Re: windows visual view ?
On December 16, 2002 01:08 pm, Francois Gouget wrote: > Would this be more or less what you are looking for? That's OK, but I was looking for some of them architecture diagrams that explain the interaction between processes and wineserver. Something to shed some light on the more obscure corners of wine. This ones are OK, but we kinda already know that yeah, Wine is build on top on Linux... :) -- Dimi.
Re: PATCH: implemented Fiber APIs
On December 16, 2002 02:37 pm, Alberto Massari wrote:
> +void WINAPI SwitchToFiber(LPVOID lpFiber)
> +{
> +#ifdef HAVE_UCONTEXT_H
> +PHFIBER pFiber,pCurrentFiber;
> +pFiber=(PHFIBER)lpFiber;
> +pCurrentFiber=(PHFIBER)TlsGetValue(fiber_tls_index);
> +TRACE("Switching from %p to %p\n",pFiber,pCurrentFiber);
> +TlsSetValue(fiber_tls_index,pFiber);
> +swapcontext(&pCurrentFiber->context,&pFiber->context);
> +#else
> +FIXME("stub\n");
> +#endif
> +}
Can you (pretty) please do without the ifdefs? They can probably
be located in only one spot, if you need to test for HAVE_UCONTEXT_H.
--
Dimi.
Re: stub handling
On December 16, 2002 08:42 pm, Alexandre Julliard wrote:
> We need to implement them, importing stubs no longer works (but of
> course you can do a stub implementation instead ;-)
Of, of, of, of, o... :)
ChangeLog
Add stubs for {Start,End}{Doc,Page}Printer.
Index: dlls/winspool/info.c
===
RCS file: /var/cvs/wine/dlls/winspool/info.c,v
retrieving revision 1.61
diff -u -r1.61 info.c
--- dlls/winspool/info.c12 Nov 2002 02:22:24 - 1.61
+++ dlls/winspool/info.c17 Dec 2002 06:43:09 -
@@ -1274,6 +1274,51 @@
}
/*
+ * EndDocPrinter [WINSPOOL.@]
+ */
+BOOL WINAPI EndDocPrinter(HANDLE hPrinter)
+{
+FIXME("(hPrinter=%p): stub\n", hPrinter);
+return FALSE;
+}
+
+/*
+ * EndPagePrinter [WINSPOOL.@]
+ */
+BOOL WINAPI EndPagePrinter(HANDLE hPrinter)
+{
+FIXME("(hPrinter=%p): stub\n", hPrinter);
+return FALSE;
+}
+
+/*
+ * StartDocPrinterA [WINSPOOL.@]
+ */
+DWORD WINAPI StartDocPrinterA(HANDLE hPrinter, DWORD Level, LPBYTE pDocInfo)
+{
+FIXME("(hPrinter=%p, Level=0x%lx, pDocInfo=%p): stub\n", hPrinter, Level,
+pDocInfo);
+return FALSE;
+}
+
+/*
+ * StartDocPrinterW [WINSPOOL.@]
+ */
+DWORD WINAPI StartDocPrinterW(HANDLE hPrinter, DWORD Level, LPBYTE pDocInfo)
+{
+FIXME("(hPrinter=%p, Level=0x%lx, pDocInfo=%p): stub\n", hPrinter, Level,
+pDocInfo);
+return FALSE;
+}
+
+/*
+ * StartPagePrinter [WINSPOOL.@]
+ */
+BOOL WINAPI StartPagePrinter(HANDLE hPrinter)
+{
+FIXME("(hPrinter=%p): stub\n", hPrinter);
+return FALSE;
+}
+
+/*
* GetFormA [WINSPOOL.@]
*/
BOOL WINAPI GetFormA(HANDLE hPrinter, LPSTR pFormName, DWORD Level,
Index: dlls/winspool/winspool.drv.spec
===
RCS file: /var/cvs/wine/dlls/winspool/winspool.drv.spec,v
retrieving revision 1.16
diff -u -r1.16 winspool.drv.spec
--- dlls/winspool/winspool.drv.spec 4 Nov 2002 23:53:42 - 1.16
+++ dlls/winspool/winspool.drv.spec 17 Dec 2002 06:01:17 -
@@ -56,8 +56,8 @@
@ stdcall DocumentPropertiesA(long long ptr ptr ptr long) DocumentPropertiesA
@ stdcall DocumentPropertiesW(long long ptr ptr ptr long) DocumentPropertiesW
@ stub EXTDEVICEMODE
-@ stub EndDocPrinter
-@ stub EndPagePrinter
+@ stdcall EndDocPrinter(long) EndDocPrinter
+@ stdcall EndPagePrinter(long) EndPagePrinter
@ stub EnumFormsA
@ stub EnumFormsW
@ stdcall EnumJobsA(long long long long ptr long ptr ptr) EnumJobsA
@@ -127,8 +127,8 @@
@ stub SpoolerInit
@ stub StartDocDlgA
@ stub StartDocDlgW
-@ stub StartDocPrinterA
-@ stub StartDocPrinterW
-@ stub StartPagePrinter
+@ stdcall StartDocPrinterA(long long ptr) StartDocPrinterA
+@ stdcall StartDocPrinterW(long long ptr) StartDocPrinterW
+@ stdcall StartPagePrinter(long) StartPagePrinter
@ stub WaitForPrinterChange
@ stdcall WritePrinter(long ptr long ptr) WritePrinter
--
Dimi.
Re: wrc options
On December 9, 2002 04:57 pm, Francois Gouget wrote:
> I agree with that one. As it is wrc just convert the filenames to
> lowercase before trying to include them which is just plain wrong. In
> particular it causes compilation errors if the file really has a
> mized-case which is not that uncommon.
ChangeLog
Remove the -L wrc option, as no automatic fudging of filenames
can ever be correct.
Index: tools/wrc/parser.y
===
RCS file: /var/cvs/wine/tools/wrc/parser.y,v
retrieving revision 1.33
diff -u -r1.33 parser.y
--- tools/wrc/parser.y 15 Aug 2002 21:57:36 - 1.33
+++ tools/wrc/parser.y 17 Dec 2002 05:33:45 -
@@ -2694,11 +2694,6 @@
*cptr = '/';
}
- /* Convert to lower case. Seems to be reasonable to do */
- for(cptr = str->str.cstr; !leave_case && *cptr; cptr++)
- {
- *cptr = tolower(*cptr);
- }
return str;
}
Index: tools/wrc/wrc.c
===
RCS file: /var/cvs/wine/tools/wrc/wrc.c,v
retrieving revision 1.21
diff -u -r1.21 wrc.c
--- tools/wrc/wrc.c 12 Dec 2002 22:23:55 - 1.21
+++ tools/wrc/wrc.c 17 Dec 2002 05:32:44 -
@@ -104,7 +104,6 @@
" -I path Set include search dir to path (multiple -I allowed)\n"
" -J Do not search the standard include path\n"
" -l lan Set default language to lan (default is neutral {0, 0})\n"
- " -L Leave case of embedded filenames as is\n"
" -m Do not remap numerical resource IDs\n"
" -n Do not generate .s file\n"
" -N Do not preprocess input\n"
@@ -252,12 +251,6 @@
int auto_register = 0;
/*
- * Set when the case of embedded filenames should not be converted
- * to lower case (-L option)
- */
-int leave_case = 0;
-
-/*
* The output byte-order of resources (set with -B)
*/
int byteorder = WRC_BO_NATIVE;
@@ -433,9 +426,6 @@
error("Language %04x is not supported",lan);
currentlanguage = new_language(PRIMARYLANGID(lan),
SUBLANGID(lan));
}
- break;
- case 'L':
- leave_case = 1;
break;
case 'm':
remap = 0;
Index: tools/wrc/wrc.h
===
RCS file: /var/cvs/wine/tools/wrc/wrc.h,v
retrieving revision 1.24
diff -u -r1.24 wrc.h
--- tools/wrc/wrc.h 31 May 2002 23:41:01 - 1.24
+++ tools/wrc/wrc.h 17 Dec 2002 05:32:50 -
@@ -64,7 +64,6 @@
extern DWORD codepage;
extern int pedantic;
extern int auto_register;
-extern int leave_case;
extern int byteorder;
extern int preprocess_only;
extern int no_preprocess;
--
Dimi.
Re: Serious problem with / and mount points
--- Tony Lambregts <[EMAIL PROTECTED]> a écrit : > Jeff Smith wrote: > > > What I have so far: > > in files/dos_fs.c:1343, drive is receiving the value -1. > > at line 1358, this is blindly added to 'A'. As you may > > know, in ASCII, 'A' - 1 = '@'. The problem is that we don't check the status of DRIVE_FindDriveRoot, as you spotted it. The sources clearly states that -1 is error. > > > > The next step is to determine why DRIVE_FindDriveRoot is > > returning -1. > > Well the reason DRIVE_FindDriveRoot is returning -1 is that > DOSDrives[drive].ino does not equal st.st_ino anymore. Line 440 of > drive.c is as follows > >if ((DOSDrives[drive].dev == st.st_dev) && >(DOSDrives[drive].ino == st.st_ino)) > > > > > > > > > > BTW, the same occurred in the reverse situation: > >(konsole) mount /mnt/cdrom > >(konsole) wcmd > >(wcmd) D: > >(wcmd) dir > >(konsole) umount /mnt/cdrom > >(wcmd) dir > > Now I am at @> > > If I remount and dir again, I am back to D> > > > Same problem with floppys too. > > -- > > Tony Lambregts > > > = Sylvain Petreolle [EMAIL PROTECTED] Fight against Spam ! http://www.euro.cauce.org/en/index.html ICQ #170597259 "Don't think you are. Know you are." Morpheus in Matrix, chapter 15. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com
Re: linking between wine books
Bill Medland wrote: What to people think? Should we be linking between different books in the wine-doc set? I had put a link from wine-user to a section in wine-devel. The pdf was happy enough with it but the sgm-to-html choked on it. Easiest solution is to drop the link. Otherwise can someone point me at an example where it works? Linking between books at a local level only works if you are building as a set. If you are building just one book then AFAIK there is no way to tell it about an id in another book. (One way around it is to point to the tag point in the winehq documentation). If anyone knows different please speak up. -- Tony Lambregts
Re: include msvcrt/ fix (take 2)
"Dimitrie O. Paun" <[EMAIL PROTECTED]> writes:
> What about this one? Currently {ming,wine}wrap are badly broken.
>
> I hope this one is rather uncontroversial, but it does depend on
> that bit from crtdll, and I think that is the problem.
I'd prefer that we simply get rid of these includes. This avoids
having to do hacks in crtdll that will need to be removed later
anyway.
--
Alexandre Julliard
[EMAIL PROTECTED]
Re: include msvcrt/ fix (take 2)
On December 14, 2002 02:10 pm, Dimitrie O. Paun wrote:
> ChangeLog
> Do not explicitly include msvcrt files from msvcrt headers.
> Include windef.h instead of winnt.h to get all appropriate types.
Alexandre,
What about this one? Currently {ming,wine}wrap are badly broken.
I hope this one is rather uncontroversial, but it does depend on
that bit from crtdll, and I think that is the problem.
--
Dimi.
Re: How long does it take *you* to build wine?
Bill Medland wrote: On December 15, 2002 10:51 pm, Dan Kegel wrote: snip So how long does it take *you* to do a clean build (just the 'make' part after 'make depend', say) of Wine-20021125? - Dan (I was wondering if mine was the slowest!) real - 47m0.357 user-43m1.631 sys-2m51.668 Celeron 450 with 128MB running RedHat 7.3 snip No yours is fast compared to my 300mhz pentium with 96mb of ram (around 3 hours) -- Tony Lambregts
Re: stub handling
"Dimitrie O. Paun" <[EMAIL PROTECTED]> writes: > printing.o: In function `printer_start_job': > printing.o(.text+0x1fa): undefined reference to `EndDocPrinter' > printing.o(.text+0x20a): undefined reference to `EndPagePrinter' > printing.o(.text+0x237): undefined reference to `StartDocPrinterA' > printing.o(.text+0x24a): undefined reference to `StartPagePrinter' > printing.o: In function `printer_finish_job': > printing.o(.text+0x2b6): undefined reference to `EndPagePrinter' > printing.o(.text+0x2bd): undefined reference to `EndDocPrinter' > collect2: ld returned 1 exit status > > How shall we proceed? We need to implement them, importing stubs no longer works (but of course you can do a stub implementation instead ;-) -- Alexandre Julliard [EMAIL PROTECTED]
Re: How long does it take *you* to build wine?
On December 15, 2002 10:51 pm, Dan Kegel wrote: snip > So how long does it take *you* to do a clean build > (just the 'make' part after 'make depend', say) of > Wine-20021125? > - Dan (I was wondering if mine was the slowest!) real - 47m0.357 user-43m1.631 sys-2m51.668 Celeron 450 with 128MB running RedHat 7.3 snip -- Bill Medland
Re: regression - 'wine notepad.exe' now has fixme
Uwe Bonnes <[EMAIL PROTECTED]> writes: > > "Dan" == Dan Kegel <[EMAIL PROTECTED]> writes: > > Dan> With today's cvs, I see fixme:win:SetWindowTextA cannot set text > Dan> "Notepad - (untitled)" of other process window (nil) when I run > Dan> notepad.exe. This also happens with wine20021125, but not with > Dan> wine20020904. > > SetWindowText can not yet set the text across process boundary. But more > functions do the right thing and use SetWindowText and not a workaround or > don't use it at all. So this is not a regression. Actually the window is (nil) so this is not an inter-process issue at all; the error message is a bit misleading in that case. And if it didn't happen before then it's clearly a regression. -- Alexandre Julliard [EMAIL PROTECTED]
Re: [RESENT (20021013)]RelayExclude/Include
Uwe Bonnes <[EMAIL PROTECTED]> writes: > Changelog: > relay32/relay386.c: check_relay_include > Treat whole dlls as advertised IMO the current behavior is correct. If the documentation says otherwise it needs to be fixed. -- Alexandre Julliard [EMAIL PROTECTED]
Re: winmm test extension
On Mon, 16 Dec 2002, Paul Millar wrote: [...] > In the future, there might be many interactive tests: or at least, ones > that require someone to sit in front of the computer. These could all be > turned off by default, but could be turned on through an option to > runtest, or via an environment variable (WINETEST_INTERACTIVE). > > Does this sound reasonable? Sounds reasonable to me. -- Francois Gouget [EMAIL PROTECTED]http://fgouget.free.fr/ I haven't lost my mind, it's backed up on tape around here somewhere...
Re: How long does it take *you* to build wine?
On Sun, Dec 15, 2002 at 10:51:59PM -0800, Dan Kegel wrote: > So how long does it take *you* to do a clean build > (just the 'make' part after 'make depend', say) of > Wine-20021125? current cvs on an athlon 550 with 192 M Ram and gcc 3.2: real47m40.902s user40m25.200s sys 3m1.410s Ciao Jörg -- Joerg Mayer <[EMAIL PROTECTED]> I found out that "pro" means "instead of" (as in proconsul). Now I know what proactive means.
linking between wine books
What to people think? Should we be linking between different books in the wine-doc set? I had put a link from wine-user to a section in wine-devel. The pdf was happy enough with it but the sgm-to-html choked on it. Easiest solution is to drop the link. Otherwise can someone point me at an example where it works? -- Bill Medland
Re: fdopen behaviour correction
> "Bill" == Bill Currie <[EMAIL PROTECTED]> writes: Bill> fdopen is not supposed to rewind the file descriptor being given a Bill> FILE *. Proof of this comes from quakeforge working in windows Bill> natively but not in wine without the attatched patch. Bill> ChangeLog * dlls/msvcrt/file.c don't rewind the file after Bill> creating the FILE* handle It is good to have a test case for such cases in wine/dlls/msvcrt/tests. Bye -- Uwe Bonnes[EMAIL PROTECTED] Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt - Tel. 06151 162516 Fax. 06151 164321 --
Re: winmm test extension
On Mon, 16 Dec 2002, Francois Gouget wrote: > How to make it play a sound? > Simple. Just pass the 'play' parameter to the test. Playing a sound is a nice idea, but perhaps there's a more general way of choosing to hear it. In the future, there might be many interactive tests: or at least, ones that require someone to sit in front of the computer. These could all be turned off by default, but could be turned on through an option to runtest, or via an environment variable (WINETEST_INTERACTIVE). Does this sound reasonable? (no pun intended ;) Paul Millar
stub handling
Alexandre, In the past functions that were declared stubs in the .spec file got a dummy definition, so programs using them could link. This doesn't seem to be the case anymore: mingwrap -mwindows -mno-cygwin -s -o putty.exe be_all.o cmdline.o ldisc.o \ ldiscucs.o logging.o misc.o noise.o pageantc.o portfwd.o \ printing.o proxy.o raw.o rlogin.o settings.o sizetip.o ssh.o \ sshaes.o sshblowf.o sshbn.o sshcrc.o sshcrcda.o sshdes.o \ sshdh.o sshdss.o sshmd5.o sshpubk.o sshrand.o sshrsa.o \ sshsh512.o sshsha.o sshzlib.o telnet.o terminal.o tree234.o \ unicode.o version.o wcwidth.o wildcard.o win_res.res.o \ winctrls.o windlg.o window.o winnet.o winstore.o winutils.o \ x11fwd.o -ladvapi32 -lcomctl32 -lcomdlg32 -lgdi32 -limm32 \ -lshell32 -luser32 -lwinmm -lwinspool -lwsock32 printing.o: In function `printer_start_job': printing.o(.text+0x1fa): undefined reference to `EndDocPrinter' printing.o(.text+0x20a): undefined reference to `EndPagePrinter' printing.o(.text+0x237): undefined reference to `StartDocPrinterA' printing.o(.text+0x24a): undefined reference to `StartPagePrinter' printing.o: In function `printer_finish_job': printing.o(.text+0x2b6): undefined reference to `EndPagePrinter' printing.o(.text+0x2bd): undefined reference to `EndDocPrinter' collect2: ld returned 1 exit status How shall we proceed? -- Dimi.
Re: resend: ConvertThreadToFiber stub
On Mon, 16 Dec 2002 13:47:10 -0600, you wrote: > On Monday 16 December 2002 01:08 pm, Rein Klazes wrote: > > Use this one, the previous patch wasn't clean > > > > > > Changelog: > > > > dlls/kernel : kernel32.spec > > scheduler : thread.c > > > > Add a stub for ConvertThreadToFiber > > > > Rein. > > funny, I recently implemented same but never submitted. using HTML Help > by any chance? Yes, so I can use the the MSDN library help files. Rein. -- Rein Klazes [EMAIL PROTECTED]
Re: winmm test extension
Hi Dan, Dan Kegel wrote: [...] I may have a counterexample. The patch I submitted recently was to fix a hang in playing sound. So if you add a timeout to your test, it would at least verify that sounds played in the proper amount of time, and didn't hang wine. Yes, there are a number of failure modes that could be tested automatically. Wine could crash for instance... What I could do is modify the test to play silence by default, and only play an actual tone if the user asks for it. I should also document the goals of playing a tone a bit more in the test itself. Maybe printing a trace telling the user how to play a tone could be nice (otherwise it's hard for someone running the test to know that the test has options). The message could look something like: winmm.c: Playing silence, use 'tone' option for an actual tone The idea of timing the sound and verifying that it does not play in less time than it's supposed to could work too. I'm a bit worried by that but I guess all that can happen is that we will overestimate the time it took to play the tone, not underestimate it. Well, if nobody objects I can add all that to the test once it's committed. -- Francois Gouget [EMAIL PROTECTED]
Re: resend: ConvertThreadToFiber stub
On Monday 16 December 2002 01:08 pm, Rein Klazes wrote: > Use this one, the previous patch wasn't clean > > > Changelog: > > dlls/kernel : kernel32.spec > scheduler : thread.c > > Add a stub for ConvertThreadToFiber > > Rein. funny, I recently implemented same but never submitted. using HTML Help by any chance? -- gmt
Re: winmm test extension
Francois Gouget wrote: This test extends the winmm test and makes it possible to play actual sound. Of course the test will not do that unless you specifically ask for it: * I doubt anyone would be happy to do 'make test' in Wine's main directory and to get annoying long beeps comming out of their computer * tests are usually run unattended and there is not much point playing the sounds if no-one listens to them to make sure they got played correctly. So what does it do? For each supported format it plays the international 'LA', i.e. a 440Hz sinusoid. It is expected that on some sound cards and for some formats the signal will have to be converted from 8 to 16 bit or from mono to stereo, or even resampled. Normally this should not affect the sound being played so you should keep hearing the same tone. If that's not the case then something went wrong during th I may have a counterexample. The patch I submitted recently was to fix a hang in playing sound. So if you add a timeout to your test, it would at least verify that sounds played in the proper amount of time, and didn't hang wine. - Dan
Re: DrawCaptionTempW patch
On Mon, 16 Dec 2002 19:45:51 +0100, you wrote: > > "Rein" == Rein Klazes <[EMAIL PROTECTED]> writes: > > Rein> Hi, This bug caused uninitialised height and width data to be fed > Rein> to the font rasterizer, allocating more then 1.2 GB of VM. > > What action or program triggered that bug? TheBat! email program. It has stopped working for some time, perhaps two months. Rein. -- Rein Klazes [EMAIL PROTECTED]
Re: DrawCaptionTempW patch
> "Rein" == Rein Klazes <[EMAIL PROTECTED]> writes: Rein> Hi, This bug caused uninitialised height and width data to be fed Rein> to the font rasterizer, allocating more then 1.2 GB of VM. What action or program triggered that bug? -- Uwe Bonnes[EMAIL PROTECTED] Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt - Tel. 06151 162516 Fax. 06151 164321 --
Re: windows visual view ?
On Mon, 16 Dec 2002, Tom Wickline wrote: > Is was just wondering if there was any type of Windows visaul view ? > view as in how evreything is layed out ? and how wine compares to Win32 ? > Say if you had a visual view of how win32 is layed out and just how > close wine comes to this . Would this be more or less what you are looking for? * Windows Architecture http://www.lugod.org/presentations/crossover/Diapositive5.html * Wine Architecture http://www.lugod.org/presentations/crossover/Diapositive6.html * API stats (the 'diff' part) http://fgouget.free.fr/wine/winapi_stats-en.shtml -- Francois Gouget [EMAIL PROTECTED]http://fgouget.free.fr/ Cahn's Axiom: When all else fails, read the instructions.
Workaround for Wine to produce dead-keys with Brazilian ABNT-2 keyboards
I spent a lot of time looking for a solution for using my Brasilian ABNT-2
keyboard. I've already sent a new patch to wine-patches in order to correct
keytables and vkey tables. But it remained a small problem: It was NOT generating
the correct dead-key codes. So, I developed a workaround. This patch
has to be applyied to dlls/x11drv/keyboard.c (It was tested with CVS 20021125).
I advice that it wil work only for ABNT-2 keyboards.
--- keyboard.c 2002-11-28 22:33:07.0 -0200
+++ keyboard.c 2002-12-16 15:10:07.0 -0200
@@ -1759,6 +1794,10 @@
else TRACE("Found keycode %d (0x%2X)\n",e.keycode,e.keycode);
ret = XLookupString(&e, (LPVOID)lpChar, 2, &keysym, NULL);
+
+/* Gambiarra para funcionarem os acentos em teclado ABNT-2 */
+if ((e.keycode==48)||(e.keycode==34)||((e.keycode==15)&&(e.state
& ShiftMask))) ret=0;
+
wine_tsx11_unlock();
if (ret == 0)
It now hangs when running an install shield
Wine, You gues have already given me loada help, but straight onto my next stage, and I get (my old friend) the Wine Console error traceback. Running an Install Shield instalation, starts OK, gets past the initial 'Install Shield will now guide you through your instalation'. Then blanks the screen for about ten minutes, returning with the console showing: Unhandles exception: page fault on write access to E:\_ITSTMP.DIR\_INS55765._WP): movl %eax,0x28(%edx,%ecx,4) Looked in /tmp/winetmp-root/_ITSTMP.DIR/* & can't see anything wrong. 'df' shows a few dozen gig free. And clues? I hope at the end of this, I can return the great feedback by answering some questions my self :) Ben
Re: How long does it take *you* to build wine?
On Sun, 15 Dec 2002, Dan Kegel wrote: > I got tired of waiting for wine to build, and started looking > for ways to speed it up. > > Wine-20021125? > - Dan Well, with a dual 1800+ MP system ( cvs from 20021216 ) takes real6m49.808s user12m0.330s done with make -j2 > > p.s. > No fair using ccache! That's the next tool I want to investigate; > it's probably the biggest win of all, at least for regression > testing. > >
Re: Wine, Windows.Forms on Linux, GC and segfaults.
On Wed, Dec 11, 2002 at 01:05:32PM -0500, Miguel de Icaza wrote: > Hello, > > > > The problem is that by the time that Wine has been initialized, > > > using setjmp/longjmp will always lead to a crash. The code > > > in pthreads > > > that performs the longjmp will first try to invoke the pthread cleanup > > > routines, and then invoke longjmp. This never happens. > > > > You can't and needn't link with -lpthread. Wine has its own > > pthread implmentation. > > > > I tried your included code and it works just fine unless you > > link with -lpthread, then it crashes in the same way as in your > > attached picture. But then you should never link Wine with > > -lpthread so that is not really a bug. > > > > Of course Wine pthread implementation it not in any way complete > > so some function might be missing and some might only be only > > partially implemented and of course some might be incorrectly > > implemented. > > > > So please try again without linking with -lpthread. > > The problem is that this is very hard for us to do as two of the > underlying libraries we are using (libmono and libgc) both link against > pthread to achieve their threading capabilities. > > It might be better to investigate what are Wine's requirements for > having its own pthread implementation, and have those changed pushed to > the maintainers of pthreads. > > In particular, we are interested in using WineLib, maybe it would be > possible to have WineLib use the standard pthread implementation instead > of rolling its own? This would be very difficult. WINE implements the Win32 threading/process and synchronisation model. It needs to use the OS native threads so it gets the stack maintenance and the correct handling of the %fs. So WINE cannot change to pthreads without major hacks to the pthread library I guess. Ciao, Marcus
windows visual view ?
Is was just wondering if there was any type of Windows visaul view ? view as in how evreything is layed out ? and how wine compares to Win32 ? Say if you had a visual view of how win32 is layed out and just how close wine comes to this . If there was a negitive of Win-ME for example and one of wine how would wine look if it was layed on top of the Win-ME negitive ? I am trying to view Win32 in my head and wine and how they ( scope ) up. could this cause a brain hemorrhage ?? . :) comments , flames , views, insight most welcome Tom
Quick silly question, registering DLL's
Forgive any stupidness implied, I am UNIX not MS! How does one check whether a DLL is registered, and how does one actually register a DLL in wine? Thanks,
Re: VarOr / oleaut32
> It's a VB6 application of our own description. I have not had a chance > to hack the source code to see where it comes into play, but probably > something to do with a third party DLL we use for remote database > connectivity. There's a significant number of Var* routines not implemented. I implemented the missing ones for a VB6 app I had written, but each app executes different ones, so there's still quite a few missing. If you try the patch already submitted, see if you get further and potentially hit other missing functions. Most of the missing fn's are not too bad to implement, so keep reporting them Jason
Re: How long does it take *you* to build wine?
On Sun, Dec 15, 2002 at 10:51:59PM -0800, Dan Kegel wrote: > It does look like distributing the load helps a tiny > bit, but it's not clear it's a big win over just building > on the dual 650MHz machine. > The fastest build I've seen so far is 19 minutes. I'm pretty > sure I'm not doing things optimally yet. > > So how long does it take *you* to do a clean build > (just the 'make' part after 'make depend', say) of > Wine-20021125? On a athlon 900 with 512 MB RAM it takes about 30 minutes for a full build, but only for the first run. On the second run it takes about 3-4 minutes :). Ok, i'm using ccache and it realy paid out during my -DSTRICT work. The cache hit rate had varied between 30% and 80%, depending what i've done. You get 30% - 40% hitrate for normal updates from cvs without doing a "make clean". > No fair using ccache! That's the next tool I want to investigate; > it's probably the biggest win of all, at least for regression > testing. I've thought about buying a new computer before i started to use ccache. With ccache i'll wait for the Hammer. bye michael -- Michael Stefaniuc Tel.: +49-711-96437-199 System Administration Fax.: +49-711-96437-111 Red Hat GmbHEmail: [EMAIL PROTECTED] Hauptstaetterstr. 58http://www.redhat.de/ D-70178 Stuttgart msg15497/pgp0.pgp Description: PGP signature
Re: VarOr / oleaut32
On Mon, Dec 16, 2002 at 09:20:18AM +, Ben Clewett wrote: > [EMAIL PROTECTED] wrote: > >Hiya, > > > >What app shows this? Can I get hold of it - I should be able to knock > >something up for it as a test regardless, but it might not be til mid next > >week. > > > >Jason > > > > It's a VB6 application of our own description. I have not had a chance > to hack the source code to see where it comes into play, but probably > something to do with a third party DLL we use for remote database > connectivity. > > Thanks for the offer of a fix. > > I was hoping there was a generic fault, and somebody might have a > generic fix. Since it seems to be more indepth, I will try any get an > exact location of location of failure first. Even though I now implemented VarOr, I guess you might want to be careful about that and use a native oleaut32 instead, since oleaut32 contains tons of variant calculation functions (mul, add, and, xor, various conversions, div, round), so if there's *any* bug in there affecting some function your program uses, boom goes your data eventually. That's why I'd use native oleaut32 for production use. OTOH I'm still waiting for some builtin oleaut32 test reports ;) -- Andreas MohrStauferstr. 6, D-71272 Renningen, Germany Tel. +49 7159 800604http://mohr.de.tt
Re: VarOr / oleaut32
[EMAIL PROTECTED] wrote: Hiya, What app shows this? Can I get hold of it - I should be able to knock something up for it as a test regardless, but it might not be til mid next week. Jason It's a VB6 application of our own description. I have not had a chance to hack the source code to see where it comes into play, but probably something to do with a third party DLL we use for remote database connectivity. Thanks for the offer of a fix. I was hoping there was a generic fault, and somebody might have a generic fix. Since it seems to be more indepth, I will try any get an exact location of location of failure first. Ben
Re: How long does it take *you* to build wine?
Here are my timings for a 1.8 GHz P4: time ./configure real0m55.767s user0m15.590s sys 0m14.530s time make clean real0m9.166s user0m2.980s sys 0m2.270s time make real16m55.724s user11m47.180s sys 1m2.460s I wonder, if it's really necessary to change configure scripts and do a "configure/make clean all" that often as it's done right now. The good thing about it is, that it shows how well Linux' mutlitasking does work compared to Windows. You can just let it build it in the background and do other things meanwhile. ;-) -- Martin Fuchs [EMAIL PROTECTED]
Re: How long does it take *you* to build wine?
Dan Kegel wrote: So how long does it take *you* to do a clean build (just the 'make' part after 'make depend', say) of Wine-20021125? OK, got some numbers building 20021125 on my home machines: 30min : 750MHz PIII laptop 17min 21sec: dual 650MHz PIII, 7200RPM disks, make -j2 16min 1sec: distcc on the dual 650MHz as localhost, 750MHz laptop, and a 1.1GHz celeron, make -j4 And on another machine not at home: 8 min : dual 1.2GHz P4 Xeon, make -j2 So the fastest build I've seen so far (not including configure or make depend) is about 8 minutes. distcc seems to help a tiny bit. it's kind of looking like I should get a faster motherboard if I want sub-ten-minute builds. Next step should be to measure benefit of using ccache when searching through cvs for the patch that broke something, but I'll probably just go out and get a faster motherboard. I want one with gigabit ethernet onboard; pricewatch says $320 for an Asus A7V8X with "2400" Athlon XP, $353 for an Asus P4PE with 2.4GHz P4. Hmm. I bet either one of those would get me Wine builds below ten minutes... Does anyone have a good recipe for building an inexpensive machine that can build Wine quickly? - Dan
