August 2010 - Wine-Devel - list.winehq.org

Wine Gecko 1.1.0 RC1

by Luca Bennati

Applied the patch to master, recompiled ok. I put the rc1 cab in /usr/share/wine/gecko/, as always. In the creation of a new prefix, it gave me an err which wasn't there before: err:mshtml:init_xpcom AutoRegister(NULL) failed: 80070057 but 'wine iexplore' works better than before (sites that crashed wine before don't crash it anymore). I investigated it a bit recreating a new prefix with 'WINEDEBUG=+mshtml winecfg' and here's the log: wine: created the configuration directory '/home/luca/.wine' trace:mshtml:register_server (1) trace:mshtml:load_gecko () trace:mshtml:install_cab (L"Z:\\usr\\bin\\..\\share\\wine\\gecko\\wine_gecko-1.1.0-rc1-x86.cab") trace:mshtml:check_version "Wine Gecko 1.1.0-rc1" trace:mshtml:load_xpcom (L"C:\\windows\\system32\\gecko\\1.1.0-rc1\\wine_gecko") fixme:system:SetProcessDPIAware stub! trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsISupports 0x33d374) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33d214) trace:mshtml:nsDirectoryServiceProvider_GetFile ("GreD" 0x33d2ec 0x33d2e8) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33d214) trace:mshtml:nsDirectoryServiceProvider_GetFile ("XCurProcD" 0x33d2ec 0x33d2e8) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33bfc4) trace:mshtml:nsDirectoryServiceProvider_GetFile ("PrfDef" 0x33c09c 0x33c098) warn:mshtml:nsDirectoryServiceProvider_QueryInterface ({2f977d4b-5485-11d4-87e2-0010a4e75ef2} 0x33bff4) warn:mshtml:nsDirectoryServiceProvider_QueryInterface ({2f977d4b-5485-11d4-87e2-0010a4e75ef2} 0x33bff4) fixme:iphlpapi:NotifyAddrChange (Handle 0xf5e914, overlapped 0xf5e918): stub trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33bc64) trace:mshtml:nsDirectoryServiceProvider_GetFile ("AChrom" 0x33bd3c 0x33bd38) err:mshtml:init_xpcom AutoRegister(NULL) failed: 80070057 trace:mshtml:nsIOService_GetProtocolHandler ("file" 0x33cff8) trace:mshtml:nsIOService_NewURI ("file:///C:/windows/system32/gecko/1.1.0-rc1/wine_gecko/components/nsTryToClose.js" (null) (nil) 0x33d15c) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33ccb4) trace:mshtml:nsDirectoryServiceProvider_GetFile ("ProfDS" 0x33cd8c 0x33cd88) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33ccb4) trace:mshtml:nsDirectoryServiceProvider_GetFile ("ProfD" 0x33cd8c 0x33cd88) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33ccb4) trace:mshtml:nsDirectoryServiceProvider_GetFile ("ProfLDS" 0x33cd8c 0x33cd88) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33ccb4) trace:mshtml:nsDirectoryServiceProvider_GetFile ("ProfLD" 0x33cd8c 0x33cd88) trace:mshtml:nsIOService_NewURI ("resource://gre/modules/XPCOMUtils.jsm" (null) (nil) 0x33acb8) trace:mshtml:nsDirectoryServiceProvider_QueryInterface (IID_nsIDirectoryServiceProvider 0x33851c) trace:mshtml:nsDirectoryServiceProvider_GetFile ("CurProcD" 0x3385f4 0x3385f0) trace:mshtml:nsIOService_NewFileURI (0x7613b8 0x338674) trace:mshtml:nsIOService_NewFileURI (0x7613b8 0x338674) trace:mshtml:nsIOService_NewURI ("jar:chrome/toolkit.jar!/res/" (null) 0x75f6c8 0x3386e4) trace:mshtml:nsIOService_ExtractScheme ("jar:chrome/toolkit.jar!/res/" 0x3363d0) trace:mshtml:nsIOService_NewURI ("chrome/toolkit.jar!/res/" "" 0x75f6c8 0x7613cc) trace:mshtml:nsIOService_NewURI not wraping trace:mshtml:nsIOService_NewChannelFromURI (0x75fa90 0x33acb4) trace:mshtml:nsIOService_NewChannelFromURI Could not get nsWineURI: 80004002 trace:mshtml:nsIOService_NewChannel ("file:///C:/windows/system32/gecko/1.1.0-rc1/wine_gecko/modules/XPCOMUtils.jsm" (null) (nil) 0x33acb4) trace:mshtml:close_gecko () wine: configuration in '/home/luca/.wine' has been updated. Hope it helps :)

Aug. 24, 2010

Re: ddraw [try 2]: SetCooperativeLevel with Full screen + exclusive + normal is authorized

by Henri Verbeet

This looks suspicious to me, you have to assume the code wasn't just added for the hell of it, even though with ddraw you're never quite sure. Do other ddraw versions behave the same? Please don't add trailing spaces, neither in the code nor in TRACEs.

Aug. 24, 2010

Re: d3dx9: Implement D3DXDeclaratorFromFVF. (try 2)

by Henri Verbeet

On 24 August 2010 05:05, Misha Koshelev <misha680(a)gmail.com> wrote: > + elements[idx].Stream = 0xFF; > + elements[idx].Offset = D3DXGetDeclVertexSize(elements, 0); > + elements[idx].Stream = 0; That's ugly. > + memcpy(Declaration, elements, size * sizeof(D3DVERTEXELEMENT9)); That's unnecessary.

Aug. 24, 2010

Re: ddraw [try 2]: SetCooperativeLevel with Full screen + exclusive + normal is authorized

by Stefan Dösinger

Hi, Can you add a test that tests in which mode ddraw is after such a call? Specfically, try to create a double buffered primary surface. If this succeeds, ddraw is in fullscreen mode. If ddraw is in normal mode you'll get DDERR_NOEXCLUSIVEMODE (or similar) back. Am 24.08.2010 um 10:00 schrieb David Adam: > Fixed a typo > > A+ > > David > <0001-SetCooperativeLevel-with-Full-screen-exclusive-norma.patch>

Aug. 24, 2010

Re: Use SUBLANG_NEUTRAL for the Gaelic resources?

by Huw Davies

On Tue, Aug 24, 2010 at 09:59:51AM +0200, Francois Gouget wrote: > > Are SUBLANG_GAELIC, SUBLANG_GAELIC_SCOTTISH and SUBLANG_GAELIC_MANX very > different languages? > > If not it would be better to use SUBLANG_NEUTRAL instead of > SUBLANG_GAELIC. For instance: I think they're different enough to warrant falling back to English rather than Irish (which is essentially what you're asking). See: http://en.wikipedia.org/wiki/Manx_language#Background Huw.

Aug. 24, 2010

Re: explorerframe/tests: Add a trailing '\n' to an ok() call.

by testbot＠testbot.winehq.org

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4711 Your paranoid android. === W2K8SE (32 bit nstc) === nstc.c:1889: Test failed: Got event 7, count 0 nstc.c:1891: Test failed: Got event 4, count 0 nstc.c:1898: Test failed: Got event 6, count 1 nstc.c:1904: Test failed: Got event 4, count 0 nstc.c:1956: Test failed: Got event 7, count 0 nstc.c:1965: Test failed: Item not expanded. nstc.c:1968: Test failed: itemstate is 0x00000000 nstc.c:1985: Test failed: itemstate is 0x00000000 nstc.c:2111: Test failed: Got event 7, count 0 nstc.c:2118: Test failed: itemstate is 0x00000000 nstc.c:2142: Test failed: itemstate is 0x0000000c nstc.c:2157: Test failed: itemstate is 0x0000000c nstc.c:2171: Test failed: itemstate is 0x00000004 nstc.c:2183: Test failed: itemstate is 0x00000004 nstc.c:2240: Test failed: Got event 7, count 0 nstc.c:2245: Test failed: Got (0x80004005) nstc.c:2246: Test failed: psi is 00000000

Aug. 24, 2010

Re: ddraw [try 2]: SetCooperativeLevel with Full screen + exclusive + normal is authorized

by testbot＠testbot.winehq.org

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4710 Your paranoid android. === W98SE (32 bit ddrawmodes) === No test summary line found === W7PROX64 (64 bit ddrawmodes) === Timeout

Aug. 24, 2010

Re: [PATCH 1/5] server: move is_removeable to fd_ops

by Alexandre Julliard

Mike Kaplinskiy <mike.kaplinskiy(a)gmail.com> writes: > On Wed, Aug 4, 2010 at 1:29 AM, Mike Kaplinskiy > <mike.kaplinskiy(a)gmail.com> wrote: >> --- >> server/change.c | 2 ++ >> server/device.c | 1 + >> server/fd.c | 18 ++++++++++++------ >> server/file.c | 1 + >> server/file.h | 3 +++ >> server/mailslot.c | 3 +++ >> server/mapping.c | 1 + >> server/named_pipe.c | 3 +++ >> server/process.c | 1 + >> server/protocol.def | 2 +- >> server/queue.c | 1 + >> server/request.c | 1 + >> server/serial.c | 1 + >> server/signal.c | 1 + >> server/sock.c | 1 + >> server/thread.c | 1 + >> 16 files changed, 34 insertions(+), 7 deletions(-) >> > > Could I have some feedback on this? I haven't had the time to go on > IRC at work and US Eastern is not a widely used timezone around here > :-\. I've been working on a better way to achieve this, it should be committed shortly. -- Alexandre Julliard julliard(a)winehq.org

Aug. 24, 2010

Re: How managing crashing tests ?

by GOUJON Alexandre

On 08/23/2010 11:47 PM, Andrew Eikum wrote: > Just don't execute the code on that platforms that don't support it. > How to check if the platform supports that codepath can be tricky, and > there are a lot of ways to do it. Some examples are to use > GetProcAddress; use a different function (if it's not the one being > tested, obviously); or call the function in a certain way that can tip > you off to platform differences. > > If you have a specific problem in mind, people on the list might be > able to help. I read again the test file and in my case, there are some broken() /* NT4 */ so I'll declare a BOOL isNT4 and add a win_skip when appropriate. Many thanks !

Aug. 24, 2010

Use SUBLANG_NEUTRAL for the Gaelic resources?

by Francois Gouget

Are SUBLANG_GAELIC, SUBLANG_GAELIC_SCOTTISH and SUBLANG_GAELIC_MANX very different languages? If not it would be better to use SUBLANG_NEUTRAL instead of SUBLANG_GAELIC. For instance: dlls/mapi32/Ga.rc | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/dlls/mapi32/Ga.rc b/dlls/mapi32/Ga.rc index e8aa3c0..104a658 100644 --- a/dlls/mapi32/Ga.rc +++ b/dlls/mapi32/Ga.rc @@ -22,7 +22,7 @@ #include "res.h" -LANGUAGE LANG_GAELIC, SUBLANG_GAELIC +LANGUAGE LANG_GAELIC, SUBLANG_NEUTRAL STRINGTABLE { -- 1.7.1

Aug. 24, 2010

Re: ddraw/tests: New visual back buffer flipping tests. (resend)

by GOUJON Alexandre

On 08/23/2010 11:28 PM, Oldřich Jedlička wrote: > My mistake, I've mixed up the C and C++ comments. I thought // is the C style > (for a while). I will send the fixed patch In fact, // is not C++ only : it has been added in the C99 revision. But for compatibility, we don't accept C99. Have a look at http://wiki.winehq.org/SubmittingPatches, there are some interesting infos there.

Aug. 24, 2010

Re: Wine Gecko 1.1.0 RC1

by Paul "TBBle" Hampson

On 24 August 2010 05:00, Jacek Caban <jacek(a)codeweavers.com> wrote: > The attached patch contains remaining bits. Haven't tested it, but I was having a flick through, and noticed that the second hunk for dlls/mshtml/nsembed.c has: ERR("Could not get nsIDontent interface: %08x\n", nsres); instead of ERR("Could not get nsIContent interface: %08x\n", nsres); -- Paul "TBBle" Hampson, Paul.Hampson(a)Pobox.com

Aug. 24, 2010

Re: user32: Optimize the storage of default values

by Dmitry Timoshkov

André Hentschel <nerv(a)dawncrow.de> wrote: > -static const char * const DefSysColors[] = > +static struct { > + const char *name; > + const COLORREF rgb; > +} DefSysColors[] = It should be 'static const struct {'. -- Dmitry.

Aug. 24, 2010

Re: RFC: test of 16 bit wine (DOS)

by Dmitry Timoshkov

Morten Rønne <morten.roenne(a)tdcadsl.dk> wrote: > 16 bit programs require a compiler that is able to create the old style > address method in order to be able to test those parts of wine. I have > been told by Dmitry Timoshkov that I should use OpenWatcom for that. So > that is what I have done. I didn't tell you that. I told you that you don't need any compiler at all for the tests to be integrated into the Wine make system. Obviously you need the compiler (or rather assembler) when you create the test, but then you just convert it into a binary blob which you copy into %TEMP% and run from there. -- Dmitry.

Aug. 24, 2010

Re: ddraw/tests: New visual back buffer flipping tests. (try 2)

by Oldřich Jedlička

On Tuesday 24 August 2010 00:14:31 Octavian Voicu wrote: > 2010/8/24 Oldřich Jedlička <oldium.pro(a)seznam.cz>: > > + ok((GetRValue(color) == 0xFF && GetGValue(color) == 0xFF && > > GetBValue(color) == 0xFF) || + broken(GetRValue(color) == 0xFF > > && GetGValue(color) == 0 && GetBValue(color) == 0), // broken driver + > > "got R %02X G %02X B %02X, expected R FF G FF B FF\n", + > > GetRValue(color), GetGValue(color), GetBValue(color)); > > I hate to be the bearer of bad news, but you missed one :) Fixed in "try 3" :-) Oldřich.

Aug. 24, 2010

Re: [PATCH 1/5] server: move is_removeable to fd_ops

by Mike Kaplinskiy

Aug. 24, 2010

Re: [PATCH 1/2] quartz/tests: move shareable code to a separate file

by James McKenzie

Anton Khirnov wrote: > On Sat, Aug 21, 2010 at 07:50:50PM -0700, James McKenzie wrote: > >> Looks like you did not test the build before running it against the >> testbot (I've been guilty of this as well.) >> >> > actually i did test it and it worked for some reason, must be magic :) > in any case, after rebasing it doesn't -- and the reason seems to be > that the build system assumes that each .c file under tests/ actually > contains a test. is there a way around that (besides the > obvious-but-ugly empty test in the common file)? > > One hint is if you are introducing a new function that exists in MSDN, look for an existing test case. If there is none, one will be required before Wine implementation code will be accepted. If the test exists in a file, then the implementing code should go in the corresponding .c file in the base dll directory from what I've read here over the years. If you are introducing a new function class, then it may be best to put the tests and code into new files. Empty test case files are frowned upon unless there really is no way to successfully test what you are trying to implement, but that is extremely rare. It is best to ask in the Wine Development list as to the most appropriate manner of implementation. The folks in the list are much more knowledgeable about this, I'm just giving suggestions. James McKenzie

Aug. 24, 2010

re: RFC: test of 16 bit wine (DOS)

by Dan Kegel

Morten Rønne wrote: > I have been working on creating tests for the 16 bit implementation of wine Great! Have you seen http://code.google.com/p/win16test/ ? That's for Win16 binaries, not DOS, but it might have some similarities. For instance, it also uses openwatcom. I had trouble back then getting the Linux version of openwatcom to produce win16 binaries, so I used 'winetricks openwatcom' to install the windows version. - Dan

Aug. 23, 2010

Re: ddraw: SetCooperativeLevel with Full screen + exclusive + normal is authorized

by testbot＠testbot.winehq.org

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4696 Your paranoid android. === W98SE (32 bit ddrawmodes) === No test summary line found === W7PROX64 (64 bit ddrawmodes) === Timeout

Aug. 23, 2010

Re: ddraw/tests: New visual back buffer flipping tests. (try 2)

by Octavian Voicu

2010/8/24 Oldřich Jedlička <oldium.pro(a)seznam.cz>: > + ok((GetRValue(color) == 0xFF && GetGValue(color) == 0xFF && GetBValue(color) == 0xFF) || > + broken(GetRValue(color) == 0xFF && GetGValue(color) == 0 && GetBValue(color) == 0), // broken driver > + "got R %02X G %02X B %02X, expected R FF G FF B FF\n", > + GetRValue(color), GetGValue(color), GetBValue(color)); I hate to be the bearer of bad news, but you missed one :)

Aug. 23, 2010

Re: gdiplus: add a stub for GdipWindingModeOutline

by Austin English

On Mon, Aug 23, 2010 at 4:47 PM, Vincent Povirk <madewokherd(a)gmail.com> wrote: > You need to add it to the .spec file too. Thanks. Bad cut and paste... -- -Austin

Aug. 23, 2010

Re: gdiplus: add a stub for GdipWindingModeOutline

by Vincent Povirk

You need to add it to the .spec file too. On Mon, Aug 23, 2010 at 4:45 PM, Austin English <austinenglish(a)gmail.com> wrote: > -- > -Austin > > > >

Aug. 23, 2010

Re: How managing crashing tests ?

by Andrew Eikum

On 08/23/2010 12:59 PM, GOUJON Alexandre wrote: > Anyway, I've seen mainly two approaches > - if(0) > - /* comment > the crashing ok() > */ > > What's the difference ? And why not #if 0 ? Using if(0) means the code in the body of the if-statement still has to compile. This can check against errors while modifying headers, help protect against misplaced braces in some cases, etc. Commenting the code out loses that verification, to no advantage. Really, it doesn't matter much. > I maybe lack some basic C knowledges but I think that /**/ and #if 0 are > not added to the executable (from preprocessing stage). > If it's right, it means we can then 'enable' these blocks maybe with > (gdb) edit something Likely not, since the optimizer probably just removes the body of the if-statement entirely. > And what about a test crashing on a specific Windows ? Just don't execute the code on that platforms that don't support it. How to check if the platform supports that codepath can be tricky, and there are a lot of ways to do it. Some examples are to use GetProcAddress; use a different function (if it's not the one being tested, obviously); or call the function in a certain way that can tip you off to platform differences. If you have a specific problem in mind, people on the list might be able to help. Andrew

Aug. 23, 2010

Re: ddraw/tests: New visual back buffer flipping tests. (resend)

by Oldřich Jedlička

On Monday 23 August 2010 23:00:24 Octavian Voicu wrote: > 2010/8/23 Oldřich Jedlička <oldium.pro(a)seznam.cz>: > > + // Check it out > > You have a C++ style comment in this patch and I don't think it's allowed. My mistake, I've mixed up the C and C++ comments. I thought // is the C style (for a while). I will send the fixed patch. Oldřich. > > Octavian

Aug. 23, 2010

Re: ddraw/tests: New visual back buffer flipping tests. (resend)

by Octavian Voicu

2010/8/23 Oldřich Jedlička <oldium.pro(a)seznam.cz>: > + // Check it out You have a C++ style comment in this patch and I don't think it's allowed. Octavian

Aug. 23, 2010