lambregts <tony_lambregts(a)telusplanet.net> > Add bug.php to better integrate AppDB with Bugzilla > > Patch: http://cvs.winehq.org/patch.py?id=12051 > > Old revision New revision Changes Path > 1.4 1.5 +1 -2 appdb/appview.php -- Hatky, Worshiper of wine (http://www.winehq.org/) Impossible is only an opinion.

3 2

List other downloads on winehq
by Vincent Béron鲯n April 19, 2004

April 19, 2004

Following a discussion on IRC with a new user, we probably want to list on the Download page on winehq what other packages are available on sf, besides Wine per se. Also, besides DCOM95, there are a couple other Microsoft packages that people are sometimes encouraged to install to get some functionality that Wine hasn't matched yet. Does somebody have a list of what can be found (and installed on a non-Windows platform) on Microsoft's site? Vincent

3 2

Re: d3d8 CVS brokenness?
by Ivan Leo Murray-Smith April 19, 2004

April 19, 2004

> Like several people here, I haven't been able to build from CVS for the > last few days. Maybe you haven't used -PAd when updating, or maybe you should try pulling down an entire new source tree. Or you have out of date headers. In any case whoever maintains that code will presumably read the messages on the list, and look into it. Ivan.

3 2

Re: Widl: Conditionals, Dotted Versions, Handles
by Robert Shearman April 19, 2004

April 19, 2004

>>> Ove Kaaven <ovek(a)arcticnet.no> 04/18/04 15:25 PM >>> lør, 17.04.2004 kl. 18.53 skrev Robert Shearman: > > Hi, > > > > This patch adds support for several features I found lacking whilst > > attempting to generate headers for RPC interfaces. > > Is this needed? > > @@ -350,6 +354,7 @@ > | tSWITCHTYPE '(' type ')' { $$ = make_attrp(ATTR_SWITCHTYPE, type_ref($3)); } > | tUUID '(' aUUID ')' { $$ = make_attrp(ATTR_UUID, $3); } > | tV1ENUM { $$ = make_attr(ATTR_V1ENUM); } > + | tVERSION '(' version '.' version ')' { $$ = make_attrv2(ATTR_VERSION, $3, $5); } > | tVERSION '(' version ')' { $$ = make_attrv(ATTR_VERSION, $3); } > | tWIREMARSHAL '(' type ')' { $$ = make_attrp(ATTR_WIREMARSHAL, type_ref($3)); } > | pointer_type { $$ = make_attrv(ATTR_POINTERTYPE, $1); } > > Doesn't the "version" rule already handle this? It would appear not. The 'version' type is declared as a number further up and when it attempts to parse a dotted version it just returns 0. Rob

2 1

Re: List other downloads on winehq
by Ivan Leo Murray-Smith April 19, 2004

April 19, 2004

You don't need windows installed at all. You only need a windows license, the EULA says you must have a valid license for the computer where you're installing IE. It doesn't matter if windows is installed or not. I got this confirmed by Microsoft. Ivan.

1 0

Re: d3d8 CVS brokenness?
by brettholcomb＠charter.net April 19, 2004

April 19, 2004

I'm not at my machine now but I tried to build CVS yesterday (see my post) and got errors with d3d undefined variables, etc. In short I couldn't build it - is this part of the same issue? > > From: Joel Konkle-Parker <jjk3(a)msstate.edu> > Date: 2004/04/19 Mon PM 03:06:43 GMT > To: wine-devel(a)winehq.com > Subject: d3d8 CVS brokenness? > > I've noticed that d3d8 has been broken the past couple days in CVS. Is > this a work in progress that'll get fixed soon, or does it need to be > looked at? > > > -- > Joel Konkle-Parker > Webmaster [Ballsome.com] > > E-mail [jjk3(a)msstate.edu] > > >

1 0

Re: Make Unicode strings 'static const' (part 3)
by Dimitrie O. Paun April 19, 2004

April 19, 2004

On Mon, 19 Apr 2004, Francois Gouget wrote: > As for the previous patch it's important to check that the function does > not actually try to write to these strings. And indeed there are a few > cases where the function modifies the string, typically to change the > drive letter in strings like 'A:\'. What about a list of things that weren't "const static"-ied? Maybe we should have a policy of having such string as *always* "const static" (and maybe have puk's sparse tool warn for non conforming ones), and in places where they need to modify them, they just copy them on stack explicitly... -- Dimi.

2 1

dcom95
by Ivan Leo Murray-Smith April 19, 2004

April 19, 2004

We currently have dcom95 on the sf download page. This is a violation of the sf TOS, and a waste of bandwidth for sf. I propose we link the microsoft download page, instead of having the download on a page that should host only open source. Any objections? Ivan.

1 0

Still trying to install ZooTycoon (demo-version)
by pvriens＠xs4all.nl April 19, 2004

April 19, 2004

Hi, I'm still trying (won't give up) to install the ZooTycoon demo from M$. Unfortunatly I've been debugging for the last weeks and cannot seem to find the cause/error/reason it fails. I'm downloading new versions from CVS daily but only seem to find other stuff :-). The installation (copying of files) stops at the point when a new background picture should be shown during the copying. I still have the feeling it all boils down to messaging. I see that all PostMessage's are done to the main thread (0009). The PeekMessage's are all done from a relevant thread, ergo : nothing found. If I look at the debug-log the only entries at the end are thousends of PeekMessage's. I also see when debugging it on windows that the PeekMessages are done from the main thread ! Does somebody have any hint how to tackle this one ? It appears that several M$ demo programs suffer the same problem, so it could be worthwile to pursue this one. Cheers, Paul.

1 0

d3d8 CVS brokenness?
by Joel Konkle-Parker April 19, 2004

April 19, 2004

I've noticed that d3d8 has been broken the past couple days in CVS. Is this a work in progress that'll get fixed soon, or does it need to be looked at? -- Joel Konkle-Parker Webmaster [Ballsome.com] E-mail [jjk3(a)msstate.edu]

1 0

err:clipping:CLIPPING_UpdateGCRegion hVisRgn is zero. Please report this.
by Brad Campbell April 19, 2004

April 19, 2004

G'day all, Due to restrictions in my current setup I'm not subscribed to wine-devel at the moment, please cc me. This is a longstanding bug with Word97. It exists in Crossover office 2.1.0 as well and was reported in the Codeweavers database months ago. It usually results in the "style" drop down box on the far left of the toolbar blanking it's entries. I have a pretty complex default template that has a number of styles in it that all render with funky true-type fonts. When you scroll up and down this list a couple of times, wine ceases to render the entries. It used to lock up in an endless loop somewhere, but now it bombs with the below error message. I compile from CVS a couple of times a week, and subject to time I'm happy to do anything at all to assist in debugging this if it will help someone. (I have worked around it by creating tool bars with all my styles on it so I don't have to use this box) Regards, Brad fixme:hook:NotifyWinEvent (32773,0x30035,-4,11)-stub! fixme:hook:NotifyWinEvent (32773,0x30035,-4,10)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,9)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,8)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,7)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,6)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,5)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,4)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (32773,0x30035,-4,3)-stub! err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 116 bytes fixme:hook:NotifyWinEvent (5,0x10022,-3,0)-stub! fixme:ole:I_RpcWindowProc (0x10026,0000001c,00000000,00000000): stub err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:local:LOCAL_GetBlock not enough space in GDI heap 10c7 for 28 bytes err:clipping:CLIPPING_UpdateGCRegion hVisRgn is zero. Please report this.

2 1

Re: Installing DCOM98 with latest CVS
by pvriens＠xs4all.nl April 18, 2004

April 18, 2004

Went back to the new implementation of GetDiskFreeSpace*. This gives me the following tracing after starting 'wine dcom98.exe' with full tracing on: 0009:Call kernel32.GetDiskFreeSpaceA(00000000,406cfc40,406cfc38,406cfc3c,406cfc28) ret=01004f8a 0009:trace:volume:GetDiskFreeSpaceA (null),0x406cfc40,0x406cfc38,0x406cfc3c,0x406cfc28 0009:trace:volume:GetDiskFreeSpaceW (null),0x406cfc40,0x406cfc38,0x406cfc3c,0x406cfc28 0009:Call ntdll.RtlDosPathNameToNtPathName_U(4039d540 L"\\",406cf770,00000000,00000000) ret=40340d1e 0009:trace:file:RtlDosPathNameToNtPathName_U (L"\\",0x406cf770,(nil),(nil)) 0009:trace:file:RtlGetFullPathName_U (L"\\" 520 0x406cf488 (nil)) 0009:trace:heap:RtlAllocateHeap (0x401a0000,00000002,00000018): returning 401d6b70 0009:trace:file:RtlDosPathNameToNtPathName_U At the end, returning TRUE 0009:Ret ntdll.RtlDosPathNameToNtPathName_U() retval=00000001 ret=40340d1e 0009:Call ntdll.NtOpenFile(406cf7e8,00000000,406cf778,406cf768,00000000,00000021) ret=40340d74 0009:trace:ntdll:NtCreateFile handle=0x406cf7e8 access=00000000 name=L"\\??\\E:\\" objattr=00000040 root=(nil) sec=(nil) io=0x406cf768 alloc_size=(nil) attr=00000000 sharing=00000000 disp=1 options=00000021 ea=(nil).0x00000000 0009:trace:heap:RtlAllocateHeap (0x401a0000,00000002,00000128): returning 401d9ff0 0009:trace:heap:RtlFreeHeap (0x401a0000,00000002,401d9ff0): returning TRUE 0009:warn:ntdll:NtCreateFile L"\\??\\E:\\" not found (c000003a) Hope this helps someone with better understanding of the code. Cheers, Paul.

2 2

Re: dinput/joystick_linuxinput #2
by Christoph Frick April 18, 2004

April 18, 2004

On Sun, Apr 18, 2004 at 09:37:07PM +0200, Christoph Frick wrote: *sic* stupid brain... License: LGPL > - added code to have the joystick state be set to the assumed center > values of the joystick mapped to the range the application wants them > - generate events after setting the apropiate values in This->js > - replaced magic ints with defines from linux/input.h > - removed mapping code that generated wrong values for axes, that have > the center on the min or max -- cu

1 0

Re: GetModuleFileName
by Juan Lang April 18, 2004

April 18, 2004

Alexandre wrote: > I doubt it fills the buffer with nulls if it is too > large though, so strncpy is probably not the right > thing to use here (in fact strncpy is almost always > the wrong function to use, we should probably have > a janitorial task to fix this everywhere). What is the right thing? strlcpy? --Juan __________________________________ Do you Yahoo!? Yahoo! Photos: High-quality 4x6 digital prints for 25� http://photos.yahoo.com/ph/print_splash

2 1

Re: Add a win32 hello world program
by Alexandre Julliard April 18, 2004

April 18, 2004

Mike Hearn <mike(a)navi.cx> writes: > Mike Hearn <mike(a)navi.cx> > Add a win32 hello world program which sets up a simple window and runs > the message loop. This can be used as a starting point for test programs. I think we have already enough small winelib apps that can be used as starting points, like winver or winepath. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: GetModuleFileName
by Alexandre Julliard April 18, 2004

April 18, 2004

Eric Pouech <pouech-eric(a)wanadoo.fr> writes: > Interestingly enough, GetModuleFileName[AW] doesn't '\0'-terminate the > buffer if it is too small > Our current implementation was terminating the buffer. > This patch fixes this, and adds a test case for that behavior and > patches all the places in wine which are potentially impacted by it I doubt it fills the buffer with nulls if it is too large though, so strncpy is probably not the right thing to use here (in fact strncpy is almost always the wrong function to use, we should probably have a janitorial task to fix this everywhere). -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: dlls/kernel/console.c::SetConsoleCtrlHandler() cleanup/fixup
by Alexandre Julliard April 18, 2004

April 18, 2004

Peter Riocreux <par+wine_devel(a)silistix.com> writes: > It gives every impression of working for me, but I have only tested it > with one application. > > Suggestions for better extended error codes appreciated. Error codes should not be based on suggestions, what you should do is write a small test program under Windows to find out the real codes. -- Alexandre Julliard julliard(a)winehq.org

9 16

Re: msvcrt
by Uwe Bonnes April 18, 2004

April 18, 2004

>>>>> "Eric" == Eric Pouech <pouech-eric(a)wanadoo.fr> writes: Eric> I was about to implement the fd inheritance in msvcrt, when I came Erich, the attachment seems broken again... -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

3 2

Re: List other downloads on winehq
by Ivan Leo Murray-Smith April 18, 2004

April 18, 2004

> Internet Explorer ..... Installing IE requires a windows license for the computer where you're installing it. This sort of restricts it's use with wine. Ivan.

2 1

arch tip #2
by Mike Hearn April 18, 2004

April 18, 2004

Hi everyone, I just checked in a new update to the genpatch script. It can now send an email to wine-patches automatically given a set of local changesets. Use it like this: ./genpatch -m 5 6 7 if you did three commits, patch-5, patch-6 and patch-7 and want to send them all to wine-patches as one email. The log message and summary of the first patch will be used, and all three patches will be combined into one. You get to edit the email before it's sent and will be asked for confirmation, so you can read it over one last time. By default it will send mail to localhost - if you don't want to setup sendmail you can have "export GENPATCH_RELAY=smtp.myisp.com" in your shell init scripts to make it use a different mail relay. So in summary, development on a series of related improvements to wine goes like this: (setup arch and checkout the cvs gateway tree) ./branch some-branch cd .. cd some-branch tla make-log (make some changes, noting what they are in the arch-created log file as you go) tla commit * committed whoever(a)where-ever.org--2004/wine--some-branch--0.9--patch-2 tla make-log (more changes) tla commit * committed whoever(a)where-ever.org--2004/wine--some-branch--0.9--patch-3 ./merge (this brings your tree up to date with winehq, only bother if you think there has been enough drift to make it worthwhile) ./genpatch -m 2 3 (read generated email, to confirm delete the first line + save + quit) thanks -mike

1 0

Security of WINE and ReactOS: was Documents and settings
by Steven Edwards April 18, 2004

April 18, 2004

Hello Alex, I am going to drag this semi-off topic for a bit while I plant some ideas in people heads. --- Aleksey Bragin <aleksey(a)studiocerebral.com> wrote: > If ROS will support (and I bet it will!) NT 4.0 security model (or, > better, > Win2k then) it would be just great! > Certainly people with deeper knowledge of this security model + > knowledge of > unix and also linux implementation model can specify the differences, > and > what's missing in them. Note I only know a little about the Local Securty Authority but I think its not going to be to hard to implement if Wine and ReactOS work together on it. /me just doubts how much he can write. The unix security design of users and groups with permissions is not bad its just outdated. The nice thing about Unix is adding new security modules via PAM is not to bad except they are only for authentication. The unix concept of groups, users and permissions needs to be moved forward about 20 years. The SELinux stuff has really helped alot in this regard. (Please dont flame its the truth) I recently addressed this in a discussion about ReactOS. Currently our lsass does not exist. I think we have what we have parts of the security reference monitor already implemented in ntoskrnl and most of the parts are there for winlogon and the SAM database so we need to develop the lsass services and build out the authentaticaion modules for MSV1_0 auth. One of the nice things about the design of the Windows security module is that we can make plugins at both ends so that users can be granted access either based on a "domain" concept in winlogin/Gina using plugins for LDAP and PAM or via the lsass so users can be authencated locally. It would be nice if we could work with the Winehq people on this as I think we can share the parts of the security subsystem that reside in lsass. Think of it like this (Kernel support) wineserver/ntosknrl need to both implement the security refernce montior for privlaged use of the local system resources. I dont know how much of this wineserver really needs to take in to account in the initial incarnation. (Local security subsystem) Lsass works interactivly with services and the login system and can accept all sorts of nice plugins so we are not limited to just the standard Windows authentication. As a matter of fact you could replace large parts of they authentication system if you are supper paronoid. Think of lsass as the sentry for Windows. It talks to the SRM to make sure you are not doing anything you shouldnt be. (User interaction) Interaction with the user will come from Winlogon on ReactOS or on Unix the user should already be authenticated via Linux security. This can of course be done via the authentication modules described above. Once the user I wish I could attach the nice chart from Inside Windows NT so you could see the security subsystem. Its quite a piece of work and is quite a shame Windows security gets a bad name due to slack administrators. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html

4 3

Installing DCOM98 with latest CVS
by pvriens＠xs4all.nl April 18, 2004

April 18, 2004

Hi, I was trying to install dcom98.exe with the latest CVS and run into several errors ranging from file-not-found errors upto problems with getting the free diskspace of drives. Tracing showed (amongst others): trace:file:GetLongPathNameW L"E:\\TOOLS",0x406cf144,1024 trace:file:RtlDosPathNameToNtPathName_U (L"E:\\TOOLS",0x406cde4c,0x406cde54,(nil)) trace:file:RtlGetFullPathName_U (L"E:\\TOOLS" 520 0x406cdc08 0x406cde54) trace:file:RtlDosPathNameToNtPathName_U At the end, returning TRUE trace:file:GetLongPathNameW not found L"E:\\TOOLS"! so we didn't get an handle from RtlDosPathNameToNtPathName_U ! After some extra tracing I decided to revert some patches and got this installation working after reverting the following patches: http://cvs.winehq.org/cvsweb/wine/dlls/kernel/volume.c.diff?r1=1.7&r2=1.8 log : Reimplemented GetDiskFreeSpaceW and GetDiskFreeSpaceExW http://cvs.winehq.org/cvsweb/wine/dlls/kernel/change.c.diff?r1=1.4&r2=1.5 log : Make FindFirstFileExW and FindFirstChangeNotificationW use NtOpenFile directly http://cvs.winehq.org/cvsweb/wine/files/drive.c.diff?r1=1.114&r2=1.115 log : Reimplemented GetDiskFreeSpaceW and GetDiskFreeSpaceExW I will investigate further, but just to let you know. I didn't try other programs. Everything done on a clean non-windows installation. Cheers, Paul.

1 0

Re: Widl: Conditionals, Dotted Versions, Handles
by Ove Kaaven April 18, 2004

April 18, 2004

lør, 17.04.2004 kl. 18.53 skrev Robert Shearman: > Hi, > > This patch adds support for several features I found lacking whilst > attempting to generate headers for RPC interfaces. Is this needed? @@ -350,6 +354,7 @@ | tSWITCHTYPE '(' type ')' { $$ = make_attrp(ATTR_SWITCHTYPE, type_ref($3)); } | tUUID '(' aUUID ')' { $$ = make_attrp(ATTR_UUID, $3); } | tV1ENUM { $$ = make_attr(ATTR_V1ENUM); } + | tVERSION '(' version '.' version ')' { $$ = make_attrv2(ATTR_VERSION, $3, $5); } | tVERSION '(' version ')' { $$ = make_attrv(ATTR_VERSION, $3); } | tWIREMARSHAL '(' type ')' { $$ = make_attrp(ATTR_WIREMARSHAL, type_ref($3)); } | pointer_type { $$ = make_attrv(ATTR_POINTERTYPE, $1); } Doesn't the "version" rule already handle this?

1 0

Re: List other downloads on winehq
by Ivan Leo Murray-Smith April 18, 2004

April 18, 2004

> Does somebody have a list of what can be > found (and installed on a non-Windows platform) on Microsoft's site? It's not a complete list, but I've got the following links vb runtime http://www.microsoft.com/downloads/details.aspx?FamilyID=bf9a24f9-b5c5-48f4… the windows installer for win9x http://www.microsoft.com/downloads/details.aspx?FamilyID=cebbacd8-c094-4255… the windows installer for winnt/2k http://www.microsoft.com/downloads/details.aspx?FamilyID=4b6140f9-2d36-4977… The comctl32 dll, the eula says you can only distribute it with an app that runs on windows, but it also says you can use it, and doesn't say that to do so you need a windows license (The eula is the same one you get with the sdk), so it's ok to use it on wine. http://www.microsoft.com/downloads/details.aspx?FamilyID=6f94d31a-d1e0-4658… gdi+ components, same license as comctl32 http://www.microsoft.com/downloads/details.aspx?FamilyID=bfc0b436-9015-43e2… CAPICOM.DLL http://www.microsoft.com/downloads/details.aspx?FamilyID=860ee43a-a843-462f… ShFolder DLL http://www.microsoft.com/downloads/details.aspx?FamilyID=6ae02498-07e9-48f1… I think all these windows components can be used without problems on wine, but probably winehq can't redistribute any of them. BTW, we have dcom95 on the sourceforge page, this is a violation of the sf TOS, wouldn't it be a good idea to link to the microsoft dcom95 download page instead? Like this we'll user their bandwidth, and save a little for sf.net. Ivan.

1 0

Re: (no subject)
by Ivan Leo Murray-Smith April 18, 2004

April 18, 2004

> If not, I would like to give it a try, but I need some > documentation c.q. information on the functionallity > of these functions /what they do etc http://msdn.microsoft.com/library/default.asp?url=/library/en-us/winsock/wi… Ivan.

1 0

Change to CompareStringW introduced regression in QuArK (bug #2159)
by David Costanzo April 18, 2004

April 18, 2004

I have been investigating a regression in the functionality of QuArK, a popular level editor for Quake. The regression is filed as bug #2159. I have identified the cause of the regression as changes made to CompareStringW from revision 1.25 to revision 1.26 of wine/dlls/kernel/locale.c. I do not know how to proceed from here. The relevant modification is changing CompareStringW from calling strncmpiW/strncmpW to calling wine_compare_string. I hacked my code to call both the new and old version and print out whenever the two return different values. Calling wine_compare_string seems to have changed the behavior of comparing strings that have accent marks with strings that don't have accent marks when the NORM_IGNORECASE flag is set. I don't understand unicode collation, nor do I know how QuArK is using CompareStringW. I don't doubt that CompareStringW should call wine_compare_string, but if this is a bug in wine_compare_string, I don't know how to fix it. Does anyone have any suggestions on what my next step should be? Thanks, David Costanzo __________________________________ Do you Yahoo!? Yahoo! Photos: High-quality 4x6 digital prints for 25� http://photos.yahoo.com/ph/print_splash

1 0

Re: [D3D8] Fix on Hardware Vertex Shaders and D3DTADDRESS_MIRROR
by Lionel Ulmer April 18, 2004

April 18, 2004

> - fix the SetTextureStageState for D3DTADDRESS_MIRROR using dynamic GL > detection (and not build time GL headers) By the way, you should also check for the 'GL_IBM_texture_mirrored_repeat' extension. I know of no OpenGL drivers who has the IBM and not the ARB extension, but well, it does cost nothing to be pedantic :-) Lionel -- Lionel Ulmer - http://www.bbrox.org/

1 0

RE: Security of WINE and ReactOS: was Documents and settings
by Joshua Walker April 18, 2004

April 18, 2004

The one known as "Steven Edwards" hath scripted: --------------------------------- The unix security design of users and groups with permissions is not bad its just outdated. The nice thing about Unix is adding new security modules via PAM is not to bad except they are only for authentication. The unix concept of groups, users and permissions needs to be moved forward about 20 years. The SELinux stuff has really helped alot in this regard. (Please dont flame its the truth) ----------------------------------- Looking at security from a Unix up to windows through WINE as opposed to looking down to UNIX from Windows through WINE, I have do disagree. The UNIX security system is not really "outdated" just "diffrent". One of the reasons why it's still around after 20 years is because it works. As I have never owned Windows after ME drug it's carcass across the computing landscape, I don't really quite grok windws security anyway. From futzing with an XP box at work, I don't see any real obvious way of locking down permissions on files and such. Right click/properties on a file gives me the same tierd DOS flags that haven't changed since DOS 3.0 I don't see anyting that's blocking me from deleting, say, advapi32.dll with extreame prejudice. If I went into my /lib dir and tried a rm -fr as a normal user, my system will kindly tell me to go away and play in my /home directory. The question is quite intresting. How does one "translate" a security philosophy? Do we want to give WINE the ability to hijack wsock32.dll? (wsock32.dll.so) This is often done with ISP proprietary installers, *COUGH*-AOL-*COUGH*, and various spyware apps. Has this been considered?

2 1

(no subject)
by Robbert Xerox April 17, 2004

April 17, 2004

hi, I'm wondering if anybody is working on the WSALookupServiceBegin functions in winsock (socket.c) . If not, I would like to give it a try, but I need some documentation c.q. information on the functionallity of these functions /what they do etc. Any help appreciated. Robbert ____________________________________________________________ Yahoo! Messenger - Communicate instantly..."Ping" your friends today! Download Messenger Now http://uk.messenger.yahoo.com/download/index.html

1 0

[PATCH?] Finally found a solution to serial port writes
by Geoffrey Hausheer April 17, 2004

April 17, 2004

The following patch is probably wrong, but it fixes my problem with WriteFile failing when writing to a COM port. The problem appears to be that NtWaitForSingleObject sends a count of '1' which results in the wine_server_call in NTDLL_wait_for_multiple_objects returning 'STATUS_USER_APC' which aborts the execution before 'call_apcs' gets called (which actually does the write). calling 'NTDLL_wait_for_multiple_objects' with count = 0 allows the 'call_apcs' function to get called, and TzMax50 actually works perfectly (along with my nasty hack to allow the menus to work) as a reminder, the COM port is opened in non-overlapped mode, and all timeouts are set to 0. (I didn't submit this to wine-patches, since i'm not at all sure it is generally correct) I can also try to write a test-case that shows this works (sort of...I can test that WriteFile returns the right result, but not that data was actually written on the serial port. I know of no way to do that without external hardware). I don't know how to make it general enough though (what to do when there is no COM port attached to the machine?) .Geoff --- orig/dlls/ntdll/file.c +++ mod/dlls/ntdll/file.c @@ -585,7 +585,7 @@ ret = register_new_async(&ovp->async); if (ret != STATUS_SUCCESS) return ret; - if (flags & FD_FLAG_TIMEOUT) + if (hEvent != NULL && flags & FD_FLAG_TIMEOUT) { NtWaitForSingleObject(hEvent, TRUE, NULL); NtClose(hEvent);

1 0

WININET: make WCHAR strings const where possible
by Mike McCormack April 17, 2004

April 17, 2004

"Dmitry Timoshkov" <dmitry(a)baikal.ru> wrote: > It would be much better if you could mark all string constants > as such using 'const' key word in all your patches (including > already committed, submitted and future ones). Perhaps we should make a janitorial task to CONST-ify strings where possible? Mike ChangeLog: * make WCHAR strings const where possible Index: dlls/wininet/http.c =================================================================== RCS file: /home/wine/wine/dlls/wininet/http.c,v retrieving revision 1.56 diff -u -r1.56 http.c --- dlls/wininet/http.c 31 Mar 2004 20:09:54 -0000 1.56 +++ dlls/wininet/http.c 12 Apr 2004 14:12:48 -0000 @@ -470,8 +470,8 @@ { UINT len; LPWSTR in, out; - WCHAR szBasic[] = {'B','a','s','i','c',' ',0}; - WCHAR szColon[] = {':',0}; + const WCHAR szBasic[] = {'B','a','s','i','c',' ',0}; + const WCHAR szColon[] = {':',0}; len = lstrlenW( username ) + 1 + lstrlenW ( password ) + 1; in = HeapAlloc( GetProcessHeap(), 0, len*sizeof(WCHAR) ); @@ -504,7 +504,7 @@ { HTTPHEADERW hdr; INT index; - WCHAR szProxyAuthorization[] = { + const WCHAR szProxyAuthorization[] = { 'P','r','o','x','y','-','A','u','t','h','o','r','i','z','a','t','i','o','n',0 }; hdr.lpszValue = HTTP_EncodeBasicAuth( username, password ); @@ -538,10 +538,9 @@ WCHAR proxy[MAXHOSTNAME + 15]; /* 15 == "http://" + sizeof(port#) + ":/\0" */ WCHAR* url, szNul[] = { 0 }; URL_COMPONENTSW UrlComponents; - WCHAR szHttp[] = { 'h','t','t','p',':','/','/',0 }, szSlash[] = { '/',0 } ; - /*WCHAR szColon[] = { ':',0 }; */ - WCHAR szFormat1[] = { 'h','t','t','p',':','/','/','%','s',':','%','d',0 }; - WCHAR szFormat2[] = { 'h','t','t','p',':','/','/','%','s',':','%','d',0 }; + const WCHAR szHttp[] = { 'h','t','t','p',':','/','/',0 }, szSlash[] = { '/',0 } ; + const WCHAR szFormat1[] = { 'h','t','t','p',':','/','/','%','s',':','%','d',0 }; + const WCHAR szFormat2[] = { 'h','t','t','p',':','/','/','%','s',':','%','d',0 }; int len; memset( &UrlComponents, 0, sizeof UrlComponents ); @@ -607,7 +606,7 @@ LPWSTR lpszUrl = NULL; DWORD nCookieSize; HINTERNET handle; - WCHAR szUrlForm[] = {'h','t','t','p',':','/','/','%','s',0}; + const WCHAR szUrlForm[] = {'h','t','t','p',':','/','/','%','s',0}; DWORD len; TRACE("--> \n"); @@ -670,7 +669,7 @@ if (NULL == lpszVerb) { - WCHAR szGet[] = {'G','E','T',0}; + const WCHAR szGet[] = {'G','E','T',0}; lpwhr->lpszVerb = WININET_strdupW(szGet); } else if (strlenW(lpszVerb)) @@ -698,7 +697,7 @@ if (hIC->lpszAgent) { WCHAR *agent_header; - WCHAR user_agent[] = {'U','s','e','r','-','A','g','e','n','t',':',' ','%','s','\r','\n',0 }; + const WCHAR user_agent[] = {'U','s','e','r','-','A','g','e','n','t',':',' ','%','s','\r','\n',0 }; len = strlenW(hIC->lpszAgent) + strlenW(user_agent); agent_header = HeapAlloc( GetProcessHeap(), 0, len*sizeof(WCHAR) ); @@ -716,8 +715,8 @@ if (InternetGetCookieW(lpszUrl, NULL, NULL, &nCookieSize)) { int cnt = 0; - WCHAR szCookie[] = {'C','o','o','k','i','e',':',' ',0}; - WCHAR szcrlf[] = {'\r','\n',0}; + const WCHAR szCookie[] = {'C','o','o','k','i','e',':',' ',0}; + const WCHAR szcrlf[] = {'\r','\n',0}; lpszCookies = HeapAlloc(GetProcessHeap(), 0, (nCookieSize + 1 + 8)*sizeof(WCHAR)); @@ -789,9 +788,9 @@ LPHTTPHEADERW lphttpHdr = NULL; BOOL bSuccess = FALSE; LPWININETHTTPREQW lpwhr; - WCHAR szFmt[] = { '%','s',':',' ','%','s','%','s',0 }; - WCHAR szcrlf[] = { '\r','\n',0 }; - WCHAR sznul[] = { 0 }; + const WCHAR szFmt[] = { '%','s',':',' ','%','s','%','s',0 }; + const WCHAR szcrlf[] = { '\r','\n',0 }; + const WCHAR sznul[] = { 0 }; if (TRACE_ON(wininet)) { #define FE(x) { x, #x } @@ -1419,12 +1418,12 @@ do { - WCHAR szSlash[] = { '/',0 }; - WCHAR szSpace[] = { ' ',0 }; - WCHAR szHttp[] = { 'h','t','t','p',':','/','/', 0 }; - WCHAR szcrlf[] = {'\r','\n', 0}; - WCHAR sztwocrlf[] = {'\r','\n','\r','\n', 0}; - WCHAR szSetCookie[] = {'S','e','t','-','C','o','o','k','i','e',0 }; + const WCHAR szSlash[] = { '/',0 }; + const WCHAR szSpace[] = { ' ',0 }; + const WCHAR szHttp[] = { 'h','t','t','p',':','/','/', 0 }; + const WCHAR szcrlf[] = {'\r','\n', 0}; + const WCHAR sztwocrlf[] = {'\r','\n','\r','\n', 0}; + const WCHAR szSetCookie[] = {'S','e','t','-','C','o','o','k','i','e',0 }; TRACE("Going to url %s %s\n", debugstr_w(lpwhr->lpszHostName), debugstr_w(lpwhr->lpszPath)); loop_next = FALSE; @@ -1540,7 +1539,7 @@ { if (lpwhr->StdHeaders[i].wFlags & HDR_ISREQUEST) { - WCHAR szFmt[] = { '\r','\n','%','s',':',' ','%','s', 0}; + const WCHAR szFmt[] = { '\r','\n','%','s',':',' ','%','s', 0}; cnt += sprintfW(requestString + cnt, szFmt, lpwhr->StdHeaders[i].lpszField, lpwhr->StdHeaders[i].lpszValue); TRACE("Adding header %s (%s)\n", @@ -1554,7 +1553,7 @@ { if (lpwhr->pCustHeaders[i].wFlags & HDR_ISREQUEST) { - WCHAR szFmt[] = { '\r','\n','%','s',':',' ','%','s', 0}; + const WCHAR szFmt[] = { '\r','\n','%','s',':',' ','%','s', 0}; cnt += sprintfW(requestString + cnt, szFmt, lpwhr->pCustHeaders[i].lpszField, lpwhr->pCustHeaders[i].lpszValue); TRACE("Adding custom header %s (%s)\n", @@ -1565,7 +1564,7 @@ if (lpwhr->lpszHostName) { - WCHAR szFmt[] = { '%','s','%','s',0 }; + const WCHAR szFmt[] = { '%','s','%','s',0 }; cnt += sprintfW(requestString + cnt, szFmt, HTTPHOSTHEADER, lpwhr->lpszHostName); } @@ -1658,7 +1657,7 @@ { LPHTTPHEADERW setCookieHeader; int nPosStart = 0, nPosEnd = 0, len; - WCHAR szFmt[] = { 'h','t','t','p',':','/','/','%','s','/',0}; + const WCHAR szFmt[] = { 'h','t','t','p',':','/','/','%','s','/',0}; setCookieHeader = &lpwhr->pCustHeaders[CustHeaderIndex]; @@ -1678,7 +1677,7 @@ /* fixme: not case sensitive, strcasestr is gnu only */ int nDomainPosEnd = 0; int nDomainPosStart = 0, nDomainLength = 0; - WCHAR szDomain[] = {'d','o','m','a','i','n','=',0}; + const WCHAR szDomain[] = {'d','o','m','a','i','n','=',0}; LPWSTR lpszDomain = strstrW(&setCookieHeader->lpszValue[nPosEnd], szDomain); if (lpszDomain) { /* they have specified their own domain, lets use it */ @@ -1952,8 +1951,8 @@ BOOL bSuccess = FALSE; INT rc = 0; WCHAR value[MAX_FIELD_VALUE_LEN], field[MAX_FIELD_LEN]; - WCHAR szStatus[] = {'S','t','a','t','u','s',0}; - WCHAR szHttp[] = { 'H','T','T','P',0 }; + const WCHAR szStatus[] = {'S','t','a','t','u','s',0}; + const WCHAR szHttp[] = { 'H','T','T','P',0 }; char bufferA[MAX_REPLY_LEN]; TRACE("-->\n"); @@ -2087,32 +2086,32 @@ INT HTTP_GetStdHeaderIndex(LPCWSTR lpszField) { INT index = -1; - WCHAR szContentLength[] = { + const WCHAR szContentLength[] = { 'C','o','n','t','e','n','t','-','L','e','n','g','t','h',0}; - WCHAR szStatus[] = {'S','t','a','t','u','s',0}; - WCHAR szContentType[] = { + const WCHAR szStatus[] = {'S','t','a','t','u','s',0}; + const WCHAR szContentType[] = { 'C','o','n','t','e','n','t','-','T','y','p','e',0}; - WCHAR szLastModified[] = { + const WCHAR szLastModified[] = { 'L','a','s','t','-','M','o','d','i','f','i','e','d',0}; - WCHAR szLocation[] = {'L','o','c','a','t','i','o','n',0}; - WCHAR szAccept[] = {'A','c','c','e','p','t',0}; - WCHAR szReferer[] = { 'R','e','f','e','r','e','r',0}; - WCHAR szContentTrans[] = { 'C','o','n','t','e','n','t','-', + const WCHAR szLocation[] = {'L','o','c','a','t','i','o','n',0}; + const WCHAR szAccept[] = {'A','c','c','e','p','t',0}; + const WCHAR szReferer[] = { 'R','e','f','e','r','e','r',0}; + const WCHAR szContentTrans[] = { 'C','o','n','t','e','n','t','-', 'T','r','a','n','s','f','e','r','-','E','n','c','o','d','i','n','g',0}; - WCHAR szDate[] = { 'D','a','t','e',0}; - WCHAR szServer[] = { 'S','e','r','v','e','r',0}; - WCHAR szConnection[] = { 'C','o','n','n','e','c','t','i','o','n',0}; - WCHAR szETag[] = { 'E','T','a','g',0}; - WCHAR szAcceptRanges[] = { + const WCHAR szDate[] = { 'D','a','t','e',0}; + const WCHAR szServer[] = { 'S','e','r','v','e','r',0}; + const WCHAR szConnection[] = { 'C','o','n','n','e','c','t','i','o','n',0}; + const WCHAR szETag[] = { 'E','T','a','g',0}; + const WCHAR szAcceptRanges[] = { 'A','c','c','e','p','t','-','R','a','n','g','e','s',0 }; - WCHAR szExpires[] = { 'E','x','p','i','r','e','s',0 }; - WCHAR szMimeVersion[] = { + const WCHAR szExpires[] = { 'E','x','p','i','r','e','s',0 }; + const WCHAR szMimeVersion[] = { 'M','i','m','e','-','V','e','r','s','i','o','n', 0}; - WCHAR szPragma[] = { 'P','r','a','g','m','a', 0}; - WCHAR szCacheControl[] = { + const WCHAR szPragma[] = { 'P','r','a','g','m','a', 0}; + const WCHAR szCacheControl[] = { 'C','a','c','h','e','-','C','o','n','t','r','o','l',0}; - WCHAR szUserAgent[] = { 'U','s','e','r','-','A','g','e','n','t',0}; - WCHAR szProxyAuth[] = { + const WCHAR szUserAgent[] = { 'U','s','e','r','-','A','g','e','n','t',0}; + const WCHAR szProxyAuth[] = { 'P','r','o','x','y','-', 'A','u','t','h','e','n','t','i','c','a','t','e', 0};

4 7

Re: some more dbghelp fixes & enhancements
by Dimitrie O. Paun April 17, 2004

April 17, 2004

On April 17, 2004 3:02 pm, Eric Pouech wrote: > change log says it all > A+ > @@ -29,22 +29,21 @@ > > /* TODO The patch is b0rken. -- Dimi.

2 1

Visual FoxPro
by Cecil Champenois April 17, 2004

April 17, 2004

I have heard that there is a lot of interest in moving the WINE for Visual FoxPro forward. About a year ago, a major FoxPro developer by the name of Whil Hentzen was telling our User Group www.ocfox.org about the Visual FoxPro work you are doing with WINE. He also demonstrated Visual FoxPro working on a Linux laptop. We were all very impressed by this and anted to jump right into development but were told that the conversion or whatever you want to call it was at least 1 to 1.5 years away. This discouraged us a bit, so what we'd like to know is this: Where does the WINE work in relation to VFP 6, 76 and 8 stand now? We would like to begin some development in Linux, but need to know where the WINE work stands at this time and what the future looks like. Cecil

3 2

RE: Proposed solution for COMCTL32 subclassing problems
by Robert Shearman April 17, 2004

April 17, 2004

Filip Navara wrote: > > Hi, > > here is my proposed solution for crashes in COMCTL32 observed in > specific situations while calling the window procedure subclassing > functions. If there were removed all subclassed window procedures during > the execution of SubclassWndProc there happened a crash, because the > structure that holds the subclassing stack was freed in DefSubclassProc. Yes, that was my fault. Sorry. > Also if there were some window procedure subclassings removed during the > execution of SubclassWndProc, the stack position counting could have > gone wrong. My solution is to let the subclassing stack grow the > opposite way, so that item 0 is the most nested one. I also removed the > stacknew handling which I strongly believe is redundant now. It looks much simpler now, but I don't understand why reversing the direction of the stack should make any difference. I don't think you submitted the required patch to comctl32.h. Rob

2 4

Re: [DXDIAG2] First Implementation
by Alexandre Julliard April 17, 2004

April 17, 2004

Raphael <fenix(a)club-internet.fr> writes: > + * @TODO: fix wine headers dependencies > + */ > +#if !defined( _MSC_VER ) || defined( __WINE__ ) > +# include "windef.h" > +# include "winbase.h" > +# include "wingdi.h" > +# include "winuser.h" > +/** GetProp is a define by user.h */ > +# undef GetProp > +/* VARIANT */ > +# include <oleauto.h> > +#endif > +/** > + * End Hack > + */ This is because on Windows many headers include windows.h and we don't want to do that in Wine. There's no reason to add #ifdefs though, it should build just fine on Windows even with the extra includes. > +#undef EXPORT_STDAPI > +#if defined( _MSC_VER ) > +# define EXPORT_STDAPI __declspec(dllexport) extern HRESULT > +#else > +# define EXPORT_STDAPI extern HRESULT > +#endif We don't need that sort of thing in Wine dlls, and it confuses winapi_check, so please don't do it. Also you forgot to include the spec file in your patch... -- Alexandre Julliard julliard(a)winehq.org

2 3

Re: [dx104] better hardware vertex shader activation
by Alexandre Julliard April 17, 2004

April 17, 2004

Raphael <fenix(a)club-internet.fr> writes: > Changelog: > - Use a config entry (instead of the registrry) to control the HW vertex > shader activation Actually the registry is probably a better idea, this way we don't need to move it again once we get rid of the config file. I'd suggest Sofware\Wine\d3d instead of Sofware\Wine\Wine\d3d though, this is more likely to be the correct place once the config reorganization is done. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: abnt2 keyboard patch
by Rafael Ávila de Espíndola April 17, 2004

April 17, 2004

> That's why I said that it's an X11 bug and we can do nothing about it. > > Please resubmit the patch without that part. Attached. I will contact xfree and try to fix the rest of the problem. Thanks. > Dmitry. Rafael

2 1

Re: Visual FoxPro
by Dan Kegel April 17, 2004

April 17, 2004

Cecil Champenois wrote: > I have heard that there is a lot of interest in moving the WINE for > Visual FoxPro forward. About a year ago, a major FoxPro developer by the > name of Whil Hentzen was telling our User Group www.ocfox.org about the > Visual FoxPro work you are doing with WINE. He also demonstrated Visual > FoxPro working on a Linux laptop. We were all very impressed by this and > anted to jump right into development but were told that the conversion > or whatever you want to call it was at least 1 to 1.5 years away. This > discouraged us a bit, so what we'd like to know is this: Where does the > WINE work in relation to VFP 6, 76 and 8 stand now? > > We would like to begin some development in Linux, but need to know where > the WINE work stands at this time and what the future looks like. Hi Cecil! I went to the VFP UG in Los Angeles that week just to meet Whil, he's a nice guy. I think what he was probably talking about when he said "1 to 1.5 years away" was that Wine had some rough edges. It still does, but they're steadily being cleaned up, and Wine is getting smoother and smoother every month. You should probably give it a try. Build it from fresh sources (it's not too hard), set up a small test app in VFP, and give it a try on Wine. If you run into problems, let us know. - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

1 0

Compilation problems in glu32/glu.c
by Luke Stras April 16, 2004

April 16, 2004

I've recently downloaded a fresh copy out of CVS, and get the following errors while compiling: make[2]: Entering directory `/home/stras/src/wine/wine/dlls/glu32' ../../tools/winegcc/winegcc -B../../tools/winebuild -shared ./glu32.spec glu.o glu32.dll.dbg.o -o glu32.dll.so -L../../dlls -lkernel32 -lntdll -L../../libs/wine -lwine -L/usr/X11R6/lib -lSM -lICE -lXxf86dga -lXxf86vm -lXv -lXext -lX11 -lGL -lGLU -L../../libs/port -lwine_port glu.o(.text+0x765): In function `wine_gluTessBeginPolygon': /home/stras/src/wine/wine/dlls/glu32/glu.c:437: undefined reference to `gluTessBeginPolygon' glu.o(.text+0x786): In function `wine_gluTessEndPolygon': /home/stras/src/wine/wine/dlls/glu32/glu.c:447: undefined reference to `gluTessEndPolygon' glu.o(.text+0x9b2): In function `wine_gluTessBeginContour': /home/stras/src/wine/wine/dlls/glu32/glu.c:577: undefined reference to `gluTessBeginContour' glu.o(.text+0x9d6): In function `wine_gluTessEndContour': /home/stras/src/wine/wine/dlls/glu32/glu.c:587: undefined reference to `gluTessEndContour' glu.o(.text+0xa2b): In function `wine_gluTessProperty': /home/stras/src/wine/wine/dlls/glu32/glu.c:608: undefined reference to `gluTessProperty' glu.o(.text+0xa60): In function `wine_gluTessNormal': /home/stras/src/wine/wine/dlls/glu32/glu.c:618: undefined reference to `gluTessNormal' collect2: ld returned 1 exit status winegcc: gcc failed. Any ideas? Thanks. -- Luke Stras <stras(a)utias.toronto.edu> "The meek can have the Earth; the rest of us have other plans" --Henry Spencer -- Scanned by ClamAv - http://www.clamav.net

1 0

Re: compiling wine on Darwin/MacOS X 8/10
by Alexandre Julliard April 16, 2004

April 16, 2004

emmanuel maillard <e.rsz(a)libertysurf.fr> writes: > Changelog > Fix a lstat problem on darwin I don't think you want to remove the check. What error does lstat return in that case? -- Alexandre Julliard julliard(a)winehq.org

2 2

COMCTL32 Subclassing
by Filip Navara April 16, 2004

April 16, 2004

Hi, I have found a bug in Wine's comctl32.dll that causes crashes with subclassed windows. There can happen a situation when DefSubclassProc ends with all subclassings removed, in this case it cleans up the subclassing stack stored in a window property and frees it's memory. If DefSubclassProc was called from SubclassWndProc, there will happen crash because the subclassing stack is no longer valid. Anyone has a solution for this? - Filip --- lib/comctl32/commctrl.c 15 Apr 2004 08:29:56 -0000 1.9 +++ lib/comctl32/commctrl.c 16 Apr 2004 23:32:26 -0000 @@ -1284,9 +1284,7 @@ } stackpos = stack->stackpos; stack->stackpos = stack->stacknum; - ret = DefSubclassProc(hWnd,uMsg,wParam,lParam); - stack->stackpos = stackpos; - return ret; + return DefSubclassProc(hWnd,uMsg,wParam,lParam); }

1 0

Problem with winealsa
by Francois Gouget April 16, 2004

April 16, 2004

I decided to run the winmm test in interactive mode with wineoss and winealsa to see if everything is ok. So first, to do that just do: cd dlls/winmm/tests WINETEST_INTERACTIVE=1 ../../../tools/runtest -P wine -M winmm.dll -T ../../.. -p winmm_test.exe.so wave.c That will most likely run it with wineoss as the sound driver, then edit ~/.wine/config so it looks like this: [WinMM] ; Uncomment the "Drivers" line matching your sound setting. ;"Drivers" = "wineoss.drv" ; default for most common configurations ;"Drivers" = "winearts.drv" ; for KDE "Drivers" = "winealsa.drv" ; for ALSA users ;"Drivers" = "winejack.drv" ; for Jack sound server ;"Drivers" = "winenas.drv" ; for NAS sound system ;"Drivers" = "wineaudioio.drv" ; for Solaris machines ;"Drivers" = "" ; to disable sound And you should get exactly the same result but with the winealsa driver. Actually, it may play a couple extra or a couple less tones depending on what Alsa's OSS emulation and Alsa-native report as their supported sound formats. Anyway, each tone should play for 1 second. This is the case when using wineoss but when using winealsa they play for only 0.5 second! Does anyone have any insight into this? While I was at it I also tested winearts and it played the tones at a rate of 1 per second as expected. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ In theory, theory and practice are the same, but in practice they're different.

1 0

winapi_check broken
by Stefan Leichter April 16, 2004

April 16, 2004

Hello, winapi_check --cross-call-unicode-ascii does not longer work for me. Looks like it loops when searching for the spec-files. Does anyone else noticed this problem? Normaly a run took less than 10 minutes for me. Today i stopped it after two hours. There was only one output line: .: searching for /\.spec$/ My system is SuSE 9.0 Bye Stefan

1 0

Re: controls/menu.c
by Dimitrie O. Paun April 16, 2004

April 16, 2004

On Fri, 16 Apr 2004, Ge van Geldorp wrote: > This fixes the black text on white background in AbiWord menus > > Changelog: > Ge van Geldorp <gvg(a)reactos.com> > Setup menu drawing colors before calling owner draw routine Good catch Ge, this was really annoying, and maybe the most upsetting visual glitch. We still have a refresh/invalidate off-by-one bug somewhere, the dialogs leave a black margin (to the right and at the bottom). Did you see that? -- Dimi.

2 1

Re: dlls/richedit/reader.c
by Phil Krylov April 16, 2004

April 16, 2004

Hello Ge, Friday, April 16, 2004, 3:56:41 PM, you wrote: GvG> Changelog: GvG> Ge van Geldorp <gvg(a)reactos.com> GvG> Work around problem in NSIS installers which can't handle 1 char GvG> at a time reading of RTF text. To make it fully compatible with MS richedit, the buffer size should be not 10 chars, but 4096. Although it's not documented, all versions of MS Richedit always use this value. -- Best regards, Phil mailto:phil(a)newstar.rinet.ru

2 1

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 16, 2004

April 16, 2004

Could you take a look at this fix? I would think I have resolved the other issue with SetDefButton as well. Tests with msg ran fine. Regards Santosh Siddheshwar <<patch.diff>> > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Friday, April 16, 2004 1:35 PM > To: Santosh Siddheshwar; Alexandre Julliard > Cc: Robert Shearman; wine-devel(a)winehq.org > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > This test fails with WINE even after I create the DIALOGINFO structure > the > > first time it enters DefDlgProc. Not sure why this test should work on > > Windows. > > Since I wrote it under Windows it apparently works there. > > > Basically at the stage where you are setting DM_SETDEFID there are > > no child windows on the dialog. > > Yes, I was also confused when I saw this for the first time. > > > Then what does DM_SETDEFID and DM_GETDEFID do when there are no child > > controls on the dialog? > > Looks like it should store id in the dialog structure, and only then try > to change the look of a control. > > > My guess is implementation of DEFDLG_SetDefButton > > may need to change. > > Yes, but that's another bug, not related to your patch. Although it would > be > great if you could fix that as well, don't bother if you have no time, > I'll > fix it later, this should be a very simple patch. > > -- > Dmitry. ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

[Peter Riocreux] Re: dlls/kernel/console.c::SetConsoleCtrlHandler() cleanup/fixup
by Peter Riocreux April 16, 2004

April 16, 2004

Sorry, mistyped the wine-patches address before ...

1 0

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 16, 2004

April 16, 2004

This test fails with WINE even after I create the DIALOGINFO structure the first time it enters DefDlgProc. Not sure why this test should work on Windows. Basically at the stage where you are setting DM_SETDEFID there are no child windows on the dialog. Then what does DM_SETDEFID and DM_GETDEFID do when there are no child controls on the dialog? My guess is implementation of DEFDLG_SetDefButton may need to change. Regards Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Thursday, April 15, 2004 5:48 PM > To: Santosh Siddheshwar; Alexandre Julliard > Cc: Robert Shearman; wine-patches(a)winehq.com > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > OK.One place where this could be done is WM_NCCREATE which comes before > > WM_CREATE. Other place (possibly better) where it might be possible to > do > > this is in CBTProc hook procedure. Not sure about the second option > though. > > Any thoughts? > > Attached test case shows that dialog info is initialized on the very first > DefDlgProc invocation. > > So, DefDlgProc16/A/W as the very first thing should call DIALOG_get_info > and > if it returns NULL initialize dialog structures. > > Alexandre, please apply the test as soon as the dialog initialization is > fixed. > > Changelog: > Dmitry Timoshkov <dmitry(a)codeweavers.com> > Add a message test for custom dialog initialization. > > -- > Dmitry. << File: msg.c.diff >> ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

Re: [Darwine] compiling wine on Darwin/MacOS X 5/10
by Pierre d'Herbemont April 16, 2004

April 16, 2004

Emmanuel, Index: libs/wine/loader.c =================================================================== RCS file: /home/wine/wine/libs/wine/loader.c,v retrieving revision 1.16 diff -u -r1.16 loader.c --- libs/wine/loader.c 2 Jan 2004 21:08:05 -0000 1.16 +++ libs/wine/loader.c 15 Apr 2004 18:17:07 -0000 @@ -182,6 +182,9 @@ +#ifdef __APPLE__ + if ((test_only == 0x2) && (ret = wine_dlopen(path, RTLD_NOW | RTLD_GLOBAL, error, errorsize ))) break; +#endif +#ifdef __APPLE__ + if (!(ntdll = dlopen_dll( "ntdll.dll", error, error_size, 0x2, &file_exists ))) return; +#else if (!(ntdll = dlopen_dll( "ntdll.dll", error, error_size, 0, &file_exists ))) return; +#endif These tests are no longer useful on Mac OS X. It was when ntdll was built as a dylib. However the check for environ is useful. Pierre On 15 avr. 04, at 23:36, emmanuel maillard wrote: > Hi, > here come some patches to build wine on Darwin/OS X > > Changelog > Build dylib on Darwin > > emmanuel > > <libs.diff>

1 0

Re: Wine fixes
by Dimitrie O. Paun April 15, 2004

April 15, 2004

On April 16, 2004 12:16 am, Dimitrie O. Paun wrote: > IMO should just nuke some of these useless includes. > How can we test that they are really needed? Oops, sorry folks, this message was meant for the AbiWord guys! :) -- Dimi.

1 0

Re: abnt2 keyboard patch
by Rafael Ávila de Espíndola April 15, 2004

April 15, 2004

> I still see problems with your patch. > > 1. You are using 'keysym == 0x2C' for comma, but what happens if NumLock > is on, and you press comma on the main keyboard, not on the keypad? > Shouldn't X11 send XK_KP_xxx event in that case (something like XK_KP_Decimal)? The file /usr/include/X11/keysymdef.h has the line #define XK_KP_Separator 0xFFAC /* separator, often comma */ This appears to be appropriate. But X11 is not sending it in debian (sid) or fedora. The output from xev when, with the numlock initially off, I press comma, kp_comma, numlock, comma and kp_comma is: KeyPress event, serial 22, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69979707, (124,72), root:(364,205), state 0x0, keycode 59 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," KeyRelease event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69979799, (124,72), root:(364,205), state 0x0, keycode 59 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," KeyPress event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69980420, (124,72), root:(364,205), state 0x0, keycode 91 (keysym 0xff9f, KP_Delete), same_screen YES, XLookupString gives 0 bytes: "" KeyRelease event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69980521, (124,72), root:(364,205), state 0x0, keycode 91 (keysym 0xff9f, KP_Delete), same_screen YES, XLookupString gives 0 bytes: "" KeyPress event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69980955, (124,72), root:(364,205), state 0x0, keycode 77 (keysym 0xff7f, Num_Lock), same_screen YES, XLookupString gives 0 bytes: "" KeyRelease event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69981057, (124,72), root:(364,205), state 0x10, keycode 77 (keysym 0xff7f, Num_Lock), same_screen YES, XLookupString gives 0 bytes: "" KeyPress event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69981755, (124,72), root:(364,205), state 0x10, keycode 59 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," KeyRelease event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69981850, (124,72), root:(364,205), state 0x10, keycode 59 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," KeyPress event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69982436, (124,72), root:(364,205), state 0x10, keycode 91 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," KeyRelease event, serial 27, synthetic NO, window 0x2c00001, root 0x3f, subw 0x0, time 69982535, (124,72), root:(364,205), state 0x10, keycode 91 (keysym 0x2c, comma), same_screen YES, XLookupString gives 1 bytes: "," > 2. Please generate unified diffs, 'using diff -u' for patches. Sorry. I am really embarrassed. An unified diff is attached. > > -- > Dmitry. Rafael

2 1

Re: dlls/kernel/console.c::SetConsoleCtrlHandler() cleanup/fixup
by Dimitrie O. Paun April 15, 2004

April 15, 2004

On April 15, 2004 11:18 am, Peter Riocreux wrote: > * This doesn't yet matter, since these handlers are not yet called...! > + * > + * Peter Riocreux (par(a)silistix.com) 2004-04-15 > + * Added error checking and now uses SetLastError() > */ One more thing: please don't add this sort of thing, it belongs in the CVS log, not in the source code. -- Dimi.

1 0

Re: [D3D8] Add support for vertex shader (using ARB_vertex_program extension)
by Christian Costa April 15, 2004

April 15, 2004

Maxime Bellengé wrote: >Hi Christian > >Are you looking for someone to test the patch on an NVIDIA card ? >If you tell me how to test or the game to run, I can give you a >feedback. > >a+ > >Max > > > Hi Maxime, You can try the vertexblend, vertexshader & dolphinvs demos from the DX8 SDK. You can switch between HAL to REF devices to see the difference. Thanks for testing! :-) Bye, Christian

1 0

RE: More serial access issues
by Geoffrey Hausheer April 15, 2004

April 15, 2004

Well, after tracing the events through wineserver, I see 'serial_queue_event' gets called, serial_get_poll_events returns 'POLLOUT', async_notify gets called (but async->thread->wait is 0, so it doesn't do anything besides setting the status to STATUS_ALERTED. The secod serial_get_poll_events returns 0 (due to the status no longer being STATUS_PENDING), but this doesn't really matter as the 'register_async' call doesn't create a reply. Back in my application, __register_async does not reset u.Status (since 'ret' is 0), which means that 'finish_async' never gets called, which appears to be the only way that FILE_AsyncWriteService can be called (which actually does the write) ,and no data is written to the serial port. It all seems to operate as the code dictates, so I'm not sure what the right fix is. It would seem to me that something is not working right in __register_async. Perhaps finish_async should not be gated by 'u.Status'? Or u.Status should unconditionally be set equal to 'ret'? The second seems more likely to me, as that is the only way the WriteFile could return TRUE (in fact I see no way that WriteFile could return true in today's code, regardless of what the server does (since u.Status can only be changed from STATUS_PENDING if ret != 0 (and STATUS_SUCCESS == 0), so u.Status can never equal STATUS_SUCCESS. I am having a lot of difficulty understanding exactly what the server call is doing (I think it is checking that it is clear to write to the Serial port, though it seems to be awfully complex if that is all). Also, I don't know what the conditions are under which finish_async should be called. I changed the code to unconditionally set u.Status = ret, and WriteFile is no longer returning an error, but this didn't help (I'm never seeing FILE_AsyncWriteService get called, because call completion is disabled). So it looks like there are several points that are preventing FILE_AsyncWriteService from being called, and maybe I'm barking up the wrong tree. I can see a few possibilities: (1) some other routine is supposed to do the write, and I'm missing it, or (2) the server is supposed to do the write (perhaps via asyncy_notify?), in which case I don't see any place where the write buffer is passed to the server, or (c) the server is supposed to tell the client to do the write (again by async_notify?), in which case, I'm not seeing how that works. This code seems to run around in circles to me. .Geoff -----Original Message----- From: Geoffrey Hausheer Sent: Wednesday, April 14, 2004 2:05 PM Subject: More serial access issues Well, I broke down and pulled out the oscope and rs232 breakout box to try to figure out why my app isn't able to communicate with the device. turns out that Rein's patch is correctly deasserting DSR, however, I am not getting any data over the Tx pin. And after I checked the output of WriteFile, it is failing. GetLastError doesn't return anything useful (a value of 997 which FormatMessage() can't do anything with). But some debugging led me to determine that NtWriteFile is returning 0x103 which is STATUS_PENDING. After that I'm somewhat lost as to what is going on. Obvioulsy, 'register_new_async' is returning this status, which apparently happens when wine_server_call returns 0 or STATUS_PENDING (not sure if the second is even possible). But about here I am getting completely confused as to what we are trying to accomplish. The file is opened via: CreateFile("COM1", 0, NULL, OPEN_EXISTING, 0, NULL) so we're not in overlapped mode. SetCommTimeout is called with the COMMTIMEOUTS structure containing all 0s. All flow control is disabled using SetCommState. Before a write, we always call 'PurgeComm' and ClearCommError. In Windows, this combination results in a write always occurring when WriteFile is called, regardless of whether there is anything connected to the serial port. Wine appears to be waiting for something, and I'm not sure what. If anyone has any ideas how I might proceed, i'd be glad to hear them. Thanks, .Geoff

1 0

Compiling MFC with latest CVS checkout.
by Manjunath Sripadarao April 15, 2004

April 15, 2004

I am trying to compile MFC with latest CVS checout, I copied the source from the windows MFC dir and also the include dir (both for MFC and the VC98 include dir). I ran winemaker --interactive . and specified the -I paths. I got a couple of makefiles. Now if I do a make I get this error. [manju(a)manju MSVS]$ make make[1]: Entering directory `/home/manju/MSVS/mfc/src' wineg++ -mwindows -mno-cygwin -o mfcdb.so mfcdb.res mfcdb.dbg.o -lodbc32 -lole32 -loleaut32 -lwinspool -luuid mfcdb.exe-ueIUVp.spec.c:451: error: invalid conversion from `void (*)()' to ` void*' winegcc: gcc failed. make[1]: *** [mfcdb.so] Error 2 make[1]: Leaving directory `/home/manju/MSVS/mfc/src' make: *** [mfc/src] Error 2 Any ideas/tips. Am I even following the correct procedure ? There doesn't seem to be any document detailing this process. Any tips/instructions/suggestions would be greatly appreciated. I am doing this on a fedora core 1 (with prelink fixed and stuff, wine runs well), with a 2 Ghz P4 machine. If any other info is needed please feel free. Thanks in advance, Manju

3 3

wine/tools wineinstall
by Saulius Krasuckas April 15, 2004

April 15, 2004

> Log message: > Only create the device symlinks the first time around. <skipped /> > + # Check if dosdevices exists and create it if necessary > + if [ ! -d ~/.wine/dosdevices ] > + then > + mkdir ~/.wine/dosdevices > + ln -s /mnt/fd0 ~/.wine/dosdevices/a: > + ln -s $CROOT ~/.wine/dosdevices/c: Alexandre, what about the case when ~/.wine/dosdevices exists but is empty? PS eghm, should it be bad for wineinstall and winecfg to share common code for creating links in dosdevices/ ?

2 1

help fixing up SetConsoleCtrlHandler()
by Peter Riocreux April 15, 2004

April 15, 2004

Having seen a lot of messages like: ----------------- fixme:console:SetConsoleCtrlHandler (0x41a8af,1) - no error checking or testing yet ----------------- lately, I thought I would actually see if I could fix it up and remove the FIXME. It is in dlls/kernel/console.c, and the MSDN reference is at <http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/ba…>. I think it is obvious enough that even I, a Wine newbie could do it, except for a couple of things that, as a newbie, are not obvious to me. Please point me at docs if this is covered somewhere - it is hard getting into a new project .... 1) If I am using SetLastError, it is not obvious to me what error values I should be using there. I assume there is a generic couldnt-allocate-on-heap error, but other causes of error are specific to the function, and I cannot find any documentation. Should I just use a non-zero error number and put FIXMEs in for any number I had to invent? 2) Can I assume that Rtl{Enter,Leave}CriticalSection(), HeapFree(), ERR() and WARN() cannot fail (the latter two are presumably macros). 3) The docs say that it behaves eeeeeeever so slightly differently in ME/98/95 than in others. In this case it is simply that one of the arguments may not be NULL. Should I, either in general, or in this case in particular, detect which Windows version is being used and behave appropriately, or implement the more flexible approach and assume that people writing for the more restricted OS will not have used that case. If I shoudl detect, how does one go about this? Peter

3 6

Re: winegcc: use CC to compile the .spec.c file
by Alexandre Julliard April 15, 2004

April 15, 2004

"Dimitrie O. Paun" <dimi(a)intelliware.ca> writes: > ChangeLog > Dimitrie O. Paun <dpaun(a)rogers.com> > Always use CC to compile the .spec.c file. Doesn't compile_to_object() do this already? -- Alexandre Julliard julliard(a)winehq.org

3 2

Re: Who's who - Blowing my own horn.
by Kevin Koltzau April 15, 2004

April 15, 2004

On Wednesday 14 April 2004 10:48 pm, Tony Lambregts wrote: > Blowing my own horn. ... > + <td width="300">Location: St. Albert, Alberta, Canada </td> > + <td>IRC Nick: tindalos</td></tr> Thats my IRC nick :)

1 0

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 15, 2004

April 15, 2004

Actually there is WM_GETMINMAXINFO which comes before WM_NCCREATE but I dont think we should put the DIALOGINFO initialization logic there. WM_NCCREATE should be the better option. Regards Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Thursday, April 15, 2004 3:56 PM > To: Santosh Siddheshwar; Alexandre Julliard > Cc: Robert Shearman; wine-devel(a)winehq.org > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > OK.One place where this could be done is WM_NCCREATE which comes before > > WM_CREATE. Other place (possibly better) where it might be possible to > do > > this is in CBTProc hook procedure. Not sure about the second option > though. > > Any thoughts? > > I'll add a test case for it in our message test suite. > > -- > Dmitry. ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

1 0

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 15, 2004

April 15, 2004

OK.One place where this could be done is WM_NCCREATE which comes before WM_CREATE. Other place (possibly better) where it might be possible to do this is in CBTProc hook procedure. Not sure about the second option though. Any thoughts? Regards Santosh Siddheshwar > -----Original Message----- > From: Alexandre Julliard [SMTP:julliard(a)winehq.org] > Sent: Thursday, April 15, 2004 9:43 AM > To: Santosh Siddheshwar > Cc: Dmitry Timoshkov; Robert Shearman; wine-devel(a)winehq.org > Subject: Re: Handling dialog messages for non-template based dialogs > > Santosh Siddheshwar <santosh.s(a)sonata-software.com> writes: > > > But dont you think that the structure is not really required till the > > window has been created successfully? > > Well, I don't think so, but the only way to know for sure is to write > a test program under Windows to determine the exact behavior. > > -- > Alexandre Julliard > julliard(a)winehq.org ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

Problem with bad generation of low-level keyboard events
by Lionel Ulmer April 15, 2004

April 15, 2004

Hi all, While debugging GeneRally (a very nice game), I got some problems with DInput receiving spurious events (basically, as if key '0' was pressed). This is traced to the following problem : In windows/input.c, the function 'queue_kbd_event' stores the KBDLLHOOKSTRUCT structure on the stack and then calls 'HOOK_CallHooks' to send it. In my case, thread 000a receives the event and sends it to thread 0009. What happens is the following (the 'Hook:' lines were added by me) : 000a:trace:hook:HOOK_CallHooks calling hook in thread 0009 WH_KEYBOARD_LL code 0 wp 101 lp 4156ad6c 000a:trace:hook:HOOK_CallHooks Hook: 0x4156ad6c - 00000026 00000048 000000c1 0000b578 00000000 000a:trace:msg:send_inter_thread_message hwnd (nil) msg 80000007 ("#7") wp 101 lp 4156ad6c 000a: send_message( id=0009, type=1, flags=1, win=(nil), msg=80000007, wparam=00000101, lparam=4156ad6c, x=0, y=0, time=0000b587, info=00000000, timeout=2000, callback=(nil), data={} ) 000a: send_message() = 0 000a: set_queue_mask( wake_mask=00008040, changed_mask=00008040, skip_wait=1 ) 000a: set_queue_mask() = 0 { wake_bits=00000000, changed_bits=00000000 } 000a: select( flags=4, cookie=0x4156a578, timeout=0, handles={0x50} ) 000a: select() = PENDING 0009: get_handle_fd( handle=0xc0, access=80000000 ) 0009: get_handle_fd() = 0 { fd=33, type=1, flags=0 } (...) 0009:trace:ddraw:Main_DirectDrawSurface_Lock locked surface returning description : - DDSD_CAPS : DDSCAPS_OFFSCREENPLAIN DDSCAPS_VIDEOMEMORY (...) 000a: *wakeup* signaled=0 cookie=0x4156a578 000a: set_queue_mask( wake_mask=00008040, changed_mask=00008040, skip_wait=1 ) 000a: set_queue_mask() = 0 { wake_bits=00008000, changed_bits=00008000 } 000a: get_message_reply( cancel=1 ) 000a: get_message_reply() = TIMEOUT { result=00000000, data={} } 000a:trace:msg:retrieve_reply hwnd (nil) msg 80000007 ("#7") wp 101 lp 4156ad6c got reply 40218182 (err=258) 000a: finish_hook_chain( id=13 ) 000a: finish_hook_chain() = 0 000a:trace:key:queue_kbd_event Hook: 1) 0x4156ad6c - 00000026 00000048 000000c1 0000b578 00000000 000a: send_message( id=000a, type=7, flags=0, win=(nil), msg=00000101, wparam=00000026, lparam=c1480001, x=691, y=584, time=0000b578, info=00000000, timeout=-1, callback=(nil), data={} ) 000a: send_message() = 0 000a:trace:key:queue_kbd_event Hook: 2) 0x4156ad6c - 00000026 00000048 000000c1 0000b578 00000000 000a:trace:event:EVENT_ProcessEvent returns. (...) 0009: select( flags=12, cookie=0x4069f890, timeout=1081894976.490610 (-0.000077), handles={0x3c} ) 0009: select() = TIMEOUT 0009: get_message( flags=1, get_win=(nil), get_first=00000000, get_last=ffffffff ) 0009: get_message() = 0 { type=1, win=(nil), msg=80000007, wparam=00000101, lparam=4156ad6c, x=0, y=0, time=0000b587, info=00000000, total=0, data={} } 0009:trace:msg:MSG_peek_message got type 1 msg 80000007 ("#7") hwnd (nil) wp 101 lp 4156ad6c 0009: start_hook_chain( id=13 ) 0009: start_hook_chain() = 0 { handle=0x10022, pid=0000, tid=0000, proc=0x40cd4c70, unicode=1, module=L"" } 0009:trace:hook:HOOK_CallHooks calling hook 0x40cd4c70 WH_KEYBOARD_LL code 0 wp 101 lp 4156ad6c module L"" 0009:trace:hook:HOOK_CallHooks Hook: 0x4156ad6c - 4156ad88 00000000 00000040 4156ad8c 4156ad88 A summary of my understanding of the problem : - in 'queue_kbd_event', the structure is put on the stack on the expectation that the 'HOOK_CallHooks' won't return with the message unsent. - as seen, because the thread which receives the message does not seem to handle any messages during quite a long time (it's actually loading the track between two races), this message sending time-outs on the sender's side. - so the function 'queue_kbd_event' ends and the stack gets corrupted. - which can be seen in my last 'Hook:' dump. If it's normal for the sending to time-out, we should replace this code and add some HeapAlloc on the sender side and some HeapFree on the receiver side each time such a structure is sent ... Which will be a bit problematic for performance, but well, I do not see any other way to correct this. Lionel -- Lionel Ulmer - http://www.bbrox.org/

2 2

exec-shield patch take 5
by Mike McCormack April 15, 2004

April 15, 2004

Hi, Here's the latest version of a patch I've written to allow wine to run without disabling exec-shield or prelinking. Mike Changes in version 5: * reserve the shared heap area as well as the PE load area * allow multiple reserve areas in libwine Changes in version 4: * set the WINEPEEXEPATH environment variable correctly Changes in version 3: * reserve only the area required by the PE EXE if launched from CreateProcess (else just a default slab) * pass the reserved area to wine, so wine knows about it * unmap the preloader after it is no longer used Changes in verison 2: * replace the shared object loader instead of using a bootstrap exe Version 1: * create a bootstrap exe that reserves memory before loading wine-pthread or wine-kthread Index: dlls/kernel/process.c =================================================================== RCS file: /home/wine/wine/dlls/kernel/process.c,v retrieving revision 1.52 diff -u -r1.52 process.c --- dlls/kernel/process.c 14 Apr 2004 19:40:54 -0000 1.52 +++ dlls/kernel/process.c 15 Apr 2004 07:46:25 -0000 @@ -917,8 +917,6 @@ } found: - wine_free_pe_load_area(); /* the main binary is loaded, we don't need this anymore */ - /* build command line */ set_library_wargv( __wine_main_argv ); if (!build_command_line( __wine_main_wargv )) goto error; @@ -1060,7 +1058,8 @@ * * Build the environment of a new child process. */ -static char **build_envp( const WCHAR *envW, const WCHAR *extra_envW ) +static char **build_envp( const WCHAR *envW, const WCHAR *extra_envW, + const char *workdir, const char *unixpath ) { const WCHAR *p; char **envp; @@ -1083,7 +1082,7 @@ if (!(env = malloc( length ))) return NULL; WideCharToMultiByte( CP_UNIXCP, 0, envW, p - envW, env, length, NULL, NULL ); - count += 4; + count += 6; if ((envp = malloc( count * sizeof(*envp) ))) { @@ -1098,6 +1097,11 @@ if ((p = getenv("TMP"))) *envptr++ = alloc_env_string( "TMP=", p ); if ((p = getenv("HOME"))) *envptr++ = alloc_env_string( "HOME=", p ); if ((p = getenv("WINEPREFIX"))) *envptr++ = alloc_env_string( "WINEPREFIX=", p ); + + /* record the unix working directory */ + if (workdir) *envptr++ = alloc_env_string( "WINEWORKDIR=", workdir ); + if (workdir) *envptr++ = alloc_env_string( "WINEPEEXEPATH=", unixpath ); + /* now put the Windows environment strings */ for (p = env; *p; p += strlen(p) + 1) { @@ -1106,6 +1110,8 @@ if (is_special_env_var( p )) /* prefix it with "WINE" */ *envptr++ = alloc_env_string( "WINE", p ); else if (strncmp( p, "HOME=", 5 ) && + strncmp( p, "WINEWORKDIR=", 12 ) && + strncmp( p, "WINEPEEXEPATH=", 14 ) && strncmp( p, "WINEPREFIX=", 11 )) *envptr++ = p; } *envptr = 0; @@ -1136,7 +1142,7 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmdline, 0 ); - char **envp = build_envp( env, NULL ); + char **envp = build_envp( env, NULL, NULL, NULL ); close( fd[0] ); /* Reset signals that we previously set to SIG_IGN */ @@ -1171,14 +1177,8 @@ RTL_USER_PROCESS_PARAMETERS *params; UNICODE_STRING image_str, cmdline_str, desktop, title; NTSTATUS status; - WCHAR buffer[MAX_PATH]; - - if(!GetLongPathNameW( filename, buffer, MAX_PATH )) - lstrcpynW( buffer, filename, MAX_PATH ); - if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) - lstrcpynW( buffer, filename, MAX_PATH ); - RtlInitUnicodeString( &image_str, buffer ); + RtlInitUnicodeString( &image_str, filename ); RtlInitUnicodeString( &cmdline_str, cmdline ); if (startup->lpDesktop) RtlInitUnicodeString( &desktop, startup->lpDesktop ); if (startup->lpTitle) RtlInitUnicodeString( &title, startup->lpTitle ); @@ -1230,6 +1230,7 @@ pid_t pid; int err; char dummy = 0; + WCHAR buffer[MAX_PATH]; if (!env) { @@ -1237,7 +1238,12 @@ extra_env = DRIVE_BuildEnv(); } - if (!(params = create_user_params( filename, cmd_line, startup ))) + if(!GetLongPathNameW( filename, buffer, MAX_PATH )) + lstrcpynW( buffer, filename, MAX_PATH ); + if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) + lstrcpynW( buffer, filename, MAX_PATH ); + + if (!(params = create_user_params( buffer, cmd_line, startup ))) { if (extra_env) HeapFree( GetProcessHeap(), 0, extra_env ); return FALSE; @@ -1268,7 +1274,20 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmd_line, 1 ); - char **envp = build_envp( env, extra_env ); + char **envp; + char *unixpath; + + unixpath = wine_get_unix_file_name( buffer ); + + /* preseve the current directory, the preloader may forget it */ + if( !unixdir ) + { + char *dir = HeapAlloc( GetProcessHeap(), 0, MAX_PATH ); + getcwd( dir, MAX_PATH ); + unixdir = dir; + } + envp = build_envp( env, extra_env, unixdir, unixpath ); + HeapFree( GetProcessHeap(), 0, unixpath ); close( startfd[1] ); close( execfd[0] ); @@ -1280,8 +1299,6 @@ /* Reset signals that we previously set to SIG_IGN */ signal( SIGPIPE, SIG_DFL ); signal( SIGCHLD, SIG_DFL ); - - if (unixdir) chdir(unixdir); if (argv && envp) { Index: libs/wine/config.c =================================================================== RCS file: /home/wine/wine/libs/wine/config.c,v retrieving revision 1.3 diff -u -r1.3 config.c --- libs/wine/config.c 11 Nov 2003 22:21:29 -0000 1.3 +++ libs/wine/config.c 15 Apr 2004 07:46:31 -0000 @@ -243,6 +243,23 @@ return user_name; } +/* run wine binary in it's own directory, so it can find it's preloader */ +int wine_cwd_execve( const char *name, char **argv, char **envp ) +{ + char *dir, *p; + + dir = strdup( name ); + p = strrchr( dir, '/' ); + if( p ) + { + *p++ = 0; + chdir( dir ); + name = p; + } + + return execve( name, argv, envp ); +} + /* exec a wine internal binary (either the wine loader or the wine server) */ /* if name is null, default to the name of the current binary */ void wine_exec_wine_binary( const char *name, char **argv, char **envp ) @@ -257,7 +274,7 @@ argv[0] = xmalloc( sizeof(BINDIR "/") + strlen(name) ); strcpy( argv[0], BINDIR "/" ); strcat( argv[0], name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); free( argv[0] ); /* now try the path of argv0 of the current binary */ @@ -266,7 +283,7 @@ argv[0] = xmalloc( strlen(argv0_path) + strlen(name) + 1 ); strcpy( argv[0], argv0_path ); strcat( argv[0], name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); free( argv[0] ); } @@ -283,7 +300,7 @@ memcpy( argv[0], pos, ptr - pos ); strcpy( argv[0] + (ptr - pos), "/" ); strcat( argv[0] + (ptr - pos), name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); pos = ptr; } free( argv[0] ); Index: libs/wine/port.c =================================================================== RCS file: /home/wine/wine/libs/wine/port.c,v retrieving revision 1.8 diff -u -r1.8 port.c --- libs/wine/port.c 5 Feb 2004 02:01:35 -0000 1.8 +++ libs/wine/port.c 15 Apr 2004 07:46:31 -0000 @@ -150,36 +150,63 @@ #error You must implement wine_switch_to_stack for your platform #endif +struct reserve_area +{ + char *base; + size_t size; + struct reserve_area *next; +}; -static char *pe_area; -static size_t pe_area_size; +struct reserve_area *reserve_list; /*********************************************************************** - * wine_set_pe_load_area + * wine_reserve_area * - * Define the reserved area to use for loading the main PE binary. + * Define a reserved area (for the PE binary and system heap) + * These function are not thread safe, so we assume that + * they're called before any threads are started, and all the + * reserve areas are free'd. */ -void wine_set_pe_load_area( void *base, size_t size ) +void wine_reserve_area( void *base, size_t size ) { + struct reserve_area *area; unsigned int page_mask = getpagesize() - 1; char *end = (char *)base + size; - pe_area = (char *)(((unsigned long)base + page_mask) & ~page_mask); - pe_area_size = (end - pe_area) & ~page_mask; + area = malloc( sizeof (struct reserve_area) ); + + area->base = (char *)(((unsigned long)base + page_mask) & ~page_mask); + area->size = (end - area->base) & ~page_mask; + + area->next = reserve_list; + reserve_list = area; } /*********************************************************************** - * wine_free_pe_load_area + * wine_free_area * - * Free the reserved area to use for loading the main PE binary. + * Free a reserved area. */ -void wine_free_pe_load_area(void) +int wine_free_area( char *start, size_t size ) { + struct reserve_area *area, **p; + + for( p = &reserve_list; (area = *p); p = &area->next ) + if( start >= area->base && start + size <= area->base + area->size) + break; + + if( !area ) + return 0; + + *p = area->next; + #ifdef HAVE_MMAP - if (pe_area) munmap( pe_area, pe_area_size ); + munmap( area->base, area->size ); #endif - pe_area = NULL; + + free( area ); + return 1; } @@ -296,13 +323,8 @@ flags |= MAP_PRIVATE; #endif - if (pe_area && start && - (char *)start >= pe_area && - (char *)start + size <= pe_area + pe_area_size) - { - wine_free_pe_load_area(); + if ( start && reserve_list && wine_free_area(start, size) ) flags |= MAP_FIXED; - } if (!(flags & MAP_FIXED)) { Index: libs/wine/wine.def =================================================================== RCS file: /home/wine/wine/libs/wine/wine.def,v retrieving revision 1.10 diff -u -r1.10 wine.def --- libs/wine/wine.def 5 Feb 2004 02:01:35 -0000 1.10 +++ libs/wine/wine.def 15 Apr 2004 07:46:31 -0000 @@ -64,5 +64,5 @@ wine_pthread_init_thread wine_set_fs wine_set_gs - wine_set_pe_load_area + wine_reserve_area wine_switch_to_stack Index: libs/wine/wine.map =================================================================== RCS file: /home/wine/wine/libs/wine/wine.map,v retrieving revision 1.1 diff -u -r1.1 wine.map --- libs/wine/wine.map 12 Feb 2004 22:54:00 -0000 1.1 +++ libs/wine/wine.map 15 Apr 2004 07:46:31 -0000 @@ -64,7 +64,7 @@ wine_pthread_init_thread; wine_set_fs; wine_set_gs; - wine_set_pe_load_area; + wine_reserve_area; wine_switch_to_stack; local: *; Index: loader/Makefile.in =================================================================== RCS file: /home/wine/wine/loader/Makefile.in,v retrieving revision 1.14 diff -u -r1.14 Makefile.in --- loader/Makefile.in 22 Nov 2003 00:08:26 -0000 1.14 +++ loader/Makefile.in 15 Apr 2004 07:46:31 -0000 @@ -23,20 +23,30 @@ LIBPTHREAD = @LIBPTHREAD@ LDEXECFLAGS = @LDEXECFLAGS@ -wine-glibc: glibc.o Makefile.in +WINELDEXECFLAGS = $(LDEXECFLAGS),--dynamic-linker=$(WINELDSO) + +WINELDSO=ld-winepreload.so + +$(WINELDSO): preload.o Makefile.in + $(CC) -nostartfiles -lc -static -Wl,"-Ttext=0x7ff81000" -o $@ $< + +preload.c: preload.h main.h + +wine-glibc: glibc.o Makefile.in $(WINELDSO) $(CC) -o $@ $(LDEXECFLAGS) glibc.o $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) -wine-kthread: $(KTHREAD_OBJS) Makefile.in - $(CC) -o $@ $(LDEXECFLAGS) $(KTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(EXTRALIBS) $(LDFLAGS) +wine-kthread: $(KTHREAD_OBJS) Makefile.in $(WINELDSO) + $(CC) -o $@ $(WINELDEXECFLAGS) $(KTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(EXTRALIBS) $(LDFLAGS) -wine-pthread: $(PTHREAD_OBJS) Makefile.in - $(CC) -o $@ $(LDEXECFLAGS) $(PTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) +wine-pthread: $(PTHREAD_OBJS) Makefile.in $(WINELDSO) + $(CC) -o $@ $(WINELDEXECFLAGS) $(PTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) $(MODULE): $(MAIN_BINARY) $(RM) $(MODULE) && $(LN_S) $(MAIN_BINARY) $(MODULE) install:: $(WINE_BINARIES) $(MKINSTALLDIRS) $(bindir) + $(INSTALL_PROGRAM) $(WINELDSO) $(bindir); \ for f in $(WINE_BINARIES); do \ if [ "$(MAIN_BINARY)" = "$$f" ]; \ then $(INSTALL_PROGRAM) $$f $(bindir)/$(MODULE); \ @@ -48,6 +58,6 @@ $(RM) $(WINE_BINARIES:%=$(bindir)/%) $(bindir)/$(MODULE) clean:: - $(RM) $(WINE_BINARIES) $(MODULE) + $(RM) $(WINE_BINARIES) $(MODULE) $(WINELDSO) ### Dependencies: Index: loader/glibc.c =================================================================== RCS file: /home/wine/wine/loader/glibc.c,v retrieving revision 1.3 diff -u -r1.3 glibc.c --- loader/glibc.c 20 Feb 2004 20:19:23 -0000 1.3 +++ loader/glibc.c 15 Apr 2004 07:46:31 -0000 @@ -75,6 +75,17 @@ { const char *loader = getenv( "WINELOADER" ); const char *threads = get_threading(); + char cwd[MAX_PATH], *cwd_var; + + if( !getcwd( cwd, sizeof cwd ) ) + { + fprintf( stderr, "wine: could get current directory\n" ); + exit(1); + } + cwd_var = xmalloc( sizeof("WINEWORKDIR=") + strlen(cwd) ); + strcpy( cwd_var, "WINEWORKDIR=" ); + strcat( cwd_var, cwd ); + putenv( cwd_var ); if (loader) { @@ -86,6 +97,9 @@ new_name = xmalloc( (path - loader) + strlen(threads) + 1 ); memcpy( new_name, loader, path - loader ); + new_name[ path - loader ] = 0; + if( new_name[0] ) + chdir( new_name ); strcpy( new_name + (path - loader), threads ); /* update WINELOADER with the new name */ Index: loader/main.c =================================================================== RCS file: /home/wine/wine/loader/main.c,v retrieving revision 1.90 diff -u -r1.90 main.c --- loader/main.c 20 Jan 2004 00:28:01 -0000 1.90 +++ loader/main.c 15 Apr 2004 07:46:31 -0000 @@ -20,19 +20,54 @@ #include <stdio.h> #include <stdlib.h> +#include <unistd.h> +#include <sys/mman.h> #include "wine/library.h" +#include "main.h" + +extern void wine_reserve_area( void *base, size_t size ); + +/* this variable is set by the wine-specific shared object loader */ +struct wine_preload_info *wine_preload_vars; + /********************************************************************** * main */ int main( int argc, char *argv[] ) { - char error[1024]; + char error[1024], *dir; + if( wine_preload_vars ) + { #if 0 - static char pe_load[256*1024*1024] __attribute__((aligned(4096))); - wine_set_pe_load_area( pe_load, sizeof(pe_load) ); + printf("vars(*%p) = %p -> (%p, %08x) (%p, %08x) (%p, %08x)\n", + &wine_preload_vars, wine_preload_vars, + wine_preload_vars->pe_address, + wine_preload_vars->pe_size, + wine_preload_vars->heap_address, + wine_preload_vars->heap_size, + wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); #endif + + /* let wine know which area was mapped for the PE EXE */ + wine_reserve_area( wine_preload_vars->pe_address, + wine_preload_vars->pe_size); + + /* let wine know which area was mapped for the system heap */ + wine_reserve_area( wine_preload_vars->heap_address, + wine_preload_vars->heap_size); + + /* unmap the preloader itself */ + munmap( wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); + wine_preload_vars = NULL; + } + + dir = getenv( "WINEWORKDIR" ); + if( dir ) + chdir( dir ); wine_init( argc, argv, error, sizeof(error) ); fprintf( stderr, "wine: failed to initialize: %s\n", error ); --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/main.h 2004-04-15 17:29:17.000000000 +0900 @@ -0,0 +1,10 @@ + +struct wine_preload_info +{ + void *pe_address; + size_t pe_size; + void *heap_address; + size_t heap_size; + void *preload_base; + size_t preload_size; +}; --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.h 2004-04-15 17:29:17.000000000 +0900 @@ -0,0 +1,51 @@ +/* + mini loader for ld.so - x86 Linux 2.4.x+ kernels only + + Copyright 2004 Mike McCormack for Codeweavers + + Copyright (C) 1995,96,97,98,99,2000,2001,2002 Free Software Foundation, Inc. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. + */ + +#define LDSO_NAME "/lib/ld-linux.so.2" + +/* ELF definitions */ +#define ELF_PREFERRED_ADDRESS(loader, maplength, mapstartpref) (mapstartpref) +#define ELF_FIXED_ADDRESS(loader, mapstart) ((void) 0) + +#define __ELF_NATIVE_CLASS 32 +#define MAP_BASE_ADDR(l) 0 +#define MAP_COPY MAP_PRIVATE +#define ANONFD -1 + +#define ElfW(type) _ElfW (Elf, __ELF_NATIVE_CLASS, type) +#define _ElfW(e,w,t) _ElfW_1 (e, w, _##t) +#define _ElfW_1(e,w,t) e##w##t + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#include <sys/mman.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> + +#define GL(x) x + +extern unsigned int GL(dl_pagesize); + --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.c 2004-04-15 17:29:17.000000000 +0900 @@ -0,0 +1,749 @@ +/* + mini loader for ld.so - x86 Linux 2.4.x+ kernels only + + Copyright 2004 Mike McCormack for Codeweavers + + Copyright (C) 1995,96,97,98,99,2000,2001,2002 Free Software Foundation, Inc. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. + */ + +/* + * Design notes + * + * The goal of this program is to be a workaround for exec-shield, as used + * by the Linux kernel distributed with Fedora Core and other distros. + * + * To do this, we implement our own shared object loader that reserves memory + * that is important to Wine, and then loads the the standard shared object + * loader (ELF interpreter), usually ld-linux.so.2 + * + * This program will be invoked by linking it as the shared object loader + * using the link time flags --shared-loader=ld-winepreload.so . Since we + * are not using an absolute path, an executable using this shared object + * loader must be invoked from the directory that contains this loader. + * + * We will try to set up the stack an memory area so that the program that + * loads after us (eg. the wine binary) never knows we were here, except that + * areas of memory it needs are already magically reserved. + * + * The following memory areas are important to Wine: + * 0x00000000 - the DOS area + * 0x00400000 - the standard Win32 load address + * 0x65430000 - the shared heap + * + * If this program is used as the shared object loader. the only difference + * that the loaded programs should see is that this loader will be mapped + * into memory when it starts. + * + * Limitations: + * * using a non-absolution path for the shared object loader is not standard + */ + +/* + * References (things I consulted to understand how ELF loading works): + * + * glibc 2.3.2 elf/dl-load.c + * http://www.gnu.org/directory/glibc.html + * + * Linux 2.6.4 fs/binfmt_elf.c + * ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.4.tar.bz2 + * + * Userland exec, by <grugq(a)hcunix.net> + * http://cert.uni-stuttgart.de/archive/bugtraq/2004/01/msg00002.html + * + * The ELF specification: + * http://www.linuxbase.org/spec/booksets/LSB-Embedded/LSB-Embedded/book387.ht… + */ + +#include "preload.h" +#include "main.h" +#include <elf.h> + +/* the following headers are needed to get the definition for IMAGE_NT_HEADERS */ +#include <stdarg.h> +#include "windef.h" +#include "winbase.h" + +/* debugging */ +#undef DUMP_SEGMENTS +#undef DUMP_ARG +#undef DUMP_AUX_INFO +#undef DUMP_SYMS +#undef DUMP_PERESERVE + +/* older systems may not define these */ +#ifndef PT_TLS +#define PT_TLS 7 +#endif +#ifndef PT_GNU_EH_FRAME +#define PT_GNU_EH_FRAME 0x6474e550 +#endif +#ifndef PT_GNU_STACK +#define PT_GNU_STACK 0x6474e551 +#endif +#ifndef PT_GNU_RELRO +#define PT_GNU_RELRO 0x6474e552 +#endif + +unsigned int GL(dl_pagesize) = 0x0000; +extern const unsigned char _end[]; + +struct link_map { + ElfW(Addr) l_addr; + ElfW(Dyn) *l_ld; + const ElfW(Phdr) *l_phdr; + ElfW(Addr) l_entry; + ElfW(Half) l_ldnum; + ElfW(Half) l_phnum; + ElfW(Addr) l_map_start, l_map_end; +}; + +/* + * wld_printf - just the basics + * + * %x prints a hex number + * %s prints a string + */ +void wld_sprintf(char *str, char *fmt, int *args ) +{ + char *p = fmt; + + while( *p ) + { + if( *p == '%' ) + { + p++; + if( *p == 'x' ) + { + int ch, i, x = *++args; + for(i=7; i>=0; i--) + { + ch = (x>>(i*4))&0xf; + ch += '0'; + if(ch>'9') + ch+=('A'-10-'0'); + *str++ = ch; + } + } + else if( *p == 's' ) + { + char *s = (char*)*++args; + while(*s) + *str++ = *s++; + } + else if( *p == 0 ) + break; + p++; + } + *str++ = *p++; + } + *str = 0; +} + +int wld_strlen(char *str) +{ + int n = 0; + while( str[n] ) + n++; + return n; +} + +void wld_printf(char *fmt, ... ) +{ + char buffer[100]; + int n; + + buffer[0]=0; + wld_sprintf(buffer, fmt, (int*)&fmt ); + n = wld_strlen( buffer ); + write(2, buffer, n); +} + +void wld_strncpy(char *dest, char *src, int count ) +{ + while( (*dest++ = *src++) && --count ) + ; + *dest = 0; +} + +void wld_memset( unsigned char *buf, int val, size_t len ) +{ + while( len-- ) + *buf++=val; +} + +#ifdef DUMP_AUX_INFO +/* + * Dump interesting bits of the ELF auxv_t structure that is passed + * as the 4th parameter to the _start function + */ +static void wld_dump_auxilary( ElfW(auxv_t) *av ) +{ + for ( ; av->a_type != AT_NULL; av++) + switch (av->a_type) + { + case AT_PAGESZ: + wld_printf("AT_PAGESZ = %x\n",av->a_un.a_val); + break; + case AT_PHDR: + wld_printf("AT_PHDR = %x\n",av->a_un.a_ptr); + break; + case AT_PHNUM: + wld_printf("AT_PHNUM = %x\n",av->a_un.a_val); + break; + case AT_ENTRY: + wld_printf("AT_ENTRY = %x\n",av->a_un.a_val); + break; + case AT_BASE: + wld_printf("AT_BASE = %x\n",av->a_un.a_val); + break; + } +} +#endif + +/* + * wld_set_auxilary + * + * Set a field of the auxillary structure + */ +void wld_set_auxilary( ElfW(auxv_t) *av, int type, long int val ) +{ + for ( ; av->a_type != AT_NULL; av++) + if( av->a_type == type ) + av->a_un.a_val = val; +} + +/* + * wld_get_auxilary + * + * Get a field of the auxillary structure + */ +int wld_get_auxilary( ElfW(auxv_t) *av, int type, int *val ) +{ + for ( ; av->a_type != AT_NULL; av++) + if( av->a_type == type ) + { + *val = av->a_un.a_val; + return 1; + } + return 0; +} + +/* + * wld_map_so + * + * modelled after _dl_map_object_from_fd() from glibc-2.3.1/elf/dl-load.c + * + * This function maps the segments from an ELF object, and optionally + * stores information about the mapping into the auxv_t structure. + */ +void* wld_map_so( int fd, void* buf, size_t buflen, struct link_map *l) +{ + ElfW(Ehdr) *header = buf; + ElfW(Phdr) *phdr, *ph; + /* Scan the program header table, collecting its load commands. */ + struct loadcmd + { + ElfW(Addr) mapstart, mapend, dataend, allocend; + off_t mapoff; + int prot; + } loadcmds[header->e_phnum], *c; + size_t nloadcmds = 0, maplength; + + header = buf; + phdr = (void*) (((unsigned char*)buf) + header->e_phoff); + + if( header->e_ident[0] != 0x7f && + header->e_ident[0] != 'E' && + header->e_ident[0] != 'L' && + header->e_ident[0] != 'F' ) + { + wld_printf("Not an ELF binary... don't know how to load it\n"); + return NULL; + } + + if( header->e_machine != EM_386 ) + { + wld_printf("Not an i386 ELF binary... don't know how to load it\n"); + return NULL; + } + + maplength = header->e_phnum * sizeof (ElfW(Phdr)); + if (header->e_phoff + maplength > (size_t) buflen) + { + wld_printf("oops... not enough space for ELF headers\n"); + exit(1); + } + + l->l_ld = 0; + l->l_addr = 0; + l->l_phnum = header->e_phnum; + l->l_entry = header->e_entry; + + for (ph = phdr; ph < &phdr[l->l_phnum]; ++ph) + { + +#ifdef DUMP_SEGMENTS + wld_printf( "ph = %x\n", ph ); + wld_printf( " p_type = %x\n", ph->p_type ); + wld_printf( " p_flags = %x\n", ph->p_flags ); + wld_printf( " p_offset = %x\n", ph->p_offset ); + wld_printf( " p_vaddr = %x\n", ph->p_vaddr ); + wld_printf( " p_paddr = %x\n", ph->p_paddr ); + wld_printf( " p_filesz = %x\n", ph->p_filesz ); + wld_printf( " p_memsz = %x\n", ph->p_memsz ); + wld_printf( " p_align = %x\n", ph->p_align ); +#endif + + switch (ph->p_type) + { + /* These entries tell us where to find things once the file's + segments are mapped in. We record the addresses it says + verbatim, and later correct for the run-time load address. */ + case PT_DYNAMIC: + l->l_ld = (void *) ph->p_vaddr; + l->l_ldnum = ph->p_memsz / sizeof (Elf32_Dyn); + break; + + case PT_PHDR: + l->l_phdr = (void *) ph->p_vaddr; + break; + + case PT_TLS: + /* + * We don't need to set anything up because we're + * emulating the kernel, not ld-linux.so.2 + * The ELF loader will set up the TLS data itself. + */ + break; + + case PT_LOAD: + { + if ((ph->p_align & (GL(dl_pagesize) - 1)) != 0) + { + wld_printf("ELF load command alignment not page-aligned"); + return NULL; + } + if (((ph->p_vaddr - ph->p_offset) & (ph->p_align - 1)) != 0) + { + wld_printf("ELF load command address/offset not properly aligned"); + return NULL; + } + + c = &loadcmds[nloadcmds++]; + c->mapstart = ph->p_vaddr & ~(ph->p_align - 1); + c->mapend = ((ph->p_vaddr + ph->p_filesz + GL(dl_pagesize) - 1) + & ~(GL(dl_pagesize) - 1)); + c->dataend = ph->p_vaddr + ph->p_filesz; + c->allocend = ph->p_vaddr + ph->p_memsz; + c->mapoff = ph->p_offset & ~(ph->p_align - 1); + + c->prot = 0; + if (ph->p_flags & PF_R) + c->prot |= PROT_READ; + if (ph->p_flags & PF_W) + c->prot |= PROT_WRITE; + if (ph->p_flags & PF_X) + c->prot |= PROT_EXEC; + } + break; + + case PT_INTERP: + break; + + case PT_SHLIB: + case PT_NOTE: + case PT_GNU_EH_FRAME: + case 0x6474e551: /* FIXME: what is this? */ + break; + + default: + wld_printf("warning: unknown header type %x\n", ph->p_type); + break; + } + } + + /* Now process the load commands and map segments into memory. */ + c = loadcmds; + + /* Length of the sections to be loaded. */ + maplength = loadcmds[nloadcmds - 1].allocend - c->mapstart; + + if( header->e_type == ET_DYN ) + { + ElfW(Addr) mappref; + mappref = (ELF_PREFERRED_ADDRESS (loader, maplength, c->mapstart) + - MAP_BASE_ADDR (l)); + + /* Remember which part of the address space this object uses. */ + l->l_map_start = (ElfW(Addr)) mmap ((void *) mappref, maplength, + c->prot, MAP_COPY | MAP_FILE, + fd, c->mapoff); + /* wld_printf("set : offset = %x\n", c->mapoff); */ + /* wld_printf("l->l_map_start = %x\n", l->l_map_start); */ + + l->l_map_end = l->l_map_start + maplength; + l->l_addr = l->l_map_start - c->mapstart; + + mprotect ((caddr_t) (l->l_addr + c->mapend), + loadcmds[nloadcmds - 1].allocend - c->mapend, + PROT_NONE); + goto postmap; + } + else + { + ELF_FIXED_ADDRESS (loader, c->mapstart); + } + + /* Remember which part of the address space this object uses. */ + l->l_map_start = c->mapstart + l->l_addr; + l->l_map_end = l->l_map_start + maplength; + + while (c < &loadcmds[nloadcmds]) + { + if (c->mapend > c->mapstart) + /* Map the segment contents from the file. */ + mmap ((void *) (l->l_addr + c->mapstart), + c->mapend - c->mapstart, c->prot, + MAP_FIXED | MAP_COPY | MAP_FILE, fd, c->mapoff); + + postmap: + if (l->l_phdr == 0 + && (ElfW(Off)) c->mapoff <= header->e_phoff + && ((size_t) (c->mapend - c->mapstart + c->mapoff) + >= header->e_phoff + header->e_phnum * sizeof (ElfW(Phdr)))) + /* Found the program header in this segment. */ + l->l_phdr = (void *) (c->mapstart + header->e_phoff - c->mapoff); + + if (c->allocend > c->dataend) + { + /* Extra zero pages should appear at the end of this segment, + after the data mapped from the file. */ + ElfW(Addr) zero, zeroend, zeropage; + + zero = l->l_addr + c->dataend; + zeroend = l->l_addr + c->allocend; + zeropage = ((zero + GL(dl_pagesize) - 1) + & ~(GL(dl_pagesize) - 1)); + + /* + * This is different to the dl-load load... + * ld-linux.so.2 relies on the whole page being zero'ed + */ + zeroend = (zeroend +GL(dl_pagesize) - 1) & ~(GL(dl_pagesize) - 1); + + if (zeroend < zeropage) + { + /* All the extra data is in the last page of the segment. + We can just zero it. */ + zeropage = zeroend; + } + + if (zeropage > zero) + { + /* Zero the final part of the last page of the segment. */ + if ((c->prot & PROT_WRITE) == 0) + { + /* Dag nab it. */ + mprotect ((caddr_t) (zero & ~(GL(dl_pagesize) - 1)), + GL(dl_pagesize), + c->prot|PROT_WRITE); + } + memset ((void *) zero, '\0', zeropage - zero); + if ((c->prot & PROT_WRITE) == 0) + mprotect ((caddr_t) (zero & ~(GL(dl_pagesize) - 1)), + GL(dl_pagesize), c->prot); + } + + if (zeroend > zeropage) + { + /* Map the remaining zero pages in from the zero fill FD. */ + caddr_t mapat; + mapat = mmap ((caddr_t) zeropage, zeroend - zeropage, + c->prot, MAP_ANON|MAP_PRIVATE|MAP_FIXED, + ANONFD, 0); + } + } + + ++c; + } + + if (l->l_phdr == NULL) + { + wld_printf("no program header\n"); + exit(1); + } + + (ElfW(Addr)) l->l_phdr += l->l_addr; + (ElfW(Addr)) l->l_entry += l->l_addr; + + return (void*)l->l_entry; +} + +#define ROUND_SIZE(addr,size) \ + (((int)(size) + ((int)(addr) & page_mask) + page_mask) & ~page_mask) + +/* + * wld_reserve_wine_memory + * + * This is the reason we are here. + * We want to reserve address space so that the kernel doesn't + * map stuff there. We do this by getting in first before + * anything else is mapped (except this program) and reserve + * the memory we want. + */ +void wld_reserve_wine_memory( char *pe_file, void **base, size_t *size ) +{ + IMAGE_DOS_HEADER dos; + IMAGE_NT_HEADERS nt; + int unix_fd, page_mask = GL(dl_pagesize) - 1; + + if( !pe_file ) + return; + + unix_fd = open( pe_file, O_RDONLY, 0); + if( unix_fd < 0 ) + return; + + /* + * The following code is borrowed from + * get_image_params() in wine/server/mapping.c + */ + if (read( unix_fd, &dos, sizeof(dos) ) != sizeof(dos)) goto error; + if (dos.e_magic != IMAGE_DOS_SIGNATURE) goto error; + if (lseek( unix_fd, dos.e_lfanew, SEEK_SET ) == -1) goto error; + + if (read( unix_fd, &nt.Signature, sizeof(nt.Signature) ) != sizeof(nt.Signature)) goto error; + if (nt.Signature != IMAGE_NT_SIGNATURE) goto error; + if (read( unix_fd, &nt.FileHeader, sizeof(nt.FileHeader) ) != sizeof(nt.FileHeader)) goto error; + /* zero out Optional header in the case it's not present or partial */ + memset(&nt.OptionalHeader, 0, sizeof(nt.OptionalHeader)); + if (read( unix_fd, &nt.OptionalHeader, nt.FileHeader.SizeOfOptionalHeader) != nt.FileHeader.SizeOfOptionalHeader) goto error; + + *size = ROUND_SIZE( 0, nt.OptionalHeader.SizeOfImage ); + *base = (void *)nt.OptionalHeader.ImageBase; + +#ifdef DUMP_PERESERVE + wld_printf( "file %s\n", pe_file ); + wld_printf( "parsed PE image... base %x size %x\n", *base, *size ); +#endif + +error: + close( unix_fd ); +} + +/* + * Find a symbol in the symbol table of the executable loaded + */ +void *wld_find_symbol( const ElfW(Phdr) *phdr, int num, char *var ) +{ + const ElfW(Dyn) *dyn = NULL; + void *ptr = NULL; + const ElfW(Phdr) *ph; + const ElfW(Sym) *symtab = NULL; + const char *strings = NULL; + + /* check the values */ +#ifdef DUMP_SYMS + wld_printf("%x %x\n", phdr, num ); +#endif + if( ( phdr == NULL ) || ( num == 0 ) ) + { + wld_printf("could not find PT_DYNAMIC header entry\n"); + return NULL; + } + + /* parse the (already loaded) ELF executable's header */ + for (ph = phdr; ph < &phdr[num]; ++ph) + { + if( PT_DYNAMIC == ph->p_type ) + { + dyn = (void *) ph->p_vaddr; + num = ph->p_memsz / sizeof (Elf32_Dyn); + break; + } + } + if( !dyn ) + { + wld_printf("No PT_DYNAMIC section\n"); + return ptr; + } + + while( dyn->d_tag ) + { + if( dyn->d_tag == DT_STRTAB ) + strings = (const char*) dyn->d_un.d_ptr; + if( dyn->d_tag == DT_SYMTAB ) + symtab = (const ElfW(Sym) *)dyn->d_un.d_ptr; +#ifdef DUMP_SYMS + wld_printf("%x %x\n", dyn->d_tag, dyn->d_un.d_ptr ); +#endif + dyn++; + } + + if( (!symtab) || (!strings) ) + { + wld_printf("No symbol table found\n"); + return ptr; + } + + /* FIXME: this crashes if it doesn't find the symbol we want */ + do + { + if( ( ELF32_ST_BIND(symtab->st_info) == STT_OBJECT ) && + ( 0 == strcmp( strings+symtab->st_name, var ) ) ) + { + ptr = (void*)symtab->st_value; + break; + } + symtab++; + } + while( symtab->st_name ); + +#ifdef DUMP_SYMS + wld_printf("Found %s -> %x\n", strings+symtab->st_name, symtab->st_value ); +#endif + + return ptr; +} + +/* + * wld_start + * + * Repeat the actions the kernel would do when loading a dynamically linked .so + * Load the ELF interperter ld-linux.so and then binary itself. + * + * Note, we assume that the binary is a dynamically linked ELF shared object + * and that ld-linux.so is its ELF interpreter. + */ +void* wld_start( int argc ) +{ + int fd, phnum = 0; + unsigned char buf[0x800]; + void *start; + char **argv = (char**) &argc, **p, *pe_file = NULL; + struct link_map map; + ElfW(auxv_t)* av; + ElfW(Phdr) *elf_phdr = NULL; + struct wine_preload_info **wine_preload_vars; + static struct wine_preload_info vars; + + memset( &map, 0, sizeof map ); + + /* skip over the parameters */ + p = ++argv; +#ifdef DUMP_ARG + { + int i; + for( i = 0; i<argc; i++ ) + wld_printf("argv[%x] = %s\n", i, *p++); + if(*p++) + wld_printf("argv[argc] not NULL!!\n"); + } +#else + p+=(argc+1); +#endif + + /* skip over the environment, find the name of the PE file */ + while( *p ) + { + const char pefile[] = "WINEPEEXEPATH="; + if( 0 == memcmp( pefile, *p, sizeof pefile - 1 ) ) + pe_file = (*p) + sizeof pefile - 1 ; + p++; + } + + av = (ElfW(auxv_t)*) (p+1); + wld_get_auxilary( av, AT_PAGESZ, &dl_pagesize ); +#ifdef DUMP_AUX_INFO + wld_dump_auxilary( av ); +#endif + + /* reserve memory that Wine needs */ + vars.pe_size = 0x01000000; + vars.pe_address = (void*)0x00400000; + wld_reserve_wine_memory( pe_file, &vars.pe_address, &vars.pe_size ); + mmap( vars.pe_address, vars.pe_size, PROT_NONE, + MAP_FIXED|MAP_PRIVATE|MAP_ANON, -1, 0); + + /* reserve the memory that wine needs for the global heap */ + vars.heap_size = 0x400000; + vars.heap_address = (void*)0x65430000; + mmap( vars.heap_address, vars.heap_size, PROT_NONE, + MAP_FIXED|MAP_PRIVATE|MAP_ANON, -1, 0); + + /* load the original dynamic loader itself */ + fd = open( LDSO_NAME, O_RDONLY, 0); + if( fd < 0 ) + { + wld_printf( "error: fd = %x\n", fd); + exit(1); + } + read( fd, buf, sizeof buf ); + start = wld_map_so( fd, buf, sizeof buf, &map ); + if( !start ) + exit(0); + close( fd ); + + /* + * Pass information to the wine loader in a static variable + * so it doesn't get destroyed after we leave this stack frame + */ + wld_get_auxilary( av, AT_PHDR, (int*)&elf_phdr ); + wld_get_auxilary( av, AT_PHNUM, (int*)&phnum ); + wine_preload_vars = wld_find_symbol( elf_phdr, phnum, "wine_preload_vars" ); + if( wine_preload_vars ) + { + unsigned char *base = (void*)0x7ff81000; /* FIXME - define properly */ + vars.preload_base = (void*) base; + vars.preload_size = (size_t) (_end - base); + *wine_preload_vars = &vars; + } + + wld_set_auxilary( av, AT_BASE, map.l_addr ); + +#ifdef DUMP_AUX_INFO + wld_printf("New auxillary info:\n"); + wld_dump_auxilary( av ); + wld_printf("jumping to %x\n", start); +#endif + + return start; +} + +/* + * The _start function is the entry and exit point of this program + * + * It calls wld_start, passing a pointer to the args it receives + * then jumps to the address wld_start returns after removing the + * first argv[] value, and decrementing argc + */ +void _start(void); +__asm ( + ".align 4\n" + "\t.global _start\n" + "\t.type _start,@function\n" +"_start:\n" + "\tcall wld_start\n" + "\tpush %eax\n" + "\txor %eax,%eax\n" + "\txor %ebx,%ebx\n" + "\txor %ecx,%ecx\n" + "\txor %edx,%edx\n" + "\tret\n" +); +

1 0

Re: Re: statvfs and cdrom problem
by Christian Costa April 15, 2004

April 15, 2004

> Message du 15/04/04 00:35 > De : Alexandre Julliard > A : Christian Costa > Copie à : wined > Objet : Re: statvfs and cdrom problem > Christian Costa writes: > > > Yes, this works fine. Thanks. :-) > > When do you plan to rewrite GetFreeDiskSpace? > > In a few days I guess. There are still some missing features in > NtCreateFile that need to be done first. > > -- > Alexandre Julliard > julliard(a)winehq.org > Ok! Fine! :-) I have a small fix for the lpSectorsPerCluster param which should always contain 1 for CDROM. I will send a patch after your rewrite unless you fix it first. :-) Bye, Christian

1 0

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 14, 2004

April 14, 2004

Hi, But dont you think that the structure is not really required till the window has been created successfully? Regards Santosh Siddheshwar > -----Original Message----- > From: Alexandre Julliard [SMTP:julliard(a)winehq.org] > Sent: Thursday, April 15, 2004 9:00 AM > To: Dmitry Timoshkov > Cc: Santosh Siddheshwar; Robert Shearman; wine-devel(a)winehq.org > Subject: Re: Handling dialog messages for non-template based dialogs > > "Dmitry Timoshkov" <dmitry(a)baikal.ru> writes: > > > Yes, if Alexandre has no objections. Be patient, that may take some > time. > > Actually, I do have an objection... I think you should be allocating > the structure as soon as it is needed, WM_CREATE is not the first > message that the window will receive. > > -- > Alexandre Julliard > julliard(a)winehq.org ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 14, 2004

April 14, 2004

Hi, So can I expect this to be checked into cvs soon? Regards Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Wednesday, April 14, 2004 5:22 PM > To: Santosh Siddheshwar; Robert Shearman > Cc: wine-devel(a)winehq.org > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > Thanks for the feedback. Excellent catches! Hope my next mail to > > wine-patches gets through! > > Your new patch looks good for me. Thanks for doing this work! > > -- > Dmitry. ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

3 2

Re: abnt2 keyboard patch
by Dmitry Timoshkov April 14, 2004

April 14, 2004

"Rafael аvila de EspМndola" <rafael.espindola(a)ic.unicamp.br> wrote: > I have tested the patch with a abnt2 and a us-int keyboard. Both work > correctly. > > Dmitry: Do you think that the "magic numbers" are reasonably explained? What I > have done is basically to treat the comma similarilly to the decimal period. > The numbers are different from the first patch because in the first one I > haven't fixed the main_key_scan_abnt_qwerty and main_key_vkey_abnt_qwerty > tables. I still see problems with your patch. 1. You are using 'keysym == 0x2C' for comma, but what happens if NumLock is on, and you press comma on the main keyboard, not on the keypad? Shouldn't X11 send XK_KP_xxx event in that case (something like XK_KP_Decimal)? 2. Please generate unified diffs, 'using diff -u' for patches. -- Dmitry.

1 0

dbghelp: cross-compile
by Kevin Koltzau April 14, 2004

April 14, 2004

I decided to try cross compiling all of wine using mingw..hit a couple problems in dbghelp This fixes a few of them, it still doesn't compile (primarily because of a missing regex.h at the moment) but it does get farther I'm not sure if these fixes are quite right..I don't fully understand this code Changelog Fix a few compile errors when cross compiling

1 0

Re: MAPI32 - headers
by Alexandre Julliard April 14, 2004

April 14, 2004

Jon Griffiths <jon_p_griffiths(a)yahoo.com> writes: > -#ifndef MAPICODE_H > -#define MAPICODE_H > +#ifndef __WINE_MAPICODE_H > +#define __WINE_MAPICODE_H It's better to keep the same defines as Windows, especially since in that case they are used from other headers. > -#define MAKE_MAPI_SCODE(sev,fac,code) \ > - ( (((ULONG)(sev)<<31) | ((ULONG)(fac)<<16) | ((ULONG)(code))) ) MAKE_MAPI_SCODE and the other macros are in the Windows header so we need them, you can't remove them. > +#ifndef __FLAGS > +#define __FLAGS > +typedef unsigned long FLAGS; > +#endif There is no such #ifdef in the Windows header. If there are conflicting typedefs in the Windows headers then we want conflicts in the Wine headers too, otherwise we won't be able to build the Wine code against the Windows headers. -- Alexandre Julliard julliard(a)winehq.org

1 0

Bug 560 "Integrate" AppDB and bugzilla
by Tony Lambregts April 14, 2004

April 14, 2004

This is an update for bug 560 http://bugs.winehq.org/show_bug.cgi?id=560 I have gone through each and every bug in Bugzilla (yes all 2174 of them) and linked the ones that I could find an AppDB entry for. That translates into 756 bugs. Some of these bugs applied to more than one program. In those cases I tried to link it to the most appropriate application. That usually translated into the most widely used app. http://bugs.winehq.org/buglist.cgi?bug_file_loc_type=allwords&bug_file_loc=… About halfway through I created a new Keyword (NoAppDBEntry) to track those bugs that I could not find an AppDB entry for. There are 249 bugs with this keyword. http://bugs.winehq.org/buglist.cgi?keywords_type=allwords&keywords=NoAppDBE… There are a fair number of apps in this list that have more than one bug entered. The one that sticks out in my mind right now is Xilinx, although there were quite a few others. What strikes me most about this, is that as I got to the newer bug reports I ran into more and more esoteric programs. I think that is a good sign. Once Jerry Has committed my bugs.php patch I will consider this bug closed. In order to improve this more requires opening up the source code for our Bugzilla. -- Tony Lambregts PS: Could I get someone to give me more rights in the AppDB. My user name is "tonylambregts" (go figure).

1 0

statvfs and cdrom problem
by Christian Costa April 14, 2004

April 14, 2004

Hi, I'm trying to make GetFreeDiskSpace working with CDROM because a game needs it. However, I'm experiencing problem with statvfs... Indeed, statvfs always returns f_frsize = 0 and f_blocks = 0. If I cd to /mnt/cdrom or edit a file in it, this does not help however if I run a grep or launch Konqueror in that directory, statvfs magically works. I suspect supermount is responsible because statvfs needs the cdrom to be "physically" mounted. Is it a know issue? (I am on a Mandrake 9.1 base) Or the only solution is to keep a fd opened during the statvfs call so the drive stays mounted... Thanks in advance, Christian PS: I used the attached code for testing #include <sys/types.h> #include <sys/statvfs.h> #include <sys/statfs.h> int main() { struct statvfs info; struct statfs inf; if (statfs("/mnt/cdrom", &inf) < 0) { printf("cannot do statfs\n"); return 0; } printf("size = %ld - %ld\n", inf.f_bsize, inf.f_blocks); if (statvfs("/mnt/cdrom", &info ) < 0) { printf("cannot do statvfs\n"); return 0; } printf("size = %ld - %ld - %ld\n", info.f_bsize, info.f_frsize, info.f_blocks); return 1; }

2 3

Installer crash
by andrejoh＠c2i.net April 14, 2004

April 14, 2004

When trying to install the game Zoo Tycoon, today's CVS checkout of Wine crashes with the following output: [andrej(a)rivendell cdrom]$ nice wine ./setup err:dosmem:setup_dos_mem Cannot use first megabyte for DOS address space, please report err:wave:OSS_WaveOutInit /dev/mixer1: No such device err:wave:OSS_WaveInInit /dev/mixer1: No such device wine: Unhandled exception (thread 000d), starting debugger... err:dosmem:setup_dos_mem Cannot use first megabyte for DOS address space, please report WineDbg starting on pid c Can't find file for ELF 'wine-kthread' ((nil)) Loaded debug information from ELF 'wine-pthread' ((nil)) Breakpoint 1 at 0x0010df00 Loaded debug information from ELF '/home/andrej/local/wine/lib/libwine.so.1' (0x40001000) No debug information in ELF '/lib/tls/libpthread.so.0' ((nil)) No debug information in ELF '/lib/tls/libc.so.6' ((nil)) No debug information in ELF '/lib/libdl.so.2' ((nil)) No debug information in ELF '/lib/ld-linux.so.2' ((nil)) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/ntdll.dll.so' (0x4002c000) Loaded debug information from ELF '/home/andrej/local/wine/lib/libwine_unicode.so.1' (0x40098000) No debug information in ELF '/lib/tls/libm.so.6' ((nil)) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/kernel32.dll.so' (0x402a0000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/user32.dll.so' (0x406c0000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/gdi32.dll.so' (0x407ea000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/advapi32.dll.so' (0x40864000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/winmm.dll.so' (0x40893000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/comctl32.dll.so' (0x408fd000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/shell32.dll.so' (0x40995000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/ole32.dll.so' (0x40a18000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/rpcrt4.dll.so' (0x40a87000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/shlwapi.dll.so' (0x40ace000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/version.dll.so' (0x40b22000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/lz32.dll.so' (0x40b3b000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/imm32.dll.so' (0x40b51000) No debug information in ELF '/usr/lib/libfreetype.so.6' ((nil)) No debug information in ELF '/usr/lib/libz.so.1' ((nil)) No debug information in ELF '/usr/lib/libfontconfig.so.1' ((nil)) No debug information in ELF '/usr/lib/libexpat.so.0' ((nil)) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/x11drv.dll.so' (0x40d90000) No debug information in ELF '/usr/X11R6/lib/libSM.so.6' ((nil)) No debug information in ELF '/usr/X11R6/lib/libICE.so.6' ((nil)) No debug information in ELF '/usr/X11R6/lib/libXv.so.1' ((nil)) No debug information in ELF '/usr/X11R6/lib/libXext.so.6' ((nil)) No debug information in ELF '/usr/X11R6/lib/libX11.so.6' ((nil)) No debug information in ELF '/usr/X11R6/lib/libXcursor.so.1' ((nil)) No debug information in ELF '/usr/X11R6/lib/libXrender.so.1' ((nil)) No debug information in ELF '/usr/X11R6/lib/X11/locale/lib/common/xlcUTF8Load.so.2' (0x40021000) No debug information in ELF '/usr/X11R6/lib/X11/locale/lib/common/ximcp.so.2' (0x40e4c000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/wineoss.drv.so' (0x40e6a000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/msacm.drv.so' (0x40ea6000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/msacm32.dll.so' (0x40ebd000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/midimap.drv.so' (0x40ff0000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/vwin32.vxd.so' (0x41005000) Loaded debug information from ELF '/home/andrej/local/wine/lib/wine/winedos.dll.so' (0x41330000) No debug information in 32bit DLL 'D:\setup.exe' (0x400000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\NTDLL.DLL' (0x40050000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\KERNEL32.DLL' (0x402d0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\ADVAPI32.DLL' (0x40870000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\GDI32.DLL' (0x40800000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\USER32.DLL' (0x406e0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\WINMM.DLL' (0x408a0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\COMCTL32.DLL' (0x40910000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\RPCRT4.DLL' (0x40aa0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\OLE32.DLL' (0x40a30000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\SHLWAPI.DLL' (0x40af0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\SHELL32.DLL' (0x409b0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\LZ32.DLL' (0x40b40000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\VERSION.DLL' (0x40b30000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\IMM32.DLL' (0x40b60000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\X11DRV.DLL' (0x40da0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\WINEOSS.DRV' (0x40e80000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\MSACM32.DLL' (0x40ed0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\MSACM.DRV' (0x40eb0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\MIDIMAP.DRV' (0x41000000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\WINEDOS.DLL' (0x41340000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\VWIN32.VXD' (0x41010000) Unhandled exception: page fault on write access to 0x000d0000 in 32-bit code (0x4134b69b). In 32-bit mode. 0x4134b69b (DOSVM_AllocCodeUMB+0x27 [himem.c:121] in WINEDOS.DLL): movw %ax,0x0(%edi) 123 if (selector) Wine-dbg>bt Backtrace: =>0 0x4134b69b (DOSVM_AllocCodeUMB+0x27(size=0x3, segment=0xd0000, selector=0x0) [himem.c:121] in WINEDOS.DLL) (ebp=406be0e4) 1 0x4134b740 (DOSVM_InitSegments+0x2c [himem.c:222] in WINEDOS.DLL) (ebp=406be108) 2 0x4134b017 (DllMain+0x43(hinstDLL=0x41340000, fdwReason=0x1, lpvReserved=0x0) [dosvm.c:725] in WINEDOS.DLL) (ebp=406be11c) 3 0x413470b4 (__wine_dll_main+0x60 [winedos.dll-Iw24bO.spec.c] in WINEDOS.DLL) (ebp=406be14c) 4 0x4009368a (call_dll_entry_point+0x12 in NTDLL.DLL) (ebp=406be164) 5 0x400660cb (MODULE_InitDLL+0xdf(wm=0x401d1370, reason=0x1, lpReserved=0x0) [loader.c:724] in NTDLL.DLL) (ebp=406be1d4) 6 0x40066225 (process_attach+0x81(wm=0x401d1370, lpReserved=0x0) [loader.c:797] in NTDLL.DLL) (ebp=406be1f8) 7 0x400662bf (process_attach+0x11b(wm=0x401d12d0, lpReserved=0x0) [loader.c:789] in NTDLL.DLL) (ebp=406be218) 8 0x400679e4 (LdrLoadDll+0x84(path_name=0x401d1b08, flags=0x0, libname=0x406be27c, hModule=0x406be25c) [loader.c:1529] in NTDLL.DLL) (ebp=406be23c) 9 0x4030dfe7 (load_library+0x4b(libname=0x406be27c, flags=0x0) [module.c:668] in KERNEL32.DLL) (ebp=406be26c) 10 0x4030e0be (LoadLibraryExW+0x2e(libnameW=0x406be32c, hfile=0x0, flags=0x0) [module.c:731] in KERNEL32.DLL) (ebp=406be28c) 11 0x4030e118 (LoadLibraryW+0x1c(libnameW=0x406be32c) [module.c:761] in KERNEL32.DLL) (ebp=406be2a4) 12 0x4033300b (VXD_Open+0x163(filenameW=0x401d1af0, access=0x0, sa=0x0) [vxd.c:192] in KERNEL32.DLL) (ebp=406be358) 13 0x402e2232 (CreateFileW+0x51a(filename=0x401d1ae8, access=0x0, sharing=0x0, sa=0x0, creation=0x3, attributes=0x4000000, template=0x0) [file.c:382] in KERNEL32.DLL) (ebp=406befc4) 14 0x402e2582 (CreateFileA+0x62(filename=0x463fa4, access=0x0, sharing=0x0, sa=0x0, creation=0x3, attributes=0x4000000, template=0x0) [file.c:473] in KERNEL32.DLL) (ebp=406beffc) 15 0x0041b8ac (setup.exe.LaunchGame+0x14bfc in setup.exe) (ebp=00000003) *** Invalid address 0x00000003 (_begin+0x3) My previous compile from CVS (can't remember the date; probably a couple of weeks old) got a lot further, but ended up hanging partway into the installation. This is on Fedora Core 1/i386, with exec-shield turned off. -- Try Anarchy Online http://www.anarchy-online.com/content/downloads/tryout/ Netiquette guidelines ftp://ftp.rfc-editor.org/in-notes/rfc1855.txt How to quote properly http://www.i-hate-computers.demon.co.uk/ Hvordan sitere riktig http://home.online.no/~vidaandr/news/OBSquoting.html

2 4

More serial access issues
by Geoffrey Hausheer April 14, 2004

April 14, 2004

Well, I broke down and pulled out the oscope and rs232 breakout box to try to figure out why my app isn't able to communicate with the device. turns out that Rein's patch is correctly deasserting DSR, however, I am not getting any data over the Tx pin. And after I checked the output of WriteFile, it is failing. GetLastError doesn't return anything useful (a value of 997 which FormatMessage() can't do anything with). But some debugging led me to determine that NtWriteFile is returning 0x103 which is STATUS_PENDING. After that I'm somewhat lost as to what is going on. Obvioulsy, 'register_new_async' is returning this status, which apparently happens when wine_server_call returns 0 or STATUS_PENDING (not sure if the second is even possible). But about here I am getting completely confused as to what we are trying to accomplish. The file is opened via: CreateFile("COM1", 0, NULL, OPEN_EXISTING, 0, NULL) so we're not in overlapped mode. SetCommTimeout is called with the COMMTIMEOUTS structure containing all 0s. All flow control is disabled using SetCommState. Before a write, we always call 'PurgeComm' and ClearCommError. In Windows, this combination results in a write always occurring when WriteFile is called, regardless of whether there is anything connected to the serial port. Wine appears to be waiting for something, and I'm not sure what. If anyone has any ideas how I might proceed, i'd be glad to hear them. Thanks, .Geoff

1 0

AutoFS-ed drives
by Mark Draheim April 14, 2004

April 14, 2004

Anybody using autofs and wine? Does it work? I don't follow wine development lists, so maybe autofs is no longer a problem. Once in a while I build a CVS snapshot and test whether the odd app I still have works. And wine on autofs doesn't work here. I think Guido already sent a mail about this last month. Guido patched wine way back in Feb 2003, submitted a bug http://bugs.winehq.com/show_bug.cgi?id=1283 and even set up a page to describe the problem http://freespace.sourceforge.net/guidod/wine-vol-a/ Since then he has completely rewritten the patch in order to search the /etc/auto.* files for drives. This obsoleted the additional options in wine's config with the old patch. This last patch worked perfectly until recently. However, Guido is busy right now and cannot work on the patch. And my C is pretty bad. If anybody is interested I can provide the last patch known to work. If for some reason the patch does not meet wine's standards, would it be possible that the autofs problem is recognized and put on a TODO list for later? Or even close the bug as WONTFIX? thanks Mark

2 2

Re: ISF_MyComputer_fnParseDisplayName()
by Dmitry Timoshkov April 14, 2004

April 14, 2004

"Martin Fuchs" <martin-fuchs(a)gmx.net> wrote: > Changelog: > make drive letters in PIDLs always uppercase to enable PIDL comparison > > > Index: shfldr_mycomp.c > =================================================================== > RCS file: /CVS/ReactOS/reactos/lib/shell32/shfldr_mycomp.c,v > retrieving revision 1.10 > diff -u -p -d -r1.10 shfldr_mycomp.c > --- shfldr_mycomp.c 11 Apr 2004 13:24:29 -0000 1.10 > +++ shfldr_mycomp.c 13 Apr 2004 20:45:39 -0000 > @@ -217,6 +217,7 @@ ISF_MyComputer_fnParseDisplayName (IShel > /* do we have an absolute path name ? */ > else if (PathGetDriveNumberW (lpszDisplayName) >= 0 && lpszDisplayName[2] == (WCHAR) '\\') { > szNext = GetNextElementW (lpszDisplayName, szElement, MAX_PATH); > + szElement[0] = toupper(szElement[0]); /* make drive letter uppercase to enable PIDL comparison */ It's better to use unicode version here (toupperW) since in some linux locales uppercasing is done differently. -- Dmitry.

3 2

Re: shell32 - file not found
by Juan Lang April 14, 2004

April 14, 2004

Hi Martin, Martin wrote: > But we could use the expression > "MAKE_HRESULT(SEVERITY_ERROR, FACILITY_WIN32, > ERROR_FILE_NOT_FOUND)" instead. A shorter version is HRESULT_FROM_WIN32(ERROR_FILE_NOT_FOUND). Question: could you not change _ILCreateFromPathA to return an HRESULT? --Juan __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html

2 1