- Wine-Devel - list.winehq.org

Re: excessive stack space usage in NTDLL
by Marcus Meissner Oct. 3, 2005

Oct. 3, 2005

On Mon, Oct 03, 2005 at 02:36:55PM +0200, Alexandre Julliard wrote: > Marcus Meissner <marcus(a)jet.franken.de> writes: > > > No recursion as far as I can see. I get this backtrace: > > > > Program received signal SIGSEGV, Segmentation fault. > > [Switching to Thread 1469332400 (LWP 24408)] > > 0x5577e57d in wine_cp_wcstombs (table=0x5586e668, flags=0, src=0x57845114, srclen=Cannot access memory at address 0x57845078 > > ) at wctomb.c:181 > > 181 case 7: dst[-7] = uni2cp_low[uni2cp_high[src[-7] >> 8] + (src[-7] & 0xff)]; > > (gdb) bt > > #0 0x5577e57d in wine_cp_wcstombs (table=0x5586e668, flags=0, src=0x57845114, srclen=Cannot access memory at address 0x57845078 > > ) at wctomb.c:181 > > #1 0x55725df1 in RtlUnicodeToOemN (dst=0x57845188 "#", dstlen=7, reslen=0x0, src=0x57845114, srclen=14) at rtlstr.c:845 > > #2 0x557271f5 in RtlUpcaseUnicodeStringToCountedOemString (oem=0x57845194, uni=0x57845250, doalloc=Cannot access memory at address 0x57845113 > > ) at rtlstr.c:1127 > > #3 0x5571964c in RtlIsNameLegalDOS8Dot3 (unicode=0x57845250, oem=0x0, spaces=0x5784524f "W\016") at path.c:909 > > #4 0x55705055 in wine_nt_to_unix_file_name (nameW=0x578458f4, unix_name_ret=0x57845830, disposition=1, check_case=Cannot access memory at address 0x5784524e > > ) at directory.c:1162 > > #5 0x55709ccd in NtQueryFullAttributesFile (attr=0x578458dc, info=0x57845854) at file.c:1388 > > #6 0x55709ecc in NtQueryAttributesFile (attr=0x578458dc, info=0x578458b4) at file.c:1436 > > #7 0x55a266bb in GetFileAttributesW (name=0x57842c00) at file.c:1772 > > #8 0x55a26756 in GetFileAttributesA (name=0x57c7fec8 "C:\\Windows\\Temp\\ACDLABS 8.0 (FREE)") at file.c:1793 > > > > (gdb) x /i $eip > > 0x5577e57d <wine_cp_wcstombs+2957>: mov %edx,0xffffff40(%ebp) > > (gdb) (gdb) print $ebp-0xc0 > > $4 = (void *) 0x57844ff8 > > (gdb) > > > > cat /proc/../maps excerpt: > > 57842000-57844000 rwxp 57842000 00:00 0 > > 57844000-57845000 ---p 57844000 00:00 0 > > 57845000-57944000 rwxp 57845000 00:00 0 (1MB stack) > > 579c5000-57a1c000 rwxp 579c5000 00:00 0 > > 57a20000-57c80000 rwxp 57a20000 00:00 0 > > > > It just underflowed the 1MB stackspace here. > > > > But the above backtrace only has 2kb of stackspace usage, true. > > Something else must be using the rest of the stack, unless the initial > stack pointer was set wrong, which seems unlikely. Probably the app > was not compiled with frame pointer and gdb can't get a backtrace. > Does winedbg show anything different? I guess the app itself uses so much stuff. There is a lot more backtrace below, but just hexcode. I stopped scrolling after 20000 entries. Can I do winedbg do follow CreateProcess()es? Or attach it to running wines by just knowing the PID? Ciao, Marcus

2 1

Stop-ship 0.9 problem
by Mike Hearn Oct. 3, 2005

Oct. 3, 2005

There is a serious problem with Wine 0.9 as-is, namely that we now respect Windows antialiasing settings. If you have a patented bytecode hinter enabled FreeType, things look OK: http://plan99.net/~mike/files/hinted-fonts.jpg but if you don't (like 99% of Linux users): http://www.republika.pl/belegdol/temp/wine.png Needless to say we can't ship 0.9 when it'll look like that for most users. The attached patch fixes the issue by providing an override for antialiasing. FreeType AA makes the fonts look good again at small sizes. It also changes the logic on one line - I may be mistaken but it looked wrong to my eye: - if(!get_gasp_flags(physDev, &flags) || flags & GASP_DOGRAY) + if(get_gasp_flags(physDev, &flags) && (flags & GASP_DOGRAY)) We can patch this variable in Crossover (where CW ship a licensed freetype copy) so the correct AA settings are used which (a) looks nicer and (b) makes font metrics correct. A better long term solution suggested by Krzysztof Foltman might be to make bitmapped copies of a hinted Tahoma at small sizes, for instance. thanks -mike Index: dlls/x11drv/xrender.c =================================================================== RCS file: /home/wine/wine/dlls/x11drv/xrender.c,v retrieving revision 1.74 diff -u -p -d -r1.74 xrender.c --- dlls/x11drv/xrender.c 23 Sep 2005 10:05:54 -0000 1.74 +++ dlls/x11drv/xrender.c 1 Oct 2005 22:48:10 -0000 @@ -100,6 +100,12 @@ static INT mru = -1; static int antialias = 1; +/* we need this for WineHQ because antialiasing is necessary to cover + * up for deficiencies in the FreeType auto-hinter (a pox upon Apple + * for not releasing the patent royalty free. It expires sometime + * around 2009) */ +static BOOL force_always_antialias = TRUE; + /* some default values just in case */ #ifndef SONAME_LIBX11 #define SONAME_LIBX11 "libX11.so" @@ -442,7 +448,7 @@ static BOOL get_gasp_flags(X11DRV_PDEVIC GetTextMetricsW(physDev->hdc, &tm); ppem = abs(X11DRV_YWStoDS(physDev, tm.tmAscent + tm.tmDescent - tm.tmInternalLeading)); - gasp++; + gasp++; /* skip version field */ num_recs = get_be_word(*gasp); gasp++; while(num_recs--) @@ -474,15 +480,25 @@ static int GetCacheEntry(X11DRV_PDEVICE assert( !entry->format[format] ); } - if(antialias && plfsz->lf.lfQuality != NONANTIALIASED_QUALITY) + if (force_always_antialias) { - if(!get_gasp_flags(physDev, &flags) || flags & GASP_DOGRAY) - entry->aa_default = AA_Grey; + entry->aa_default = AA_Grey; + } + else + { + if ((antialias && plfsz->lf.lfQuality != NONANTIALIASED_QUALITY)) + { + if(get_gasp_flags(physDev, &flags) && (flags & GASP_DOGRAY)) + entry->aa_default = AA_Grey; + else + entry->aa_default = AA_None; + } else + { entry->aa_default = AA_None; + } } - else - entry->aa_default = AA_None; + return ret; }

8 13

Trouble compiling tests standalone with cl
by Dan Kegel Oct. 3, 2005

Oct. 3, 2005

While looking at http://bugs.winehq.org/show_bug.cgi?id=1899 I noticed that dlls/msvcrt/tests/file.c didn't test O_APPEND, so I thought I'd try building that test standalone, running it under Windows, and adding the missing test logic. (Rather than building under Windows, I used a shell script that invokes cl.exe -- see http://kegel.com/wine/cl-howto.html -- but that shouldn't matter here. Just to be clear about what version of cl I'm using, cl -? shows: Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077 for 80x86 Copyright (C) Microsoft Corporation 1984-2002. All rights reserved I also have the Platform SDK installed, so I have MS's headers, e.g. /home/dank/.wine/drive_c/Program Files/MicrosoftSDK/include/WinNT.h and my INCLUDE environment variable is set to C:\Program Files\msvc2003\include;C:\Program Files\MicrosoftSDK\include This used to work fine.) To compile a test standalone, one should be able to do e.g. cd dlls/msvcrt/tests cl -DSTANDALONE -D_X86_ -I../../../include file.c The first problem I ran into is that some wine .h files use wine-specific processor defines, so compilation fails with e.g. ../../../include\basetsd.h(177) : fatal error C1189: #error : Unknown CPU architecture! OK, I can live with one extra define, I guess, so I tried cl -DSTANDALONE -D_X86_ -D__i386__ -I../../../include file.c But that fails with the errors ../../../include\winnt.h(1680) : error C2054: expected '(' to follow 'inline' ../../../include\winnt.h(1681) : error C2085: 'NtCurrentTeb' : not in formal parameter list ... ../../../include\winbase.h(2271) : error C2054: expected '(' to follow 'inline' Evidently cl doesn't like the inline functions in winnt.h and winbase.h. Any suggestions? It seems a bit strange to be using wine's winnt.h when I have MS's headers, but I haven't tracked that down yet. Should Wine's headers build with cl? Thanks, Dan

3 8

excessive stack space usage in NTDLL
by Marcus Meissner Oct. 3, 2005

Oct. 3, 2005

Hi, The stackspace usage in dlls/ntdll/directory.c is affecting programs. One installer crashes because of it I suspect. objdump -d directory.o|grep "sub.*[0-9a-f][0-9a-f][0-9a-f],%esp" wine_nt_to_unix_file_name - 0x5f0 (1520 byte) (in this my sample program crashes) append_entry - 0x2cc (716 byte) NtQueryDirectoryFile - 0x22d4 (8916 byte!) problematic also cdrom.o DVD_ReadKey - 0x84c CDROM_DeviceIoControl - 0x40c thread.o start_thread - 0x830 nt.o NtQuerySystemInformation - 0x990 How best to reduce? Allocate space with RtlAllocateHeap()? Ciao, Marcus

2 1

Re: enumerate the substorage transforms
by Dmitry Timoshkov Oct. 3, 2005

Oct. 3, 2005

"Mike McCormack" <mike(a)codeweavers.com> wrote: > --- dlls/msi/action.c 30 Sep 2005 10:32:41 -0000 1.211 > +++ dlls/msi/action.c 30 Sep 2005 11:33:20 -0000 > @@ -46,6 +46,8 @@ http://msdn.microsoft.com/library/defaul > #define REG_PROGRESS_VALUE 13200 > #define COMPONENT_PROGRESS_VALUE 24000 > > +#define LPCTSTR LPCWSTR Do you really need this define? It looks like it's unused (not mentioning that it breaks Wine rules on using unicode APIs). -- Dmitry.

3 2

re: Release plans
by Dan Kegel Oct. 3, 2005

Oct. 3, 2005

Alexandre wrote: > Bugzilla has had a good cleanup lately (thanks guys!) and most of the > irrelevant bugs have been closed, so please have a look at the > remaining ones to see if there's anything you know how to fix. I'll be updating the page http://kegel.com/wine/qa/ with bugzilla statistics State2005.09.29 Unconfirmed 836 New 403 Assigned 68 Resolved 421 Closed 1476

2 6

Single sign-on SoC project summary
by Kai Blin Oct. 2, 2005

Oct. 2, 2005

Hi folks, Dimi asked me to give a little status about the single sing-on summer of code project I was doing to the list, and I'm happy to oblige. Applications running on win32 can use the Security Support Provider Interface (SSPI) to handle authentication via a number of protocols, like kerberos, NTLM, Negotiate and Schannel. My project was to implement the NTLM and Negotiate protocols, using Andrew Bartlett's GENSEC implementation from samba4. Samba offers a command line tool to provide NTLM ( and, in samba 4, Negotiate ) style authentication for other programs, so my task was to get wine to use ntlm_auth. Things that got done: --------------------- * Base64 codec ntlm_auth was first designed as a helper for squid. Squid talks to it's helpers using base64, wine just had an encoder for it. * Code to run ntlm_auth ntlm_auth doesn't store all the state information needed for the protocol handshake, so some wrapping code was needed. * Code that implements the security support providers for NTLM and Negotiate. To make the sspi functions behave properly, the api functions have to do the right thing. Work still left to do: ---------------------- * Getting the code into cvs: While most of the code was already submitted to the wine-patches list, a lot of it was still in need of improvements. I'm currently working on this. * Getting more of the stubs to actually do something While the code that exists so far can handle the basic NTLM and Negotiate authentication, a lot of functions defined in the API still are stubs. Conclusion: ----------- As soon as the code is in cvs, wine should be able to do basic NTLM and Negotiate authentication, server and client side. Client side just requires ntlm_auth, server side also requires samba to be set up correctly, and when using samba 3, access to the winbindd named pipe for the user running wine. Real use of this code seems limited, as most use cases also use the schannel protocol, which isn't implmented yet. Juan Lang is working on this, but needs to get more of crypt32 implemented for that. I will continue to work on this, there's still a lot of providers to go. I would like to use this summary to extend my thanks to Google and Chris DiBona for getting the whole Summer of Code set up. Also, I'd like to thank you guys, particularly Juan, for the help I have recieved so far. It's really fun to work with the project, I hope to keep doing so for a while. Kai -- Kai Blin, (blin at gmx dot net) "Of course power tools and alcohol don't mix. Everyone knows power tools aren't soluble in alcohol..." -- Crazy Nigel

2 1

Re: Release plans
by Tony Lambregts Oct. 2, 2005

Oct. 2, 2005

Alexandre Julliard wrote: > Folks, > > I just released 20050930, this should be considered the pre-0.9 > release, so please give it some good testing. In particular, please > test the things that new users will encounter first, like the > automatic .wine creation and winecfg. > This is fantastic. I have a question though what should I call this release in bugzilla? 20050930 20050930 pre beta 0.9.0 0.9.0 pre beta (20050930) other suggestions???? Obviously we should decide now how our release nameing structure is going to be called and I did not want to just put it in without some feedback... -- Tony Lambregts

2 1

Unable to compile CVS
by Stephen Torri Oct. 2, 2005

Oct. 2, 2005

I am having difficulty in compiling wine CVS. I synced up with the repository this afternoon. The error I am getting is: make[2]: Entering directory `/home/storri/src/wine/dlls/avifil32' ../../tools/winegcc/winegcc -B../../tools/winebuild -shared ./avifil32.spec avifile.spec.o acmstream.o api.o avifile.o editstream.o extrachunk.o factory.o getframe.o icmstream.o regsvr.o tmpfile.o wavfile.o rsrc.res -o avifil32.dll.so -L../../dlls -L../../dlls/msacm32 -L../../dlls/msvfw32 -L../../dlls/winmm -L../../dlls/ole32 -L../../dlls/user32 -L../../dlls/advapi32 -L../../dlls/kernel32 -L../../dlls/ntdll -lmsacm32 -lmsvfw32 -lwinmm -lole32 -luser32 -ladvapi32 -lkernel32 -lntdll -L../../libs/wine -lwine -luuid -L../../libs/port -lwine_port avifile.spec.o(.data.rel.ro+0x600): In function `__wine_avifile_dll_CallFrom16_p_long_lp': /home/storri/src/wine/dlls/avifil32/avifile.spec.c:136: undefined reference to `__wine_call_from_16_long' avifile.spec.o(.data.rel.ro+0x618):/home/storri/src/wine/dlls/avifil32/avifile.spec.c:136: undefined reference to `__wine_call_from_16_long' avifile.spec.o(.data.rel.ro+0x630): In function `__wine_avifile_dll_CallFrom16_p_long_lpl': /home/storri/src/wine/dlls/avifil32/avifile.spec.c:143: undefined reference to `__wine_call_from_16_long' avifile.spec.o(.data.rel.ro+0x648):/home/storri/src/wine/dlls/avifil32/avifile.spec.c:143: undefined reference to `__wine_call_from_16_long' avifile.spec.o(.data.rel.ro+0x660):/home/storri/src/wine/dlls/avifil32/avifile.spec.c:144: undefined reference to `__wine_call_from_16_long' avifile.spec.o(.data.rel.ro+0x678):/home/storri/src/wine/dlls/avifil32/avifile.spec.c:144: more undefined references to `__wine_call_from_16_long' follow collect2: ld returned 1 exit status winegcc: gcc failed. make[2]: *** [avifil32.dll.so] Error 2 make[2]: Leaving directory `/home/storri/src/wine/dlls/avifil32' make[1]: *** [avifil32] Error 2 make[1]: Leaving directory `/home/storri/src/wine/dlls' make: *** [dlls] Error 2 Stephen

3 3

Re: shlwapi: stub IUnknown_CPContainerInvokeParam
by Vitaliy Margolen Oct. 2, 2005

Oct. 2, 2005

Please ignore this patch it is not correct. > Vitaliy Margolen > changelog: > shlwapi > -stub IUnknown_CPContainerInvokeParam

1 0