September 2009 - Wine-Devel

re: fixing ReadDirectoryChangesW()
by Dan Kegel Sept. 12, 2009

Sept. 12, 2009

Michael Pujos wrote: > I'm willing to resubmit this patch with updated tests > against current Wine, but before I procceed > I need to be sure it won't be rejected simply because it changes the protocol. Server changes are tricky to get in. You have to be patient and persistent. I think your best next move is to file a bug describing the problem (and give the URL for the app it affects for you, if possible), and submit just the tests first (with appropriate todo_wine lines where wine gets it wrong currently). All that is useful, and a good first step towards getting the bug fixed. Thanks! - Dan

3 2

Re: DDraw: Add a few missing WINAPIs
by Stefan Dösinger Sept. 12, 2009

Sept. 12, 2009

Am Saturday 12 September 2009 23:41:00 schrieb Stefan Dösinger: > Am Saturday 12 September 2009 23:25:56 schrieb Stefan Dösinger: > > Oops: forgot the patch: Er, never mind that patch - those functions aren't used directly in the vtable. There are the FPUSetup and FPUPreserve wrappers around them, so the default calling convention is ok.

1 0

RE: Help with a native Windows test program needed (CreateFileW).
by Markus Stockhausen Sept. 12, 2009

Sept. 12, 2009

Sorry for the the wrong feedback. As I have limited possibilites creating executables on Windows I relied on Thomas executable for the tests. Maybe you can post a corrected version. It is not my intention to confuse someone. Im just trying to narrow down my initial analysis. The last write access to the malicous memory address before the application dump comes from CreateFileW. Best regards. -------- Weitergeleitete Nachricht -------- Von: wine-bugs(a)winehq.org An: mst(a)collogia.de Betreff: [Bug 9575] IBExpert need windows native ole32.dll to run Datum: Sat, 12 Sep 2009 05:18:49 -0500 http://bugs.winehq.org/show_bug.cgi?id=9575 --- Comment #15 from Alexandre Julliard <julliard(a)winehq.org> 2009-09-12 05:18:42 --- There's no such behavior. The test is broken.

2 5

Re: [ntdll] Add test for NtQueryDirectoryFile
by Paul Vriens Sept. 12, 2009

Sept. 12, 2009

On 09/12/2009 02:18 AM, Dan Kegel wrote: > Here's the next broken-out test patch from the series that > implements support for hidden and system attributes. > This one just adds a test for NtQueryDirectoryFile(). > It passes for me on Wine and Vista Home Basic. > > [Paul, could you test it on your box to make sure it's ok there?] Hi Dan, No failures on my NT4+ boxes (32bit). One thing I noticed is that you don't cleanup in some of the failure cases which would leave that temporary directory around. -- Cheers, Paul.

1 2

Help with a native Windows test program needed (CreateFileW).
by Markus Stockhausen Sept. 12, 2009

Sept. 12, 2009

Hi, as I have no Windows machine and cannot test this. I'm in need for someone who can build a small exe that gives further hints to my observations in bug #9575. The following is needed: call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) Return the result of second call. I expect it to be a valid handle on Windows whereas Wine will return an invalid handle. Thanks in advance. Markus

2 3

RE: Help with a native Windows test program needed (CreateFileW)
by Greg Geldorp Sept. 11, 2009

Sept. 11, 2009

Hi Markus, > From: Markus Stockhausen <markus.stockhausen(a)collogia.de> > > as I have no Windows machine and cannot test this: I'm in need for > someone who can build a small exe that gives further hints to my > observations in bug #9575. The following is needed: > > call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) > call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) > > Return the result of second call. I expect it to be a valid handle on > Windows whereas Wine will return an invalid handle. This test program: int main(int argc, char *argv[]) { HANDLE File1, File2; File1 = CreateFileW(L"test.dat", GENERIC_READ, 0, NULL, CREATE_ALWAYS, 0, NULL); File2 = CreateFileW(L"test.dat", GENERIC_READ, 0, NULL, OPEN_EXISTING, 0, NULL); printf("File1 0x%x File2 0x%x error %u\n", File1, File2, GetLastError()); return 0; } returns this result on Windows 2008: File1 0x10 File2 0xffffffff error 32 so the second CreateFile fails on Windows too, with ERROR_SHARING_VIOLATION. Ge.

1 0

RFC: registry keys of serial ports
by Stefan Leichter Sept. 11, 2009

Sept. 11, 2009

Hi, i like to know what modifikations the attached patch needs to go into git. More things need to be done in this area. I know of: - more registry keys [1] - remove the registry keys when a serial port disappears (USB) - the links in the dosdevice directory needs to be created Patches for the items above will be created, when the current patch went into git References: 1 http://www.winehq.org/pipermail/wine-patches/2009-April/071999.html - http://www.winehq.org/pipermail/wine-devel/2009-April/074476.html -- Stefan

3 5

Help with a native Windows test program needed (CreateFileW)
by Markus Stockhausen Sept. 11, 2009

Sept. 11, 2009

Hi, as I have no Windows machine and cannot test this: I'm in need for someone who can build a small exe that gives further hints to my observations in bug #9575. The following is needed: call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) call CreateFileW( 'testfile.txt', ..., dwShareMode = 0, ...) Return the result of second call. I expect it to be a valid handle on Windows whereas Wine will return an invalid handle. Thanks in advance. Markus

2 1

fixing ReadDirectoryChangesW()
by Michael Pujos Sept. 11, 2009

Sept. 11, 2009

A few months ago I wrote a patch that addressed the following issues with ReadDirectoryChangesW(): "1. move and rename of files and directories in a monitored directory are not notified 2. the current implementation allow to only get one change (FILE_NOTIFY_INFORMATION structure) at a time. No change is actually lost but the calling application using ReadDirectoryChangesW() expect to retrieve all changes since the last call. With the current implementation it would only get the first and would have to call ReadDirectoryChangesW() again to get the other. With this patch , it gets all notification (as many consecutive FILE_NOTIFY_INFORMATION structs) since the last call to ReadDirectoryChangesW(). These move operations are implemented by this patch and conform to what win32 does: - Moving a directory or a file inside the monitored directory. A FILE_NOTIFY_INFORMATION structure with action FILE_ACTION_ADDED is returned - Moving a directory or a file out of the monitored directory. A FILE_NOTIFY_INFORMATION structure with action FILE_ACTION_REMOVED is returned - Renaming a directory or a file in the monitored directory. Two consecutive FILE_NOTIFY_INFORMATION structures with actions FILE_ACTION_RENAMED_OLD_NAME and FILE_ACTION_RENAMED_NEW_NAME are returned within the same reply. This can actually be done because issue 2 above is corrected." This old patch can be found here: http://bubbleguuum.free.fr/0001-server-implement-move-rename-change-notific… This patch modify the server protocol in server_protocol.h from struct read_change_reply { struct reply_header __header; int action; /* VARARG(name,string); */ }; to: struct read_change_reply { struct reply_header __header; int action_count; /* VARARG(name,string); */ }; I'm willing to resubmit this patch with updated tests against current Wine, but before I procceed I need to be sure it won't be rejected simply because it changes the protocol.

1 0

Re: [PATCH 2/4] mshtml: Added http loading test.
by Paul Vriens Sept. 11, 2009

Sept. 11, 2009

On 09/02/2009 12:27 PM, Jacek Caban wrote: > --- > dlls/mshtml/tests/htmldoc.c | 270 > +++++++++++++++++++++++++++++-------------- > 1 files changed, 182 insertions(+), 88 deletions(-) > > > > ------------------------------------------------------------------------ > > Hi Jacek, Looks like this one introduced 2 test failures on some boxes, I couldn't find a common denominator though. The failures happen on IE6, IE7 and IE8 on different platforms (NT4 and higher). There are however also IE8 boxes without failures. The failures seem to be triggered by the "test_download(FALSE, FALSE, FALSE);" on line 4026. Could you have a look? (Some other failures happen on ME and lower but I'll investigate those). -- Cheers, Paul.

2 1