May 2007 - Wine-devel - list.winehq.org

Fedora Packages
by Andreas Bierfert 30 May '07

30 May '07

Hi, here is some news/status update on the fedora packages: As people have been complaining about the link on the winehq webpage I have taken the time to make a summary page[1] on the fedora wiki which is easier for me to maintain and keep up-to-date and is probably more helpful. I would request that the link in the download section is changed to that page. I have good news for users of RHEL/Centos: After collaborating with some other maintainers from the fedora project I finally got all dependencies for wine into the EPEL[2] repository (which is something like Fedora Extras but for RHEL/Centos). This means that there now is a wine version for users of RHEL/Centos 5 (it may take a couple of hours or a day till it is published to the repository and all mirrors have synced it). It will probably take a couple of days but I finished testing these changes for RHEL/Centos 4 as well so there should also be a version out soon. Comments and suggestions are (like always) very much welcome. Regards, Andreas [1] - http://fedoraproject.org/wiki/AndreasBierfert/Wine [2] - http://fedoraproject.org/wiki/EPEL -- Andreas Bierfert | http://awbsworld.de | GPG: C58CF1CB andreas.bierfert(a)lowlatency.de | http://lowlatency.de | signed/encrypted phone: +49 2402 102373 | cell: +49 173 5803043 | mail preferred

1 1

c++ code
by Evan Stade 30 May '07

30 May '07

Hi I want to write conformance tests for gdiplus and that requires the use of c++. Even if I were to focus on the gdi+ flat api, the windows SDK headers for that have c++ code in them, also eventually when the gdi+ c++ wrappers are written on top of the flat api then we will need to be able to compile c++ in the wine environment. What do I have to do to be able to write these tests? -Evan Stade

4 8

Re: Pretty Build System (Take 3)
by Kirill K. Smirnov 29 May '07

29 May '07

You probably forget to run 'autoconf'. I can see nice messages like: $ autoconf ... $ ./configure ... $ make depend ... $ make ... CC extrachunk.o CC factory.o CC getframe.o CC icmstream.o CC regsvr.o CC tmpfile.o >I get the following (on Slackware 11) while running the make depend >portion of ./configure && make depend && make && sudo make install: >make[1]: Entering directory `/home/speeddy/wine-git/tools' >../tools/makedep -C. -S.. -T.. -I/usr/include/freetype2 bin2res.c >fnt2bdf.c fnt2fon.c make_ctests.c makedep.c relpath.c sfnt2fnt.c >make[1]: Leaving directory `/home/speeddy/wine-git/tools' >make[1]: Entering directory `/home/speeddy/wine-git/tools' >make[1]: `makedep' is up to date. >make[1]: Leaving directory `/home/speeddy/wine-git/tools' >make[1]: Entering directory `/home/speeddy/wine-git/dlls' >make[2]: Entering directory `/home/speeddy/wine-git/dlls/acledit' >../../tools/makedep -C. -S../.. -T../.. main.c >make[2]: Leaving directory `/home/speeddy/wine-git/dlls/acledit' >make[2]: Entering directory `/home/speeddy/wine-git/dlls/acledit' >make[2]: MAKERULE_MAKEDEP@: Command not found >make[2]: *** [depend] Error 127 >make[2]: Leaving directory `/home/speeddy/wine-git/dlls/acledit' >make[1]: *** [acledit/__depend__] Error 2 >make[1]: Leaving directory `/home/speeddy/wine-git/dlls' >make: *** [dlls/__depend__] Error 2

1 0

Re: wininet: Add a new NETCON_query_data_available function.
by Stefan Leichter 29 May '07

29 May '07

Am Monday 28 May 2007 12:16 schrieb Robert Shearman: > Use it to implement the behaviour where InternetReadFileExA does a > synchronous request if the data is available and asynchronous otherwise. > --- > configure.ac | 1 + > dlls/wininet/internet.c | 49 > +++++++++++++++++++++++++++-------------- > dlls/wininet/internet.h | 1 + > dlls/wininet/netconnection.c | 35 ++++++++++++++++++++++++++++++ > 4 files changed, 69 insertions(+), 17 deletions(-) Hello, your patch broke the compilation for me. Bye Stefan make[2]: Entering directory `/usr/src/wine/wine-build/dlls/wininet' gcc -c -I../../../wine-git/dlls/wininet -I. -I../../../wine-git/include -I../../include -D__WINESRC__ -D_WINX32_ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O2 -o netconnection.o ../../../wine-git/dlls/wininet/netconnection.c ../../../wine-git/dlls/wininet/netconnection.c: In function ‘NETCON_query_data_available’: ../../../wine-git/dlls/wininet/netconnection.c:585: error: ‘WININET_NETCONNECTION’ has no member named ‘peek_msg’ ../../../wine-git/dlls/wininet/netconnection.c:586: error: ‘WININET_NETCONNECTION’ has no member named ‘peek_len’ make[2]: *** [netconnection.o] Fehler 1 make[2]: Leaving directory `/usr/src/wine/wine-build/dlls/wininet'

1 0

Happy SOC'ing all.
by Maarten Lankhorst 29 May '07

29 May '07

For all other students joining SummerOfCode, I think it's a good idea if we have a public place where we can see each other's changes. I set up my own fork, sound.git at http://repo.or.cz/w/wine.git - if others want to do the same I encourage them to do so, it's easy when you figure out how to git push. Since soc now officially started (according to date, in holland anyway), I just want to say: - Happy coding - Good luck all Cheers, Maarten

4 4

Crosscompile (mingw) failure with msxml3 tests
by Paul Vriens 29 May '07

29 May '07

Hi, I just wanted to report the current failures when crosscompiling the msxml3 tests with mingw (and thus no new winetest executable): xmldoc.cross.o:xmldoc.c:(.text+0x1a9): undefined reference to `_IID_IXMLDocument' xmldoc.cross.o:xmldoc.c:(.text+0x1c0): undefined reference to `_CLSID_XMLDocument' xmldoc.cross.o:xmldoc.c:(.text+0x149d): undefined reference to `_IID_IXMLDocument' xmldoc.cross.o:xmldoc.c:(.text+0x14b4): undefined reference to `_CLSID_XMLDocument' xmlelem.cross.o:xmlelem.c:(.text+0x22): undefined reference to `_IID_IXMLDocument' xmlelem.cross.o:xmlelem.c:(.text+0x39): undefined reference to `_CLSID_XMLDocument' xmlelem.cross.o:xmlelem.c:(.text+0x1425): undefined reference to `_IID_IXMLDocument' xmlelem.cross.o:xmlelem.c:(.text+0x143c): undefined reference to `_CLSID_XMLDocument' xmlelem.cross.o:xmlelem.c:(.text+0x1c46): undefined reference to `_IID_IXMLElement' xmlelem.cross.o:xmlelem.c:(.text+0x1f0f): undefined reference to `_IID_IXMLElement' collect2: ld returned 1 exit status make: *** [msxml3_crosstest.exe] Error 1 These started to happen after the addition of some patches made by James Hawkins (Add tests for [IXMLElement|IXMLDocument]) on May 26th. Cheers, Paul.

1 0

uninstaller: Shorten variable name length.. Whats wrong?
by Tom Spear 29 May '07

29 May '07

Just curious to know if there was something wrong with this patch, or if maybe it wasnt committed due to being a cosmetic change? If it's the latter, I'll just resend the other 2 parts of the patch (2and 3 of 3) against the code the way it is instead of changing the lines.. On 5/23/07, Tom Spear <speeddymon(a)gmail.com> wrote: > This patch shortens the length of some lines in > FetchUninstallInformation, as suggested by Peter Beutner > > -- > Thanks > > Tom > > -- Thanks Tom

1 0

[wiki] Child pages mess up the logo-location
by Paul Vriens 29 May '07

29 May '07

Hi, Every now and then I check for the 'Recent Changes' on the wiki. The following page caught my attention: http://wiki.winehq.org/MaartenLankhorst/SOC2007 This page is actually named "MaartenLankhorst/SOC2007" and is a child of the 'MaartenLankhorst' page. This page messes up the logo in the top-left corner. When looking at the properties of the logo it shows 'http://wiki.winehq.org/MaartenLankhorst/logo.png' instead of the correct 'http://wiki.winehq.org/logo.png'. The same logic is true for http://wiki.winehq.org/MacOSX/FAQs (and probably others). Just thought I mention it. Cheers, Paul.

1 0

Start of an opengl-based gdi driver
by Stefan Dösinger 29 May '07

29 May '07

Hi, One of our remaining fundamental d3d problems is the slowness of IDirectDrawSurface7::GetDC(and d3d9). This call allows performing gdi operations on surfaces. Currently this is handled through downloading the opengl surface, performing the gdi operation, and uploading it again. This is slow, especially on render targets. On textures this is rarely used, and wined3d can better cache the texture. Most applications use it to render text text on the back buffer. Applications I have are the d3d7 sdk demos, Empire Earth, Prince of Persia 3D, Age of Empires 2. Sometimes it is also used to load textures with a BitBlt. I have discussed two ways of solving this issue with alexandre. One was to tell the X server to perform the calls in opengl(EXT_texture_from_pixmap looked promissing), the other one was writing an opengl gdi driver in wined3d. The X server approach failed because there is just no API to get that handled, and EXT_texture_from_pixmap explicitly doesn't do that job, and even if there is a hacky way driver support will be bad. So I spent the last days to check how an implementation in wined3d would work - and here it is. So far this driver handles only ExtTextOutW, and I only successfully used it with the directx sdk samples. ExtTextOut uses a Helper device context(normal gdi) to get the glyphs with GetGlyphOutline, loads it into an opengl texture(GL_ALPHA) and records 256 display lists to draw textured quads. So far no unicode support, no custom pens, but custom fonts work. I have 2 screenshots of a dx7 sdk demo with classic getdc and the opengl driver(radeon M9, 1.6 ghz pentium M, 512 mb ram). There seems to be some parameter missing, the text is slightly displaced. Tex.gif(attached) is the created font texture(warning, white text on transparent background). http://stud4.tuwien.ac.at/~e0526822/gl.png http://stud4.tuwien.ac.at/~e0526822/gdi.png One problem of this approach is that the driver has to handle everything. However, I keep a DIB section of the size of the surface, and if I encounter some unsupported call I can print a FIXME, download from gl and call the dib section to do the job. After ReleaseDC the surface code has to upload it again. Its the slow old behavior, so we don't gain anything for unsupported calls, but its no loss either. So far most forward functions are missing, so the DIB driver refuses to initialize itself - Edit CreateDC in dlls/wined3d/gdi.c to activate it. ExtTextOut has an example how the forwarding looks like. A big problem I still have are operations that involve 2 device contexts. Prince of Persia 3D(good test case for custom fonts) calls StretchBlit from/to my gl dc. So in X11drv one of the physdevs is a WineD3DSurfaceImpl * instead of a WINEX11DRV_PHSYDEV * - not good. So far I haven't found a way to let x11drv access my surface data, or enforce that my function is called, fetch the hdc from the x11 phsydev and call my fallback dc, or request the device's content from x11drv and handle the blit myself.

2 3

Start of an opengl-based gdi driver
by Stefan Dösinger 28 May '07

28 May '07

Hi, One of our remaining fundamental d3d problems is the slowness of IDirectDrawSurface7::GetDC(and d3d9). This call allows performing gdi operations on surfaces. Currently this is handled through downloading the opengl surface, performing the gdi operation, and uploading it again. This is slow, especially on render targets. On textures this is rarely used, and wined3d can better cache the texture. Most applications use it to render text text on the back buffer. Applications I have are the d3d7 sdk demos, Empire Earth, Prince of Persia 3D, Age of Empires 2. Sometimes it is also used to load textures with a BitBlt. I have discussed two ways of solving this issue with alexandre. One was to tell the X server to perform the calls in opengl(EXT_texture_from_pixmap looked promissing), the other one was writing an opengl gdi driver in wined3d. The X server approach failed because there is just no API to get that handled, and EXT_texture_from_pixmap explicitly doesn't do that job, and even if there is a hacky way driver support will be bad. So I spent the last days to check how an implementation in wined3d would work - and here it is. So far this driver handles only ExtTextOutW, and I only successfully used it with the directx sdk samples. ExtTextOut uses a Helper device context(normal gdi) to get the glyphs with GetGlyphOutline, loads it into an opengl texture(GL_ALPHA) and records 256 display lists to draw textured quads. So far no unicode support, no custom pens, but custom fonts work. Attached are 2 screenshots of a dx7 sdk demo with classic getdc and the opengl driver(radeon M9, 1.6 ghz pentium M, 512 mb ram). There seems to be some parameter missing, the text is slightly displaced. Tex.gif is the created font texture(warning, white text on transparent background) One problem of this approach is that the driver has to handle everything. However, I keep a DIB section of the size of the surface, and if I encounter some unsupported call I can print a FIXME, download from gl and call the dib section to do the job. After ReleaseDC the surface code has to upload it again. Its the slow old behavior, so we don't gain anything for unsupported calls, but its no loss either. So far most forward functions are missing, so the DIB driver refuses to initialize itself - Edit CreateDC in dlls/wined3d/gdi.c to activate it. ExtTextOut has an example how the forwarding looks like. A big problem I still have are operations that involve 2 device contexts. Prince of Persia 3D(good test case for custom fonts) calls StretchBlit from/to my gl dc. So in X11drv one of the physdevs is a WineD3DSurfaceImpl * instead of a WINEX11DRV_PHSYDEV * - not good. So far I haven't found a way to let x11drv access my surface data, or enforce that my function is called, fetch the hdc from the x11 phsydev and call my fallback dc, or request the device's content from x11drv and handle the blit myself.

1 0