November 2005 - Wine-devel - list.winehq.org

RFC: Presentation for Desktop Architecture meeting, take two
by Dan Kegel 29 Nov '05

29 Nov '05

The presentation has to be really short. Essentially, everything we want to say has to fit in just two slides! Here's my attempt: http://kegel.com/wine/wine-osdl-da05.pdf Can the folks who posted issues have a look at the X.org and/or Mesa categories in https://bugs.freedesktop.org/ and send me links to the appropriate enhancement requests? (And if they're not there already, can you add them?) Dmitry Timoshkov wrote: > How about inability of X server to handle pixmaps larger > than 32767 x 32767 and an ugly workaround introduced by > the following patch: > ftp://ftp.x.org/pub/X11R6.8.2/patches/xorg-CAN-2005-2495.patch Lionel Ulmer wrote: > OK, here is my wish list from the X.org guys: > * relative mouse movement reporting. ... > * full support for resolution / depth switching. ... > * GL (GLX ?) extension which would remove the 'thread-affinity' from GL. > * GLX extension that would export the 'clip-list' functionalities of cards > * same point as before but for Xv the day we will re-add this code for > accelerated YUV display. > * have 'connection-level' configuration. ... > * have a non-connection limited mouse-grab. ... Thanks, Dan -- Why can't Johnny run Linux? See http://kegel.com/linux/comfort

1 0

Re: resend: http.c: redirect corrections.
by Robert Shearman 29 Nov '05

29 Nov '05

Aric Stewart wrote: >Index: dlls/wininet/http.c >=================================================================== >RCS file: /home/wine/wine/dlls/wininet/http.c,v >retrieving revision 1.113 >diff -u -r1.113 http.c >--- dlls/wininet/http.c 22 Nov 2005 14:53:30 -0000 1.113 >+++ dlls/wininet/http.c 22 Nov 2005 15:35:24 -0000 >@@ -629,6 +634,26 @@ > > /* We appear to do nothing with the buffer.. is that correct? */ > >+ if(!(lpwhr->hdr.dwFlags & INTERNET_FLAG_NO_AUTO_REDIRECT)) >+ { >+ DWORD dwCode,dwCodeLength=sizeof(DWORD),dwIndex=0; >+ if(HTTP_HttpQueryInfoW(lpwhr,HTTP_QUERY_FLAG_NUMBER|HTTP_QUERY_STATUS_CODE,&dwCode,&dwCodeLength,&dwIndex) && >+ (dwCode==302 || dwCode==301)) >+ { >+ WCHAR szNewLocation[2048]; >+ DWORD dwBufferSize=2048; >+ dwIndex=0; >+ if(HTTP_HttpQueryInfoW(lpwhr,HTTP_QUERY_LOCATION,szNewLocation,&dwBufferSize,&dwIndex)) >+ { >+ static const WCHAR szGET[] = { 'G','E','T', 0 }; >+ /* redirects are always GETs */ >+ HeapFree(GetProcessHeap(),0,lpwhr->lpszVerb); >+ lpwhr->lpszVerb = WININET_strdupW(szGET); >+ return HTTP_HandleRedirect(lpwhr, szNewLocation, NULL, 0, NULL, 0); >+ } >+ } >+ } >+ > TRACE("%i <--\n",rc); > return rc; > } > > This proves that HttpSendRequestEx and HttpSendRequest need to share more code. >@@ -1849,13 +1885,26 @@ > #endif > > HeapFree(GetProcessHeap(), 0, lpwhs->lpszServerName); >- lpwhs->lpszServerName = WININET_strdupW(hostName); >+ if (urlComponents.nPort != INTERNET_DEFAULT_HTTP_PORT && >+ urlComponents.nPort != INTERNET_DEFAULT_HTTPS_PORT) >+ { >+ int len; >+ static WCHAR fmt[] = {'%','s',':','%','i',0}; >+ len = lstrlenW(hostName); >+ len+=6; >+ lpwhs->lpszServerName = HeapAlloc(GetProcessHeap(),0,len*sizeof(WCHAR)); > > I submitted a patch to unbreak the Host header for proxies that means you should be changing lpwhs->lpszHostName now. >+ sprintfW(lpwhs->lpszServerName,fmt,hostName,urlComponents.nPort); >+ } >+ else >+ lpwhs->lpszServerName = WININET_strdupW(hostName); >+ >+ HTTP_ProcessHeader(lpwhr, g_szHost, lpwhs->lpszServerName, HTTP_ADDREQ_FLAG_ADD | HTTP_ADDREQ_FLAG_REPLACE | HTTP_ADDHDR_FLAG_REQ); >+ >+ > > And you should be sending lpwhs->lpszHostName. >@@ -1868,6 +1917,9 @@ > return FALSE; > } > >+ if (lstrlenW(extra)>0) >+ StrCatW(path,extra); > Why are you using a shlwapi string function? What's wrong with strcatW or lstrcatW? Also, this could easily cause a buffer overflow by a server sending a redirect with a large query. You need to check that there is enough space before blinding copying into the buffer. - Rob Shearman

1 0

Wine benchmarks
by Hans Kristian Rosbach 29 Nov '05

29 Nov '05

http://wiki.winehq.org/BenchMark is getting old, and an update using version 0.9 would be ideal. Any chance at getting an update using the same computer as before? A little before & after comparison would also be nice. -A page for Wine 0.9 vs Windows (like today) -Another for Wine 0.9 vs Wine 20050419 to see progress/regress -HK

3 6

Re: [Bug 3002] problems writting to serial ports on 2005-05-24 release
by James Hawkins 29 Nov '05

29 Nov '05

On 11/25/05, Fernando Chmielewsky <ferchimi(a)gmail.com> wrote: > sorry I ask you this in this way... > I once tracked a bug in wine using CVS, but now I can't even get to > download the cvs tree, doing exactly the sama I did in the past: > creating ~/.cvspass with the single line > :pserver:cvs(a)cvs.winehq.com:/home/wine Ah<Z > and ~/cvsrc file containing: > cvs -z 3 > update -PAd > diff -up > > and running: > cvs checkout wine > > and this just throws the following: > cvs checkout: No CVSROOT specified! Please use the `-d' option > cvs [checkout aborted]: or set the CVSROOT environment variable. > > and then, after setting export CVSROOT=~/wine, and creating ~wine/CVSROOT > says the following: > cvs checkout: cannot find module `wine' - ignored > > I know there's something wrong I am doing, only I do not know yet, > after all this time, what this things mean... > hope you can help me > You should check out the docs on winehq.org, they'll give better details then the folowing: your CVSROOT wasn't set up correctly: CVSROOT=:pserver:cvs(a)winehq.org:/home/wine -- James Hawkins

1 0

Re: [ntdll] fix GetProcAddress on mapped modules as data
by Vitaliy Margolen 28 Nov '05

28 Nov '05

Monday, November 28, 2005, 3:29:51 PM, Raphael wrote: > Hi, > fix crash on Corel 9 start > (http://bugs.winehq.org/show_bug.cgi?id=2884) > Changelog: > - don't crash on GetProcAddress when input mapped modules as data I think you need to make some tests of this. It shouldn't be to hard to load some dll as a resource and try to get proc address. Also check if you can do that for already loaded dll (like kernel32.dll). Now the patch: > + if ((ULONG_PTR)module & 1) > + { > + module = (HMODULE)((ULONG_PTR)module & ~1); > + } Check is some-what redundant. Just use: module = (HMODULE)((ULONG_PTR)module & ~1); > - void *proc = name ? find_named_export( module, exports, exp_size, name->Buffer, -1 ) > - : find_ordinal_export( module, exports, exp_size, ord - exports->Base ); > + void *proc = NULL; > + if (name) > + { > + proc = find_named_export( module, exports, exp_size, name->Buffer, -1 ); > + } > + else > + { > + proc = find_ordinal_export( module, exports, exp_size, ord - exports->Base ); > + } Any reason for this? It's the same no matter how you write it only 7 lines longer <g> Vitaliy Margolen

2 2

powerpoint: images upside down
by Cihan Altinay 28 Nov '05

28 Nov '05

Hi, When using MS Powerpoint 2003 images are drawn upside down in fullscreen (presentation) mode but not in design mode. Also, text is drawn correctly as seen on the attachments. I would appreciate a hint on what traces/logs to provide for more info as the dc & win output is quite long. Thanks, Cihan

3 4

[msvcrt] how to handle msvcr*.dll ?
by Raphael 28 Nov '05

28 Nov '05

Hi, since visual studio .net (2002,3,5) MS provides newer versions of msvcrt: - msvcr70.dll - msvcr71.dll - msvcr80.dll ... (and same equivalents for msvcp*.dll) how we will handle that on wine: - creating new dll forlders and using forwards to current msvcrt.dll or - creating only new .spec files on msvcrt folder ? Regards, Raphael

4 4

Re: Question about X11DRV_ChoosePixelFormat glXChooseFBConfig returns NULL (glError: 0)
by Daniel 28 Nov '05

28 Nov '05

On Sun, Nov 27, 2005 at 03:24:32PM +0000, L. Lenders wrote: > Hi Daniel, i opened a bug for this with requested > info: bug 3956. Thanks for the help btw! regards Louis Ok, I managed to catch up. So let's see: 1. You do have client GLX 1.3, but server GLX 1.2 Raphael: That amounts to GLX version 1.2, the lower version is important. 2. You have a different problem. The calls to glXChooseFBConfig return NULL because your card doesn't want to do the requested visual. If you look closely, you will see another call to glXChooseFBConfig that does not fail because your card wants to do this visual. 3. bug #3956 should be closed. Daniel

3 2

Window geometry problem
by Pavel Troller 28 Nov '05

28 Nov '05

Hi! I would like to report a long lasting problem with an application, which renders its main window incorrectly in wine. The app is "iRiver caption editor" and it can be freely downloaded at http://www.iriveramerica.com/download/ce.zip . Installation of the program is easy and proceeds and the program itself runs well, but its window looks like a total mess. For a snapshot, look at http://www.sinus.cz/~patrol/ircape.png . I think it's not necessary to write more, what's wrong, simply the widgets are misplaced, incorrectly sized, the window has scrollbars, which it IMHO shouldn't have etc. The window is not resizable. It's the only app known to me which exhibits such a behaviour. I don't have a possibility to test it on a real windoze system, I don't have it. I'm about to file a bug for it, but I'm asking first here to verify that it's not an easily resolvable problem with my setup etc., and also whether a similar (or the same) behaviour isn't already registered. With regards, Pavel Troller

3 2

Re: [DbgHelp]: SymGet{Next|Prev}Line64.
by Robert Shearman 28 Nov '05

28 Nov '05

Eric Pouech wrote: > /* dbghelp.c */ > extern struct process* process_find_by_handle(HANDLE hProcess); >+extern BOOL validate_addr64(DWORD64 addr); > extern HANDLE hMsvcrt; > extern BOOL validate_addr64(DWORD64 addr); > > validate_addr64 is duplicated. -- Rob Shearman

1 0