- Wine-Devel - list.winehq.org

Re: 100% CPU usage in Wine-0.9
by Mark Knecht Dec. 14, 2005

Dec. 14, 2005

On 11/15/05, Mark Knecht <markknecht(a)gmail.com> wrote: > On 11/10/05, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > <SNIP> > > > > > LATE BREAKING NEWS: It appears that when I switch to the Alsa > > > driver the CPU usage does down to normal. Were you using Alsa or OSS? > > > I tried both alsa and oss(with oss emulation in alsa), both are working fine. > > I don't use any hw acceleration. So is this program useable for you now with > > alsa? > > > > Stefan, > After having a few people test this it seems that it is not simply > enough to use the Alsa driver. A couple of people have tried that and > continue to see 100% CPU usage, so apparently there is something more > involved. > > In the bug report I have attached some debug info I was asked to > get. Hopefully that will help a developer in some way. > > I am here to test other ideas or do more debug work. I think this > problem is real. > > Cheers, > Mark > Hi, I haven't seen any responses on this in the last month. Did I possibly miss something? Thanks, Mark

1 0

josystick driver layer and support of multiple devices of one kind
by Christoph Frick Dec. 14, 2005

Dec. 14, 2005

hiho i did recently some patches to mostly make joystick_linuxdinput.c handle things more like joystick_linux.c. the main thing, that i dont like on this work, is the fact, that i am just copy-and-paste code from one end to another. i assume someone started back then the joystick_linuxdinput part and if there where changes at either one of them they where not done (or not all the time) also in the other file. also if i have understood this correctly there is no support for multiple joysticks of one kind at least in the joystick_linuxdinput part (i can not confirm this, because i had for now a js-joystick and a event-wheel - but a event-joystick is on the way). for many game-types it is much likely, that the users want to use more than one input-device (e.g. use the wheel and the pedals from two different manufactures, use the pedals from your wheel with your flight stick, use two gamepads, ...) and its most likely that such a device is nowadays a usb device poping up in /dev/input. if the current implementation works fine with more than one js/event-driven joystick - so let me know; my windows/dinput knowledge ist zero and i have not the hardware to test it yet. so my point: i want to make the joystick_... interface to be more abstract. so to bring the code, that is common for the js and event into one common part and then utilize implementations of a interface for each js and event based joysticks. with this step it would make sense to also allow the "driver" (i call it that way now) to report the joysticks that are available and so have support for multiple joysticks. a split like this would also allow an easier integration of e.g. joystick support for OSX or other opteration systems that handle joysticks in another way then unix/linux does. so my questions: - i this an approach, that would be acceptablefor wine? so if i start on something like this, is there a chance it would be submitted to the tree - or is there a problem in terms of "wine does not work like this and we would ignore such an approach" - is the sound-system of wine with its lots of implementations a place to get ideas for such a change; or is this totally off, because the user only chooses one driver to use (instead of: only build the driver if appropriate and then let each driver search for hardware) -- cu

3 3

Problems generating c++ project since wine 0.9.3
by Sergio Dec. 14, 2005

Dec. 14, 2005

Hi all, I've been having problems trying to compile one c++ project since latest wine version (0.9.3). This project compiled and worked fine with previous versions. The error messages look like: /usr/include/wine/windows/unknwn.h:26: error: expected unqualified-id before ‘{’ token ... /usr/include/wine/windows/objidl.h:66: error: expected unqualified-id before ‘:’ token ... /usr/include/wine/windows/oaidl.h:553: error: expected unqualified-id before ‘:’ token ... /usr/include/wine/windows/propidl.h:287: error: expected unqualified-id before ‘:’ token ... /usr/include/wine/windows/oleidl.h:46: error: expected unqualified-id before ‘:’ token ... Is there something that I am misisng when generating this project? Or is this a known issue? Thanks. Sergio.

2 2

Re: PATCH: handle NULL XImage in GetBitmapBits
by Alexandre Julliard Dec. 13, 2005

Dec. 13, 2005

Marcus Meissner <marcus(a)jet.franken.de> writes: > Changelog: > In GetBitmapBits() handle the correct bitmap information > when a DIB is used. I think for a DIB we want to copy the bits directly, like we already do in SetBitmapBits. Does this work for you? diff --git a/dlls/gdi/bitmap.c b/dlls/gdi/bitmap.c index a0c6b25..5c2b6c3 100644 --- a/dlls/gdi/bitmap.c +++ b/dlls/gdi/bitmap.c @@ -307,6 +307,30 @@ LONG WINAPI GetBitmapBits( if (!bmp) return 0; + if (bmp->dib) /* simply copy the bits from the DIB */ + { + DIBSECTION *dib = bmp->dib; + const char *src = dib->dsBm.bmBits; + DWORD max = dib->dsBm.bmWidthBytes * dib->dsBm.bmHeight; + if (count > max) count = max; + ret = count; + if (!bits) goto done; + + if (bmp->dib->dsBmih.biHeight >= 0) /* not top-down, need to flip contents vertically */ + { + src += dib->dsBm.bmWidthBytes * dib->dsBm.bmHeight; + while (count > 0) + { + src -= dib->dsBm.bmWidthBytes; + memcpy( bits, src, min( count, dib->dsBm.bmWidthBytes ) ); + bits = (char *)bits + dib->dsBm.bmWidthBytes; + count -= dib->dsBm.bmWidthBytes; + } + } + else memcpy( bits, src, count ); + goto done; + } + /* If the bits vector is null, the function should return the read size */ if(bits == NULL) { -- Alexandre Julliard julliard(a)winehq.org

3 2

Can Wine run "sv.net classic"? Not at the moment :-(
by Dan Kegel Dec. 13, 2005

Dec. 13, 2005

http://www.heise.de/newsticker/meldung/67265 describes what looks like a nice opportunity for wine to prove itself. German employers are required to send insurance information online. There are two allowed methods: either a web page (which apparantly sucks), or a windows app called sv.net classic, available as a free download at http://www.itsg.de/svnet/ I tried installing it, but I didn't get past square one because the mdac installer is more busted than it used to be, at least on my machine. - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

3 4

Re: gethostbyname(my_name) and IL2-Sturmovik
by Juan Lang Dec. 13, 2005

Dec. 13, 2005

Hi Pavel, > And on the other side, try to query a DNS server for "localhost". > Every DNS server should immediately return 127.0.0.1, it's RFC > compliance requirement. Just to be clear, we're not talking about gethostbyname("localhost"). We're talking about gethostbyname(the_name_of_my_machine). In Windows, this returns your external address. See the following sample code which demonstrates this: http://tangentsoft.net/wskfaq/examples/ipaddr.html On Linux, at least for me, I get only 127.0.0.1 for my hostname (juandell, when I'm running on this laptop.) If I were to have this name registered with DNS, and if I were to ask a DNS server for the address of it, I wouldn't get 127.0.0.1, of course. But we're not talking about DNS anyway. We're talking about winsock's gethostbyname function. Wine's implementation isn't correct, since it doesn't match Windows. I still think David's patch could use some work, but let's not argue about whether its behaviour is correct: matching Windows is what we're obliged to do. --Juan __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

A DirectX ToDo
by Tom Wickline Dec. 13, 2005

Dec. 13, 2005

Hello Everyone, I've just posted a first draft of a DirectX ToDo on the Wiki : http://wiki.winehq.org/DirectX-ToDo?action=show Feedback, Comments, Suggestions are welcome. Cheers, Tom

1 0

Re: WINE-NB.BLOCK
by Rüdiger Koerschulte Dec. 13, 2005

Dec. 13, 2005

> Hello, > > I`m sorry, but my English is not so good, but I`m trying to get an answer > about: > > I have installed SUSE 10.0, then WINE and then WINETOOLS. and the IE6. Now, > when I press the key comma, on the numeric-block, I get the remove > function. When I press die remove function an the cursor-block, I get a > comma. > > The function is always the same. It is the same when I put the > numeric-block on or off. > > How I can get a comma on the right key? > > Thank you for your understanding for my schoolenglish! > > Rüdiger

1 0

cross-compile error
by Steven Edwards Dec. 13, 2005

Dec. 13, 2005

Hi, When doing Cross-Compiles I am seeing this error ../../include/wine/exception.h:142: error: parse error before "sigjmp_buf" ../../include/wine/exception.h:142: warning: no semicolon at end of struct or union ../../include/wine/exception.h:146: error: parse error before '}' token ../../include/wine/exception.h:146: warning: type defaults to `int' in declaration of `__WINE_FRAME' ../../include/wine/exception.h:146: warning: data definition has no type or storage class It seems the detection of sigjmp longjmp is busted for cross-compiling. I am using this patch locally but don't know how to properly fix it. Index: include/wine/exception.h =================================================================== RCS file: /home/wine/wine/include/wine/exception.h,v retrieving revision 1.26 diff -u -r1.26 exception.h --- include/wine/exception.h 1 Sep 2004 17:36:04 -0000 1.26 +++ include/wine/exception.h 13 Dec 2005 18:03:58 -0000 @@ -139,7 +139,7 @@ /* finally data */ __WINE_FINALLY finally_func; } u; - sigjmp_buf jmp; + jmp_buf jmp; /* hack to make GetExceptionCode() work in handler */ DWORD ExceptionCode; const struct __tagWINE_FRAME *ExceptionRecord; -- Steven Edwards - ReactOS and Wine developer "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

1 0

Re: gethostbyname(my_name) and IL2-Sturmovik
by David Nolden Dec. 13, 2005

Dec. 13, 2005

@Pavel Troller I've checkt the /etc/hosts now. You were right, it says 127.0.0.1 localhost.localdomain localhost Zwirch (With Zwirch the Name of this Computer) But I can't just statically rewrite it, because I use DHCP. Also this is a fresh Ubuntu-Install, so I'm surely not the only Wine-User with such an /etc/hosts.. @Juan You're right, I've checked the patch and it indeed was ugly! But when I sent it I hadn't done the check under Windows, so I didn't know if the patch was generally useful, so I mainly ment it as a piece for discussion.(Also I didn't notice how ugly it really was ;) I've cleaned it up a bit, closed the leak, and did a few other small tweaks so now the patch itself might be acceptable. greetings, David Am Dienstag, 13. Dezember 2005 07:04 schrieb Pavel Troller: > > Ok, I've just checked doing the same thing under windows. > > > > The test(ripdaveno is the name of the Computer I tested on): > > > > { > > ... > > hostent* h = gethostbyname("ripdaveno"); > > char* str = inet_ntoa(*((in_addr*)h->h_addr_list[0])); > > ... > > } > > > > The string in str was "192.168.2.75" which is my Network IP-Adress. With > > wine(under Linux), as far as I've tested it, exactly the same code always > > generates "127.0.0.1". > > > > Since Wine tries to simulate Windows as exactly as possible, this is > > wrong. So, in my opinion this Patch is necessary. :-) > > Hi! > It's strange but there is no patch in this mail I can look at... > However, even more strange is that wine returns localhost address to > you... Why ? I just tested Your testing code (just slightly modified to get > it to compile) on both plain linux as well as in wine and in both cases it > returns my real public IP. Isn't your Linux networking setup (/etc/hosts > table) a bit wrong ? My one contains (arcus is my machine name) > > 127.0.0.1 localhost > 195.39.17.7 arcus.sinus.cz arcus > > I have already seen broken setups containing things like > > 127.0.0.1 localhost arcus > > which is of course wrong. > > Regards, Pavel Troller

4 6