- Wine-Devel - list.winehq.org

Re: [PATCH] wininet.dll compile time defines
by Vincent Béron Nov. 25, 2003

Nov. 25, 2003

Le mar 25/11/2003 à 00:37, Alex Tribble a écrit : > Hi, > When upgrading to Fedora Core 1 (the new RedHat release, sort of), > wine stopped compiling for me. I was getting errors while parsing > <openssl/ssl.h>, et al. -DOPENSSL_NO_KRB5 makes these go away. > > Any thoughts? Please CC me, as I'm not subscribed to the devel list. The patch http://cvs.winehq.com/patch.py?id=10182 (applied yesterday in CVS) fixes the issue. Vincent

1 0

Re: Mach-O Support for imageSize in NtHeader
by Alexandre Julliard Nov. 25, 2003

Nov. 25, 2003

"Pierre d'Herbemont" <stegefin(a)free.fr> writes: > Here is my second try for a replacement of the _end symbol. I think > that since there is no support of it we need a kind of emulation > function, which I thought should be located in the wine_port lib. So I > add a mach-o.c port file, with the needed function. But should it be > done this way? No, the static char _end[] thing should be enough, you don't have to store anything in there (which won't do what you want anyway). The only thing that matters is the address of the _end symbol. -- Alexandre Julliard julliard(a)winehq.com

2 1

unicows update and PE memory format
by Shachar Shemesh Nov. 25, 2003

Nov. 25, 2003

Hi list, Here is a quick update. It seems that I won't be able to use the forwarding inside the spec file. I'm not sure exactly why that is, but I think the unicows.lib link time library tries to load stuff by doing "GetModuleHandle", and then fetching relative values from that place. Is there a plausable assumption about PE memory layout? Is this something that Win32 properly defines? Do we have such a format for winelib dlls? Shachar -- Shachar Shemesh Open Source integration & consulting Home page & resume - http://www.shemesh.biz/

3 5

RFH: *ReAlloc* mess
by Dimitrie O. Paun Nov. 25, 2003

Nov. 25, 2003

Hi folks, I need some help to figure out how the following realloc functions behave: GlobalReAlloc LocalReAlloc Local32ReAlloc16 NetApiBufferReallocate SysReAllocStringLen So, I need to know: -- do they allocate memory when a NULL is passed in? -- do they free memory when the size is passed in as 0? Help with this matter is much appreciated! -- Dimi.

1 0

Re: Update comctl spec
by Alexandre Julliard Nov. 24, 2003

Nov. 24, 2003

Hans Leidekker <hans(a)it.vu.nl> writes: > Changelog: > - Export more functions by name. This will break the import libraries. Why do you want to do that? -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: Compile Bug: dlls/wininet/cookie.c .. and its cause
by Alexandre Julliard Nov. 24, 2003

Nov. 24, 2003

Ivan Gyurdiev <ivg2(a)cornell.edu> writes: > I have no clue what the FAR macro does, but wouldn't it be better to > just rename the wine FAR macro to something different if you dont want > it to conflict w/ the other FAR macro? It's a Windows thing, we can't change it. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: heap: fix RtlHeapReAllocate()
by Dimitrie O. Paun Nov. 24, 2003

Nov. 24, 2003

On November 21, 2003 03:21 pm, Eric Pouech wrote: > this will not compile on Win9x. You have to take care of: Good point, sorry about that, I was being lazy. To force people to do the dyn load, we shouldn't import ntdll, no? ChangeLog RtlHeapReAllocate() should not allocate memory. Add small test to make sure it doesn't return. Don't import ntdll for tests, we need to load it dynamically. Index: dlls/ntdll/heap.c =================================================================== RCS file: /var/cvs/wine/dlls/ntdll/heap.c,v retrieving revision 1.25 diff -u -r1.25 heap.c --- dlls/ntdll/heap.c 10 Oct 2003 00:12:17 -0000 1.25 +++ dlls/ntdll/heap.c 20 Nov 2003 21:00:39 -0000 @@ -1231,11 +1231,11 @@ HEAP *heapPtr; SUBHEAP *subheap; - if (!ptr) return RtlAllocateHeap( heap, flags, size ); /* FIXME: correct? */ + if (!ptr) return NULL; if (!(heapPtr = HEAP_GetPtr( heap ))) { set_status( STATUS_INVALID_HANDLE ); - return FALSE; + return NULL; } /* Validate the parameters */ Index: dlls/ntdll/tests/Makefile.in =================================================================== RCS file: /var/cvs/wine/dlls/ntdll/tests/Makefile.in,v retrieving revision 1.9 diff -u -r1.9 Makefile.in --- dlls/ntdll/tests/Makefile.in 8 May 2003 03:47:24 -0000 1.9 +++ dlls/ntdll/tests/Makefile.in 24 Nov 2003 07:19:04 -0000 @@ -3,12 +3,12 @@ SRCDIR = @srcdir@ VPATH = @srcdir@ TESTDLL = ntdll.dll -IMPORTS = ntdll CTESTS = \ env.c \ error.c \ generated.c \ + heap.c \ large_int.c \ path.c \ rtl.c \ --- /dev/null 2003-01-30 05:24:37.000000000 -0500 +++ dlls/ntdll/tests/heap.c 2003-11-24 02:17:07.000000000 -0500 @@ -0,0 +1,49 @@ +/* + * Unit test suite for heap functions + * + * Copyright 2003 Dimitrie O. Paun + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <stdarg.h> +#include <stdlib.h> + +#include "ntstatus.h" +#include "windef.h" +#include "winbase.h" +#include "wine/test.h" +#include "winnt.h" +#include "winnls.h" +#include "winreg.h" +#include "winternl.h" + +PVOID (WINAPI *pRtlReAllocateHeap)(HANDLE,ULONG,PVOID,ULONG); + +static void test_realloc( void ) +{ + void *mem = NULL; + + mem = pRtlReAllocateHeap(GetProcessHeap(), 0, mem, 10); + ok(mem == NULL, "memory allocated"); +} + +START_TEST(heap) +{ + HMODULE mod = GetModuleHandle("ntdll.dll"); + if (!mod) return; + pRtlReAllocateHeap = (void *)GetProcAddress(mod, "RtlReAllocateHeap"); + if (pRtlReAllocateHeap) test_realloc(); +} -- Dimi.

3 3

Compile Bug: dlls/wininet/cookie.c .. and its cause
by Ivan Gyurdiev Nov. 24, 2003

Nov. 24, 2003

Macro __WINESRC__ is defined for dlls/wininet/cookie.c in the Makefile. The file cookie.c includes wine/debug.h. wine/debug.h includes windef.h. windef.h defines the macro FAR to __ONLY_IN_WINELIB() __ONLY_IN_WINELIB() defines to do_not_use_this_in_wine if __WINESRC__ The file dlls/wininet/cookie.c also includes dlls/wininet/internet.h dlls/wininet/internet.h includes openssl/ssl.h for my machine openssl/ssl.h includes openssl/kssl.h openssl/kssl.h has the following code using the FAR macro: #ifdef KRB5_HEIMDAL typedef unsigned char krb5_octet; #endif #ifndef FAR #define FAR #endif ... Line 136: krb5_octet FAR *key; This is on a Fedora core system, which applies a patch to openssl that changes the original openssl-devel regarding this, but I think the issue's with wine, and is still there. The original openssl-devel looks like this: #ifdef KRB5_HEIMDAL typedef unsigned char krb5_octet; define FAR #endif Either way, the point is the FAR macro is being used in kssl.h, and it's defined in windef.h, as do_not_use_this_in_wine, resulting in this: gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_WINX32_ -D_REENTRANT -fPIC -Wall -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o cookie.o cookie.c In file included from /usr/include/openssl/ssl.h:179, from internet.h:36, from cookie.c:39: /usr/include/openssl/kssl.h:136: error: syntax error before '*' token /usr/include/openssl/kssl.h:149: error: syntax error before '*' token /usr/include/openssl/kssl.h:150: error: syntax error before '*' token /usr/include/openssl/kssl.h:151: error: syntax error before '*' token /usr/include/openssl/kssl.h:151: error: syntax error before '*' token /usr/include/openssl/kssl.h:152: error: syntax error before '*' token /usr/include/openssl/kssl.h:153: error: syntax error before '*' token /usr/include/openssl/kssl.h:155: error: syntax error before '*' token /usr/include/openssl/kssl.h:157: error: syntax error before '*' token /usr/include/openssl/kssl.h:159: error: syntax error before '*' token /usr/include/openssl/kssl.h:167: error: syntax error before '*' token In file included from internet.h:36, from cookie.c:39: /usr/include/openssl/ssl.h:909: error: syntax error before "KSSL_CTX" /usr/include/openssl/ssl.h:931: error: syntax error before '}' token make[2]: *** [cookie.o] Error 1 make[2]: Leaving directory `/home/phantom/tmp/firebird-downloads/wine-20031118/dlls/wininet' make[1]: *** [wininet] Error 2 make[1]: Leaving directory `/home/phantom/tmp/firebird-downloads/wine-20031118/dlls' make: *** [dlls] Error 2 ===================================================================== I am not subscribed to this list. If you'd like to email back write to ivg2(a)cornell.edu. Also, some unrelated questions: does anyone know what causes this: err:module:NE_LoadBuiltinModule loaded .so but dll winealsa.drv still not found (the file is there... oss is there too, and works great) ....or why keyboard focus is always lost (characters appear in the console, kb doesn't work in application) when playing most games under a window manager like gnome or kde, but not under twm? This bug has been there forever and is really annoying.

3 2

Re: winmm: move MCI settings from system.ini to the registry
by Andreas Mohr Nov. 24, 2003

Nov. 24, 2003

Hi, On Mon, Nov 24, 2003 at 12:10:20PM -0500, Dimitrie O. Paun wrote: > On November 22, 2003 03:02 am, Eric Pouech wrote: > > I won't have too much time to look at this in details right now, so if > > someone wants to jump in, I'll be happy to give him (her ?) a hand. > > To be honest, > My interest in doing this work is to get rid of system.ini from the > Wine tree. While I agree that there's more to do, it's just not up > my alley :) So, this version of the patch still maintains support > for system.ini, just like the old code does, but it also adds support > for registry, which should allow us to move stuff there. To be honest, I'm not sure why one would want that. After all the system.ini settings are an essential part of a Windows environment, and removing the whole file will break many (admittedly older) programs, if I'm not mistaken... (nevermind the duplication of the parameters, which does exist) Andreas Mohr

1 0

Re: documentation/samples/generic.ppd
by Marcus Meissner Nov. 24, 2003

Nov. 24, 2003

On Mon, Nov 24, 2003 at 02:32:49AM -0500, Dimitrie O. Paun wrote: > > I can't help but feel that this file does not belong in the > Wine tree, but in the print system packages (cups, etc.). > > cvs rm -f documentation/samples/generic.ppd > > ChangeLog > Remove generic.ppd, it does not belong in the Wine tree. This is the generic fallback PPD file which can be used LPR style printing, where we do not get access to a PPD easily. Ciao, Marcus

2 1