October 2002 - Wine-Devel - list.winehq.org

LVM_* defs
by Dimitrie O. Paun Oct. 21, 2002

Oct. 21, 2002

Can anyone help me with the definition of these listview messages? TIA. LVM_CANCELEDITLABEL LVM_CREATEDRAGIMAGE LVM_ENABLEGROUPVIEW LVM_GETBKIMAGE LVM_GETGROUPINFO LVM_GETGROUPMETRICS LVM_GETINSERTMARK LVM_GETINSERTMARKCOLOR LVM_GETINSERTMARKRECT LVM_GETOUTLINECOLOR LVM_GETSELECTEDCOLUMN LVM_GETTILEINFO LVM_GETTILEVIEWINFO LVM_GETUNICODEFORMAT LVM_GETVIEW LVM_HASGROUP LVM_INSERTGROUP LVM_INSERTGROUPSORTED LVM_INSERTMARKHITTEST LVM_ISGROUPVIEWENABLED LVM_MAPIDTOINDEX LVM_MAPINDEXTOID LVM_MOVEGROUP LVM_MOVEITEMTOGROUP LVM_REMOVEALLGROUPS LVM_REMOVEGROUP LVM_SETBKIMAGE LVM_SETGROUPINFO LVM_SETGROUPMETRICS LVM_SETINFOTIP LVM_SETINSERTMARK LVM_SETINSERTMARKCOLOR LVM_SETOUTLINECOLOR LVM_SETSELECTEDCOLUMN LVM_SETTILEINFO LVM_SETTILEVIEWINFO LVM_SETTILEWIDTH LVM_SETTOOLTIPS LVM_SETUNICODEFORMAT LVM_SETVIEW LVM_SETWORKAREAS LVM_SORTGROUPS LVM_SORTITEMSEX -- Dimi.

1 0

Msg help
by Dimitrie O. Paun Oct. 20, 2002

Oct. 20, 2002

Anyone know the message name for the codes: 2486 2793 The progress bar receives them, and complains because it doesn't know what they are... -- Dimi.

1 0

RtlAllocateHeap crash
by Dimitrie O. Paun Oct. 20, 2002

Oct. 20, 2002

While I was playing around with Newsbin Pro, I stumbled over this: Unhandled exception: page fault on write access to 0x0000000c in 32-bit code (0x400c2eaa). In 32-bit mode. 0x400c2eaa (RtlAllocateHeap+0xda [heap.c:1039] in libntdll.dll.so): movl %eax,0xc(%edx) 1040 pArena->prev->next = pArena->next; Is it worth investigating? -- Dimi.

1 0

Re: WebBrowser
by Dimitrie O. Paun Oct. 20, 2002

Oct. 20, 2002

On October 19, 2002 03:27 am, Malte Starostik wrote: > ChangeLog > * Fixed IWebBrowser::Navigate() parameter type > * Added IWebBrowserApp and related typedefs BTW, any taker on implementing IWebBrowser & Co. in terms of the Gecko API, so we can get Mozilla working through Wine? -- Dimi.

3 4

Re: glibc 2.3 pthread patch
by Alexandre Julliard Oct. 20, 2002

Oct. 20, 2002

Ove Kaaven <ovehk(a)ping.uio.no> writes: > Now that glibc 2.3 is available for those bleeding-edge folks on debian > unstable, it seems that Wines compiled on 2.2 won't run on 2.3 systems. > The glibc folks did warn us about this a while ago though, so here's a > patch that implements their suggestion for making wine compatible with as > many glibc versions as possible. Perhaps Alexandre wants to use his > WINE_GET_SONAME stuff to detect the name of the libc at configure time > instead of at runtime like I do, though. Funny, I was working on the same thing right now... But do you really need to fetch the real libc handle? It seems to me RTLD_NEXT should be enough. Here's how I did it: Index: scheduler/pthread.c =================================================================== RCS file: /opt/cvs-commit/wine/scheduler/pthread.c,v retrieving revision 1.26 diff -u -r1.26 pthread.c --- scheduler/pthread.c 12 Sep 2002 22:07:06 -0000 1.26 +++ scheduler/pthread.c 20 Oct 2002 21:45:13 -0000 @@ -38,13 +38,6 @@ #include "thread.h" #include "winternl.h" -static int init_done; - -void PTHREAD_init_done(void) -{ - init_done = 1; -} - /* Currently this probably works only for glibc2, * which checks for the presence of double-underscore-prepended * pthread primitives, and use them if available. @@ -62,28 +55,18 @@ asm(".globl " PSTR(alias) "\n" \ "\t.set " PSTR(alias) "," PSTR(orig)) -/* strong_alias does not work on external symbols (.o format limitation?), - * so for those, we need to use the pogo stick */ -#if defined(__i386__) && !defined(__PIC__) -/* FIXME: PIC */ -#define jump_alias(orig, alias) __ASM_GLOBAL_FUNC( alias, "jmp " PSTR(orig)) -#endif +static int init_done; -/* get necessary libc symbols */ -#if (__GLIBC__ == 2) && (__GLIBC_MINOR__ >= 1) && defined(HAVE___LIBC_FORK) -#define LIBC_FORK __libc_fork -#define PTHREAD_FORK __fork -#define ALIAS_FORK -#else -#define LIBC_FORK __fork -#define PTHREAD_FORK fork -#endif -extern pid_t LIBC_FORK(void); +static pid_t (*libc_fork)(void); +static int (*libc_sigaction)(int signum, const struct sigaction *act, struct sigaction *oldact); + +void PTHREAD_init_done(void) +{ + init_done = 1; + if (!libc_fork) libc_fork = dlsym( RTLD_NEXT, "fork" ); + if (!libc_sigaction) libc_sigaction = dlsym( RTLD_NEXT, "sigaction" ); +} -#define LIBC_SIGACTION __sigaction -extern int LIBC_SIGACTION(int signum, - const struct sigaction *act, - struct sigaction *oldact); /* NOTE: This is a truly extremely incredibly ugly hack! * But it does seem to work... */ @@ -255,15 +238,20 @@ } strong_alias(__pthread_atfork, pthread_atfork); -pid_t PTHREAD_FORK(void) +pid_t __fork(void) { pid_t pid; int i; + if (!libc_fork) + { + libc_fork = dlsym( RTLD_NEXT, "fork" ); + assert( libc_fork ); + } EnterCriticalSection( &atfork_section ); /* prepare handlers are called in reverse insertion order */ for (i = atfork_count - 1; i >= 0; i--) if (atfork_prepare[i]) atfork_prepare[i](); - if (!(pid = LIBC_FORK())) + if (!(pid = libc_fork())) { InitializeCriticalSection( &atfork_section ); for (i = 0; i < atfork_count; i++) if (atfork_child[i]) atfork_child[i](); @@ -275,9 +263,7 @@ } return pid; } -#ifdef ALIAS_FORK -strong_alias(PTHREAD_FORK, fork); -#endif +strong_alias(__fork, fork); /***** MUTEXES *****/ @@ -658,13 +644,20 @@ /***** ANTI-OVERRIDES *****/ /* pthreads tries to override these, point them back to libc */ -#ifdef jump_alias -jump_alias(LIBC_SIGACTION, sigaction); -#else int sigaction(int signum, const struct sigaction *act, struct sigaction *oldact) { - return LIBC_SIGACTION(signum, act, oldact); + if (!libc_sigaction) + { + libc_sigaction = dlsym( RTLD_NEXT, "sigaction" ); + assert( libc_sigaction ); + } + return libc_sigaction(signum, act, oldact); +} + +#else /* __GLIBC__ */ + +void PTHREAD_init_done(void) +{ } -#endif #endif /* __GLIBC__ */ -- Alexandre Julliard julliard(a)winehq.com

2 2

Known listview problems
by Dimitrie O. Paun Oct. 20, 2002

Oct. 20, 2002

Folks, Things look pretty good on the listview front. In fact, I am kindda happy the way things are, as the S-series patches were rather tricky at points... Currently, I know of the following problems: -- Newsbin dies swiftly (reported my Rein Klazes) -- FTPCommander seems to experience some listview-related faults (reported by Carlos Lozano) -- Xnews has some minor focus rectangle problems (discovered by myself :)) -- IDAdemo _may_ have some minor drawing glitches (ditto) At this point, I would like to ask everybody that is aware of a listview problem to report it. I might not be able to put as much time, and effort into listview, as I did lately, so the more of a complete picture I have now, the better the chances that I'm gonna fix it. Next on my TODO: -- Fix Newsbin's crash (BTW, Rein I need a new trace with the latest CVS, if possible) -- Store per-column info into the listview (see the second status update for details) If you feel that there is a particular are of the listview that needs attention, let me know so I can work it into my TODO, time permitting. -- Dimi.

3 6

configuration
by Eric Pouech Oct. 20, 2002

Oct. 20, 2002

I dunno if Alexandre will like this patch, but - when compiling with --disable-win16 turned on - the glue generation is based on the C_SRCS16 files as well as the C_SRCS - this, of course, brings in lots of unneed imports (triggered by the non needed glue code) the attached patch, when wine compilation is configured with --disable-win16, removes all 16 bit .c files from the inspection for glue generation this might not be sufficient: - the .glue.c file is generated in any case - however, it should compile as an empty compilation unit A+ Index: configure.ac =================================================================== RCS file: /home/cvs/cvsroot/wine/wine/configure.ac,v retrieving revision 1.84 diff -u -r1.84 configure.ac --- configure.ac 10 Oct 2002 17:54:28 -0000 1.84 +++ configure.ac 20 Oct 2002 17:43:53 -0000 @@ -25,10 +25,12 @@ AC_SUBST(OPTIONS) AC_SUBST(WIN16_FILES,"\$(WIN16_FILES)") AC_SUBST(WIN16_INSTALL,"\$(WIN16_INSTALL)") +AC_SUBST(C_SRCS16,"\$(C_SRCS16)") if test "x$enable_win16" = "xno" then WIN16_FILES="" WIN16_INSTALL="" + C_SRCS16="" fi if test "x$enable_debug" = "xno" then Index: dlls/Makedll.rules.in =================================================================== RCS file: /home/cvs/cvsroot/wine/wine/dlls/Makedll.rules.in,v retrieving revision 1.36 diff -u -r1.36 Makedll.rules.in --- dlls/Makedll.rules.in 2 Oct 2002 02:34:10 -0000 1.36 +++ dlls/Makedll.rules.in 20 Oct 2002 17:42:51 -0000 @@ -37,7 +37,7 @@ # Rule for 16-bit glue $(MODULE).glue.c: $(C_SRCS) $(C_SRCS16) $(WINEBUILD) - $(LDPATH) $(WINEBUILD) $(DEFS) -o $@ -glue -C$(SRCDIR) $(C_SRCS) $(C_SRCS16) + $(LDPATH) $(WINEBUILD) $(DEFS) -o $@ -glue -C$(SRCDIR) $(C_SRCS) @C_SRCS16@ # Rules for .dll files

2 1

Listview regression
by David D. Hagood Oct. 20, 2002

Oct. 20, 2002

I've just applied/built/installed/tested the U series patches, and I have 2 regressions in the AAA MapNGo program: First, the first column of the directions window (which shows the type of road) is missing. Second, when I do the "Along the way", the program cores out. The log, and a screen cap showing (or rather, NOT showing) the missing field in the directions window, are in a tarball: http://sktc.net/~wowbagger/listview3.tar.gz

2 10

Fw: Wine problem
by Leonardo Bocalon Oct. 20, 2002

Oct. 20, 2002

Hello it´s me again , i tried the procedure that you told me to do, I downloaded the last version of the wine, i installed it and still not working the keyboard display. Trying to explain to you more specifically the problem I´m going to send you the log which was the problem!!(I compacted whit .lha). I also want to know how to register (use) the files virtual drivers ( VXD)! Thanks again for you atention. Leonardo Bocalon Sysmo Informática / Brasil ----- Original Message ----- From: "Andreas Mohr" <andi(a)rhlx01.fht-esslingen.de> To: "Leonardo Bocalon" <leonardo(a)sysmo.com.br> Cc: <wine-devel(a)winehq.com> Sent: Wednesday, October 16, 2002 4:16 AM Subject: Re: Wine problem > On Tue, Oct 15, 2002 at 01:50:19PM -0300, Leonardo Bocalon wrote: > > Ac wine programers !! > > I am from a software company and we have a problem, our software communicate with a keyboard which has a display this software is make using Delphi (by borland) and for cominicate with the keyboard we use assembler code , wile we run the software in windows it work perfectly but when we try to use it in Linux using the wine it dosen´t work (the keyboard display). > > > > Please a need a sulution for it !!!! > Please retry with latest Wine CVS if you haven't yet, then add a BugZilla report > describing the problem if it still persists. > Attach screenshots if useful and maybe already add some log file like: > wine --debugmsg +relay,+event,+key,+keyboard MyPrg.exe &>/tmp/winekey.log > > -- > Andreas Mohr Stauferstr. 6, D-71272 Renningen, Germany > Tel. +49 7159 800604 http://mohr.de.tt

5 4

Re: DPA_InsertPtr (A1)
by Dimitrie O. Paun Oct. 20, 2002

Oct. 20, 2002

On October 20, 2002 12:33 pm, Paul Rupe wrote: > I was just looking at this same problem. Should we not also remove the > line hdpa->nItemCount++? DPA_SetPtr already does this (which is why we > needed the -1 you added), so I think we end up incrementing it twice. You are correct. I've actually sent the patch in about 5minutes ago ... :) > For comparison, here's the patch I was about to send (relative to the first > comctl32undoc.c patch): Yup, that's what we need to end up with. -- Dimi.

1 0