May 2004 - Wine-Devel - list.winehq.org

A little question o suggestion...
by Urbez Santana Roma May 17, 2004

May 17, 2004

El Lunes, 8 de Marzo de 2004 23:20, wine-devel-request(a)winehq.org escribió: The file wine/server/file.c contains in line 151 to 156 (function create_file) the next lines: ----------------------------------------------------------------------------------------------------------------- mode = (attrs & FILE_ATTRIBUTE_READONLY) ? 0444 : 0666; if (len >= 4 && (!strcasecmp( name + len - 4, ".exe" ) || !strcasecmp( name + len - 4, ".com" ))) mode |= 0111; ------------------------------------------------------------------------------------------------------------------ That is a problem for a instal.lation programs or another programs that writen a Driver *.drv or a *.dll file, or another unknown extension. When created the file, the permisions not be 0555 or 0777, and the program instal.led not works, wine cannot execute them. My opinion, is that the files in a Windows Filesystem are virtually executable (the flag not exist, and all is an executable if one loads to exec) I cray that we must written the mode|=0111, whithout query the extension '.exe' '.com' '.drv' '.dll', etc... (note that the code only wants .exe and .com) I have write these modification and solutions the problem. ---------------------------------------------------------------------------------------------- mode = (attrs & FILE_ATTRIBUTE_READONLY) ? 0444 : 0666; mode |= 0111; ---------------------------------------------------------------------------------------------- -- Another problem is in file wine/dlls/kernel/ne_module.c : If the extension is not dll, exe, or com, the dll with another name cannot be executed. for example extensions with .drv .ldd .dat that are really DLL's cannot be executed, i have write a patch to solve it, but only i have added the extensions (that is not the correct solution). If u will, i can write a patch that detect if it is a NE module really without test the extension, (testing NE signature anz MZ, of the file...) ################################################ #- Urbez Santana i Roma - #- Email: urbez(a)linuxupc.upc.es #- Private Web: http://linuxupc.upc.es/~urbez/ ################################################

1 0

stdole2.tlb and stdole32.tlb
by Jonathan Wilson May 16, 2004

May 16, 2004

Is anyone working on getting sutable-to-use-in-wine versions of these? Would having these benifit OLE apps?

2 1

Re: VirtualProtect bug?
by Dan Kegel May 16, 2004

May 16, 2004

> I'm trying to write a test that checks for data overruns > using VirtualAlloc and VirtualProtect and I think > I may have found a bug. Here is the problem: > > DWORD dwPageSize; > BYTE * twoPages, temp; > DWORD flOldProtect; > twoPages = VirtualAlloc(NULL, 2 * dwPageSize, MEM_RESERVE | MEM_COMMIT, PAGE_READWRITE); > res = VirtualProtect(twoPages + dwPageSize, dwPageSize, PAGE_NOACCESS, &flOldProtect); > temp = twoPages[dwPageSize + 1]; // should cause exception but doesn't in wine > twoPages[dwPageSize + 1] = 0; // does cause exception Say, that code fragment wasn't complete, was it? Looks like dwPageSize was uninitialized... > I would like to write a test to check the exception handling > but I don't know how to do a __try and __except in C. OK, I don't know much about it either. A little googling shows: http://www.kerneltraffic.org/wine/wn20030516_170.html#8 makes it sound like __try is hard to do with gcc. You might need to punt and, when compiling with gcc, do nothing unless building in winelib, where you'd set up a handler for SIGSEGV to catch the bad access. http://www.codeproject.com/cpp/transactions.asp?df=100&forumid=11253&exp=0&… has some examples of how __try is used in visual C. (Note that you can download a copy of visual C's commandline compiler free from Microsoft.) What confuses me is, doesn't MinGW support __try? Hrm. - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

1 0

Patch to reduce warnings from comm.c
by Scott W Gifford May 15, 2004

May 15, 2004

I have two programs I run under Wine that make use of the serial ports. I get this error and fixme when I run then: err:comm:GetCommState tcgetattr or ioctl error 'Invalid argument' fixme:comm:SetupComm insize 1050 outsize 1050 unimplemented stub I fixed these two in my copy of Wine with the attached patch. First, the GetCommState errors were because I was using a pseudo-tty as the serial port, so setserial didn't work on it. I modified it to ignore errors from setserial that set errno==EINVAL. SetupComm is a function that sets the communications buffer sizes. Wine doesn't have its own communication buffers, so I filled out this function to set the kernel's xmit buffer to the requested amount if possible, and ignore the receive buffer. With these changes, my programs work properly (they worked properly before), and I don't get these complaints from Wine. -----ScottG.

1 0

Re: symbol hiding
by Dan Timis May 15, 2004

May 15, 2004

> From: Mike Hearn <mh(a)codeweavers.com> > Subject: Re: symbol hiding > Date: Sat, 15 May 2004 17:07:15 +0100 > Organization: CodeWeavers, Inc > > I think Pauls point was that if what you are doing is so secret and so > proprietary that you're afraid symbol names will leak trade secrets > you're > probably better off with some full blown code obfuscater/encryption > system. Actually, we do have an encryption system, and it is the name of the functions used internally by the encryption library that we want to hide. It is not for hiding trade secrets, but to prevent dishonest customers from accessing features without paying for them. This has been already used successfully on Mac and Windows where it is very easy to hide those names. Doing this will not make our copy protection full proof, but will slow down the inevitable hack. Hopefully, by the time someone will crack our copy protection we will have a new stronger version. Dan

2 1

Re: combobox: combobox popups should be always-on-top
by Dimitrie O. Paun May 15, 2004

May 15, 2004

On Sat, May 15, 2004 at 11:34:08AM -0700, Steven Edwards wrote: > Changelog: Thomas Weidenmueller <w3seek(a)reactos.com> > combobox popups should be always-on-top Formatting seems to be a bit off. Shouldn't the content of the else be indented 4 spaces? -- Dimi.

1 0

Re: symbol hiding
by Dan Timis May 15, 2004

May 15, 2004

Thank you Paul and Brian, I thought it was probably OK to ask these kind of questions. But, it's nice to know for sure. Dan

1 0

Re: symbol hiding
by Dan Timis May 15, 2004

May 15, 2004

On Friday, May 14, 2004, at 05:34 AM, wine-devel-request(a)winehq.org wrote: > Message: 5 > Date: Thu, 13 May 2004 22:26:34 -0400 > From: Paul Davis <paul(a)linuxaudiosystems.com> > To: wine-devel(a)winehq.org > Subject: symbol hiding > > for this reason, i personally wouldn't feel comfortable asking in a > public forum how to go about doing this. your mileage may and probably > does vary, of course. My original question generated this new thread, which in turn became very technical. But, I think that Paul's original posting was more of a philosophical or maybe even moral nature. I'm not sure if there is a FAQ about the etiquette on this group, the kind of questions that can be asked, and the kind that cannot be asked. I have nothing to hide. Our company sells a piece a hardware that has an x86 mother board inside and runs Linux and wine. It is a consumer appliance similar to TiVo in the sense that it does not look like a computer and the customer does not have access to the Linux shell, nor can he or she run other applications. On top Linux we have a proprietary application that runs under wine (wine -- proprietary-application.exe.so). Our application is not open source. If questions relating to a commercial non open source product that uses wine heavily are not welcome on this list, I apologize and I will never ask these kinds of questions again. Dan Timis Muse Research, Inc.

3 2

Building WINE with Intel ICC for Linux
by Steven Edwards May 15, 2004

May 15, 2004

Hello, I was able to build WINE with the Intel Compiler for Linux. Intel allows free usage for non-comercial works so I figured I would check and see if changing compilers would yield better performance. According to Intels site it is supposed to give a 15 to 30% speed boost in applications. I was able to compile WINE with only about a 10 line patch. Somehow some of of the inline asm is already getting defined by icc. I am having some problems running graphical applications and maybe it is related to the code I had to disable. If anyone else wants to test I figure this might be a good tool to have if you are porting a Winelib application and want a large performance speed up. I was able to get all of WINE to build but was not able to run any graphical applications due to the crash show in the attached crash.log I was however able to run quite a few console applications including regresion tests which showed failures quite a few failures. If anyone is interested in building WINE with the Intel Compiler I will be happy to test a bit but I dont have a lot of time to put in to it. Attached is my hack.diff and the log from trying to run regedit. 1. Download ICC and run icsvar.sh 2. apply the following patch 3. Configure wine with ./configure CC=icc 4. recompile WINE. Thanks Steven __________________________________ Do you Yahoo!? Win a $20,000 Career Makeover at Yahoo! HotJobs http://hotjobs.sweepstakes.yahoo.com/careermakeover

4 6

Broken link in WWN archive
by Stefan Leichter May 15, 2004

May 15, 2004

Hello, to reproduce the problem do this sequence starting from winehq's main page: 1. click on More Issues... <http://www.winehq.org/?issue=back> in Weekly Newsletters 2. scroll to the bottom of the page and click Next 12 Issues >> <http://www.winehq.org/?issue=back;pos=12> 3. scroll to the bottom of the page and click << Prev 12 Issues <http://www.winehq.org/?wwn=back;pos=0> you should now be back on the page of step 2 but you are back on the main page. The problem can be solved be replaceing the 'wwn' of the link against 'issue'. I did not find the template. Therefore I could not send a path :-( Bye Stefan

2 1