- Wine-Devel - list.winehq.org

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 21, 2004

April 21, 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