- Wine-Devel - list.winehq.org

./configure message
by dmitry serpokryl Feb. 28, 2006

Feb. 28, 2006

configure: WARNING: linux/capi.h: present but cannot be compiled configure: WARNING: linux/capi.h: check for missing prerequisite headers? configure: WARNING: linux/capi.h: see the Autoconf documentation configure: WARNING: linux/capi.h: section "Present But Cannot Be Compiled" configure: WARNING: linux/capi.h: proceeding with the preprocessor's result configure: WARNING: linux/capi.h: in the future, the compiler will take precedence configure: WARNING: ## ------------------------------------ ## configure: WARNING: ## Report this to wine-devel(a)winehq.org ## configure: WARNING: ## ------------------------------------ ## SuSE 10 x86-64, gcc-4.0.2 (release)

4 4

Re: [Dnsapi.dll] Implement Dnsapi.dll
by Mike McCormack Feb. 28, 2006

Feb. 28, 2006

MattK wrote: > This is a patch to implement Dnsapi.dll. It contains "bare bones". Please submit a seperate patch for the windns.h header file. > @@ -1193,6 +1193,7 @@ AC_CHECK_FUNCS(\ > strcasecmp \ > strerror \ > strncasecmp \ > + strsep \ Adding strsep here doesn't guarantee that it will exist. You need to check usage with #ifdef HAVE_STRSEP, or just assume it's always there and remove the configure check (prefered). Looks like that code is using tab indent too, so if this was needed, you should use the same to keep it consistent. > RCS file: /home/wine/wine/dlls/Makefile.in,v This is automatically generated. No patch is needed. > +DNS_STATUS WINAPI DnsValidateName_A(LPCSTR pszName, DNS_NAME_FORMAT); Hmmm. The Windows Platform SDK doesn't define this function as WINAPI. > +# Generated from /mnt/windows/windows/system32/dnsapi.dll by winedump This comment is unnecessary. > +107 stdcall DnsValidateName_A(str long) Maybe cdecl instead, if the Platform SDK header is right? > +EXTRALIBS = -lc Pretty sure libc is linked by default, so you don't need to add it. > + /* The Platform SDK documentation for this says that it ignores all periods. */ > + /* So there's gonna be no periods! */ > + > + if(dots1 && dots2) > + { > + strsep(&name1,dots1); > + strsep(&name2,dots2); > + pName1 = name1; > + pName2 = name2; > + } This ignores the first period and everything after it. It also modifies the input (ie. the program's copy of the string). > + equal = strcmp(pName1,pName2); Probably better to use lstrcmpA to use the Wine codepage. > + if(equal == 0) > + { > + return TRUE; > + } > + > + if(equal > 0 || equal < 0) > + { > + return FALSE; > + } > +} How about "return !equal;" ? It would be nice if you could write a test case for DnsNameCompare_A. If you do, please submit it seperately from the rest of the patch. Thanks for the patch! Mike

1 0

Re: [PATCH 06/12] [WineDbg]: rewrite auto mode
by Alexandre Julliard Feb. 27, 2006

Feb. 27, 2006

Eric Pouech <eric.pouech(a)wanadoo.fr> writes: > @@ -880,7 +861,14 @@ enum dbg_start dbg_active_launch(int > */ > enum dbg_start dbg_active_auto(int argc, char* argv[]) > { > + HANDLE hFile; > + > argc--; argv++; > - dbg_action_mode = automatic_mode; > - return dbg_active_attach(argc, argv); > + hFile = parser_generate_command_file("echo Modules:", "info share", > + "echo Threads:", "info threads", > + NULL); > + if (hFile == INVALID_HANDLE_VALUE || !dbg_active_attach(argc, argv)) > + return FALSE; > + dbg_main_loop(hFile); > + return TRUE; > } This completely breaks auto mode, because dbg_active_attach returns an enum dbg_start, not a boolean. There's the same mistake in the next patch too. -- Alexandre Julliard julliard(a)winehq.org

2 1

CRYPT32: iod.c compile problem (gcc 2.95)
by Vitaly Lipatov Feb. 27, 2006

Feb. 27, 2006

I try compile last version with gcc 2.95 and get it failed: $ make gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_CRYPT32_ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -gstabs+ -Wpointer-arith -O2 -march=i586 -o oid.o oid.c oid.c: In function `init_oid_info': oid.c:940: structure has no member named `Algid' oid.c:969: structure has no member named `Algid' oid.c: In function `CryptFindOIDInfo': oid.c:1037: structure has no member named `Algid' oid.c:1089: structure has no member named `Algid' oid.c: In function `CertOIDToAlgId': oid.c:1126: structure has no member named `Algid' There is unnamed union, and such construction does no recognized old the compiler. What the best way to fix it? -- Vitaly Lipatov, ALT Linux Team Russia, Saint-Petersburg, www.etersoft.ru

2 1

Early warning: kernel address space change
by Andreas Mohr Feb. 27, 2006

Feb. 27, 2006

Hi, FYI (just in case it happens to affect Wine): http://lkml.org/lkml/2006/2/27/159 Andreas

2 1

[Dnsapi.dll] Implement Dnsapi.dll
by MattK Feb. 27, 2006

Feb. 27, 2006

This is a patch to implement Dnsapi.dll. It contains "bare bones". This was done mainly to help with Bug 4323 <http://bugs.winehq.org/show_bug.cgi?id=4323>. This patch has the DllMain entry point, a function and a stub, header files, implementation into the tree, and a .spec file. Please tell me how to correct anything if there is a problem.

1 0

DirectX 8 test: Gothic 2 on wine
by Olaf Leidinger Feb. 27, 2006

Feb. 27, 2006

Hello List! Gothic2 seems to start working with wine. Well - as long as you disable sound in system/gothic.ini. There are some issues with DirectDraw/Direct3D I'd like to report (not sure if they are known): - most textures are missing - everything is really dark (I have to set gamma to 2.5 (via xgamma -gamma 2.5) ) - fonts are missing/crippled I took some pictures and put it together with a log on a simple webpage. Please have a look at http://leidola.newcon.de/gothic2 Thanks a lot for your attention, Ciao, Olaf Leidinger

5 9

Re: ddraw.dll
by Segin Feb. 26, 2006

Feb. 26, 2006

Alexander N. Sørnes wrote: > Hello, > > I have found that some Westwood installers, such as the ones for Nox, > C&C Tiberian Sun, C&C Yuri's Revenge and C&C Tiberain Sun: Firestorm > require ddraw.dll to be present in the windows/system 32 directory; if > not, they give a message saying they failed to initialize DirectX. > Can someone submit a patch so this file is created by > wineprefixcreate? ddraw.dll is present system directories of both > Windows 98SE and Windows XP. > > > Regards, > > Alexander N. Sørnes > > > I wouldn't know why it does this, but a highly likely explanation is that the installers dlopen() C:\WINDOWS\DDRAW.DLL using an absolute or semi-absolute (flexible Windows directory, in case Windows is installed on D: or even in "D:\WINNT\") path. It probably also tries to stat() thew file before opening it so that the lack of the file causes an error before it is dlopen()d. A good suggestion is if a DLL file is stat()d in the c:\windows\system(32) directory in Wine, Wine should check it against any .so files in $INSTALLPREFIX/lib/wine, and claim the file exist if a compatable .dll.so is found. Wine should also 'remember' this mapping in case of fopen() or compatable calls.

1 0

Windows Device driver for NTFS
by Gerold J. Wucherpfennig Feb. 25, 2006

Feb. 25, 2006

Can anybody help me to code a little driver for a virtual hard disk drive for examination of undocumented NTFS filesystem stuctures i.e. the $LOGFILE (journaling of NTFS)? The data should be written to e.g. C:\image.bin, filesize e.g. 50MB which will correspond to a drive e.g. Z:, drive size 50MB. After every write the driver should be paused to examine the C:\image.bin to get the undocumented structure of the file $LOGFILE? Can anybody help me? I'm a little confused by the windows DDK documentation and tutorials found on the internet. I've already downloaded the windows DDK. Please CC me and excuse me for crossposting. Regards, Gerold

2 1

Re: [PATCH] please commit this FCI work
by Detlef Riekenberg Feb. 25, 2006

Feb. 25, 2006

Am Freitag, den 24.02.2006, 14:28 +0100 schrieb Gerold J. Wucherpfennig: > Please commit. I have no knowledge about fci, but some code is strange: +#define fci_set_error(A,B,C) \ + p_fci_internal->perf->erfOper = A; \ + p_fci_internal->perf->erfType = B; \ + p_fci_internal->perf->fError = C; + You use SetLastError() for B almost everywhere following the Macro. Why not add this statement to the Macro? Exceptions where you did not use SetLastError(): Macro-Parameter B is 0 - Did you mean NO_ERROR / ERROR_SUCCESS? - Does Windows use a SetLastError(NO_ERROR) in this case? (Tests for cabinet.dll are only in one file: "extract.c". can you add tests for fci/compress?) When windows does a SetLastError(NO_ERROR): append "SetLastError(B)" to the Macro When windows does not do a SetLastError(NO_ERROR): append "if (B) SetLastError(B)" to the Macro + /* write error */ + fci_set_error( FCIERR_TEMP_FILE, ERROR_WRITE_FAULT, TRUE ); + p_fci_internal->perf->erfType = ERROR_WRITE_FAULT; + p_fci_internal->perf->fError = TRUE; + SetLastError(ERROR_WRITE_FAULT); the two p_pci_* - lines are duplicates (already done in the Macro) btw: What should be added to the changelog? cabinet/fci: cleanup error-handling with a MACRO -- By By ... ... Detlef

2 1