October 2011 - Wine-Devel - list.winehq.org

Re: vbscript: Added support for hex literals not ending with '&'

by Marvin

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=14970 Your paranoid android. === WNT4WSSP6 (32 bit) === No test summary line found === W2KPROSP4 (32 bit) === No test summary line found === WXPPROSP3 (32 bit) === No test summary line found === W2K3R2SESP2 (32 bit) === No test summary line found === WVISTAADM (32 bit) === No test summary line found === W2K8SE (32 bit) === No test summary line found === W7PRO (32 bit) === No test summary line found === W7PROX64 (32 bit) === No test summary line found === TEST64_W7SP1 (32 bit) === No test summary line found === W7PROX64 (64 bit) === No test summary line found === TEST64_W7SP1 (64 bit) === No test summary line found

Oct. 18, 2011

Re: Regression testing breakthrough

by Henri Verbeet

On 18 October 2011 10:45, Damjan Jovanovic <damjan.jov(a)gmail.com> wrote: > (especially during "reverse regression testing"), users find it too long and > technical, and only a small minority of regressions are ever bisected. And Not true. Even for the regressions that are still open it's currently 276 bisected vs. 99 not bisected, and Alexandre said about 90% of opened regressions get fixed. > I haven't figured out how to make the binaries available to users. Few users > can clone a 26 gigabyte repository, and even fewer places can serve that > much to multiple users. Maybe Git can compress it further? The other idea I > had is that users should be able to regression test through a GUI tool. > Maybe the GUI tool can just download and run the +/- 122 MB binary snapshots > for specific commits, instead of having the entire binary repository > locally? > > Any other ideas? Would you like to see this tool? Can I send an attachment > with it? > Scripts for making old versions compile may be useful, but for the most part it sounds like you're essentially duplicating ccache.

Oct. 18, 2011

Re: Regression testing breakthrough

by Ferenc Gergely Szilagyi

On Tue, Oct 18, 2011 at 10:45 AM, Damjan Jovanovic <damjan.jov(a)gmail.com>wrote: > Hi > > Since the beginning, I've had issues with regression testing. Despite the > fact it's very useful, it takes forever, it's easy to make a mistake > (especially during "reverse regression testing"), users find it too long and > technical, and only a small minority of regressions are ever bisected. And > several patches need backporting to allow older versions of Wine to compile > and run on today's make, gcc, and libraries - this is the case even for the > 1.0.x releases from less than 3 years ago! > > The problem is of course compilation. "configure" takes at least 40 > seconds, without any way to speed it up on multi-core CPUs. "make" takes > 5 > minutes, and it's only taking longer as Wine gets bigger. Compilation is > fundamentally complex and technical to users. > > But what if we had precompiled binaries, and regression testing consisted > of just running different versions of Wine? > > Wine binaries take up about 122 MB and take over 5 minutes to compile. > There's now 35770 commits between 36def4af0ca85a1d0e66b5207056775bcb3b09ff > (Release 1.0) and "origin". That's about 4.4 terrabytes of storage and over > 4 months of compilation, if each of those versions had to be compiled and > installed into its own prefix, way beyond what most users are willing or > able to store or do. Most patches however end up affecting only a few binary > files in the end, and compiling successive versions allows "make" to be very > quick. > > So I've written a tool that compiles Wine and adds each commit's binaries > into a Git repository. It knows how to compile old versions of Wine > (currently as far back as 1.0). It knows that commits affecting only > ANNOUNCE, .gitignore, and files in dll/ or programs/ ending with .c and such > don't need to go through the endlessly slow "configure", only "make". It is > stateless: if interrupted, it can resume from the last successful commit. It > works around bugs in GNU make (you won't believe how many there are...). > > This tool compiled all 35000 or so commits from Wine 1.0 to around 4th > October 2011 in only 7 days, generating a Git repository of Wine binaries > that's only 26 gigabytes in size. Regression testing with binaries is a > pleasure: it takes only a few seconds :-) on each bisection. I bisected a 16 > step regression in just 20 minutes, and most of that time was spent running > the application and dealing with 2 X-server crashes. > > I haven't figured out how to make the binaries available to users. Few > users can clone a 26 gigabyte repository, and even fewer places can serve > that much to multiple users. Maybe Git can compress it further? The other > idea I had is that users should be able to regression test through a GUI > tool. Maybe the GUI tool can just download and run the +/- 122 MB binary > snapshots for specific commits, instead of having the entire binary > repository locally? > > Any other ideas? Would you like to see this tool? Can I send an attachment > with it? > > Thank you > Damjan Jovanovic > > Hi, while I agree that 26 GB is a lot for most people, it would be no problem for me. Long compilation times were exactly the most important factor that kept me away from testing regressions so far. And I think I am not the only one on this side, so even if the repository is that huge, I think it might still worth to grant at least a few people the possibility to clone it. If I could arrange with my VPS provider to buy more disk space, I would even be willing to hold a mirror of it online. I would also definiately like to see how it is all achieved. Regards, Ferenc Gergely Szilagyi

Oct. 18, 2011

Regression testing breakthrough

by Damjan Jovanovic

Hi Since the beginning, I've had issues with regression testing. Despite the fact it's very useful, it takes forever, it's easy to make a mistake (especially during "reverse regression testing"), users find it too long and technical, and only a small minority of regressions are ever bisected. And several patches need backporting to allow older versions of Wine to compile and run on today's make, gcc, and libraries - this is the case even for the 1.0.x releases from less than 3 years ago! The problem is of course compilation. "configure" takes at least 40 seconds, without any way to speed it up on multi-core CPUs. "make" takes > 5 minutes, and it's only taking longer as Wine gets bigger. Compilation is fundamentally complex and technical to users. But what if we had precompiled binaries, and regression testing consisted of just running different versions of Wine? Wine binaries take up about 122 MB and take over 5 minutes to compile. There's now 35770 commits between 36def4af0ca85a1d0e66b5207056775bcb3b09ff (Release 1.0) and "origin". That's about 4.4 terrabytes of storage and over 4 months of compilation, if each of those versions had to be compiled and installed into its own prefix, way beyond what most users are willing or able to store or do. Most patches however end up affecting only a few binary files in the end, and compiling successive versions allows "make" to be very quick. So I've written a tool that compiles Wine and adds each commit's binaries into a Git repository. It knows how to compile old versions of Wine (currently as far back as 1.0). It knows that commits affecting only ANNOUNCE, .gitignore, and files in dll/ or programs/ ending with .c and such don't need to go through the endlessly slow "configure", only "make". It is stateless: if interrupted, it can resume from the last successful commit. It works around bugs in GNU make (you won't believe how many there are...). This tool compiled all 35000 or so commits from Wine 1.0 to around 4th October 2011 in only 7 days, generating a Git repository of Wine binaries that's only 26 gigabytes in size. Regression testing with binaries is a pleasure: it takes only a few seconds :-) on each bisection. I bisected a 16 step regression in just 20 minutes, and most of that time was spent running the application and dealing with 2 X-server crashes. I haven't figured out how to make the binaries available to users. Few users can clone a 26 gigabyte repository, and even fewer places can serve that much to multiple users. Maybe Git can compress it further? The other idea I had is that users should be able to regression test through a GUI tool. Maybe the GUI tool can just download and run the +/- 122 MB binary snapshots for specific commits, instead of having the entire binary repository locally? Any other ideas? Would you like to see this tool? Can I send an attachment with it? Thank you Damjan Jovanovic

Oct. 18, 2011

Re: [4/7] wined3d: YUY2 and UYVY are block based

by Stefan Dösinger

On Monday 17 October 2011 20:32:43 Henri Verbeet wrote: > That's mostly a limitation of the individual blitters. You could > certainly make YUY2 and UYVY work with e.g. FBO blits as long as you > only blit on block boundaries, and I wouldn't be all that surprised if > you have to blit on block boundaries anyway in D3D. Only as long as you don't have to filter. When the sizes don't match even point filters break, even if you're reading and writing on block boundaries. But you could handle that in a fragment shader if the circumstances allow you to calculate the values for the whole block in each shader run. Should be possible for blits at least.

Oct. 17, 2011

Re: Still need patch review please (wineps.drv)

by Daniel Santos

Oh, but I forgot to add that it does fix the problem (Finale 2010 prints dashed lines correctly now). http://bugs.winehq.org/show_bug.cgi?id=26091 On 10/17/2011 03:53 PM, Daniel Santos wrote: > Of particular concern is making sure my dynamically allocated memory is > being freed. The new data is stored in PSPEN and is allocated when > SelectObject (on a wineps.drv device of course) passing a pen and is > freed whenever a subsequent pen is selected or the device context is > deleted. > > I'm just not intimately familiar enough with this part of the API to > know for certain that this eliminates all room for leaks. > > Thanks! > Daniel

Oct. 17, 2011

Still need patch review please (wineps.drv)

by Daniel Santos

Of particular concern is making sure my dynamically allocated memory is being freed. The new data is stored in PSPEN and is allocated when SelectObject (on a wineps.drv device of course) passing a pen and is freed whenever a subsequent pen is selected or the device context is deleted. I'm just not intimately familiar enough with this part of the API to know for certain that this eliminates all room for leaks. Thanks! Daniel

Oct. 17, 2011

Re: [4/7] wined3d: YUY2 and UYVY are block based

by Henri Verbeet

On 17 October 2011 19:54, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > The fixup is a separate issue. We can write to formats with fixups just fine if > the input is in the correct format or we have a shader that can convert TO > that format. The problem with the blocks is that if one pixel is modified the > entire block may have to be modified, and we can't do that even with shaders. > So we can't blit to e.g. YUY2 surfaces even if the input is YUY2 as well and > the underlying luminance-alpha textures may be fbo attachable. > That's mostly a limitation of the individual blitters. You could certainly make YUY2 and UYVY work with e.g. FBO blits as long as you only blit on block boundaries, and I wouldn't be all that surprised if you have to blit on block boundaries anyway in D3D. YV12 would be harder, but at least d3d10/11 hardware could probably do it.

Oct. 17, 2011

Re: [4/7] wined3d: YUY2 and UYVY are block based

by Stefan Dösinger

On Monday 17 October 2011 15:38:18 Henri Verbeet wrote: > I think the "gl_" prefix is somewhat questionable since the format > being compressed or not is a property of the D3D format, it doesn't > depend on what GL format we use, at least in the way the flag is > currently used. You can change the meaning of the flag to be specific > to the GL format, but then you'd have to set it in the > format_texture_info[] table. The idea is to mark formats that need glCompressedTexImage2D instead of glTexImage2D, so format_texture_info sounds about right. > > + /* Block-based formats can't be rendered to with FBOs even if GL > > accepts them + * (e.g. WINED3DFMT_YV12, which is just a GL_ALPHA > > texture). Even fragment shaders + * can't account for that if a > > block is partially drawn. > > There's no such restriction on the GL side, the internal format just > has to be color-renderable. The reason we can't render directly to > these is because there a fixup associated with the format, but that > has nothing to do with either being FBO-attachable or being > block-based, and applies to e.g. P8 as well. The fixup is a separate issue. We can write to formats with fixups just fine if the input is in the correct format or we have a shader that can convert TO that format. The problem with the blocks is that if one pixel is modified the entire block may have to be modified, and we can't do that even with shaders. So we can't blit to e.g. YUY2 surfaces even if the input is YUY2 as well and the underlying luminance-alpha textures may be fbo attachable. But following the logic of the compressed flag above, that should also be part of the format defintion in format_texture_info[]. When a native GL format is used(e.g. GL_APPLE_ycbcr_422 or standard s3tc) the check if the format is FBO attachable should be used. I can also move this to a separate patch. On a somewhat related note, Windows 7 has overlay d3dx9 swapchains. The SDK claims that those may support YUV primaries, but according to http://www.virtualdub.org/blog/pivot/entry.php?id=282 and a follow-up post HW drivers only support RGB formats.

Oct. 17, 2011

Re: ntdll: avoid buffer overrun in lookup_manifest_file

by Alexandre Julliard

Dan Kegel <dank(a)kegel.com> writes: > @@ -1877,7 +1877,7 @@ static WCHAR *lookup_manifest_file( HANDLE dir, struct assembly_identity *ai ) > if (build == min_build && revision < min_revision) continue; > tmp = strchrW(tmp, '_') + 1; > tmp = strchrW(tmp, '_') + 1; > - if (!strcmpiW( tmp, wine_trailerW )) > + if (!strncmpiW( tmp, wine_trailerW, dir_info->FileNameLength/sizeof(WCHAR) - (tmp - dir_info->FileName) )) That doesn't do the same thing. -- Alexandre Julliard julliard(a)winehq.org

Oct. 17, 2011

Re: vbscript: Added lexer support for '_'

by Bruno Jesus

Hi, Jacek. On Mon, Oct 17, 2011 at 10:30, Jacek Caban <jacek(a)codeweavers.com> wrote: > Hi Bruno, > > On 10/14/11 20:07, Bruno Jesus wrote: > ... > I've tested it and that doesn't seem to be true. You're right, in a vbs file the limit does not exist, it looks like to be a Visual Basic 6 editor limit. Sorry for the noise. >> And this is another test that must pass too (there is no new operand >> after the last _ ) : >> x = _ >> 3 _ >> + 3 _ >> + 3 _ >> + 3 _ > > This will work only if the next line is empty (exactly like my > implementation does). Great =) > Jacek > Best wishes, Bruno

Oct. 17, 2011

Re: [4/7] wined3d: YUY2 and UYVY are block based

by Henri Verbeet

On 16 October 2011 23:05, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + BOOL gl_compressed; I think the "gl_" prefix is somewhat questionable since the format being compressed or not is a property of the D3D format, it doesn't depend on what GL format we use, at least in the way the flag is currently used. You can change the meaning of the flag to be specific to the GL format, but then you'd have to set it in the format_texture_info[] table. > static const struct wined3d_format_compression_info format_compression_info[] = > { > - {WINED3DFMT_DXT1, 4, 4, 8}, > - {WINED3DFMT_DXT2, 4, 4, 16}, > - {WINED3DFMT_DXT3, 4, 4, 16}, > - {WINED3DFMT_DXT4, 4, 4, 16}, > - {WINED3DFMT_DXT5, 4, 4, 16}, > - {WINED3DFMT_ATI2N, 4, 4, 16}, > + {WINED3DFMT_DXT1, 4, 4, 8, TRUE }, > + {WINED3DFMT_DXT2, 4, 4, 16, TRUE }, > + {WINED3DFMT_DXT3, 4, 4, 16, TRUE }, > + {WINED3DFMT_DXT4, 4, 4, 16, TRUE }, > + {WINED3DFMT_DXT5, 4, 4, 16, TRUE }, > + {WINED3DFMT_ATI2N, 4, 4, 16, TRUE }, > + > + {WINED3DFMT_YUY2, 2, 1, 4, FALSE }, > + {WINED3DFMT_UYVY, 2, 1, 4, FALSE }, > }; If you decouple "block-based" and "compressed", it doesn't make sense anymore to set them in the same table like this. The "compressed" property should go to either the format_texture_info[] or the format_base_flags[] table, depending on whether it's considered a basic property of the format, or specific to the GL format we use for it. > + /* Block-based formats can't be rendered to with FBOs even if GL accepts them > + * (e.g. WINED3DFMT_YV12, which is just a GL_ALPHA texture). Even fragment shaders > + * can't account for that if a block is partially drawn. There's no such restriction on the GL side, the internal format just has to be color-renderable. The reason we can't render directly to these is because there a fixup associated with the format, but that has nothing to do with either being FBO-attachable or being block-based, and applies to e.g. P8 as well. > + * > + * Furthermore some drivers crash instead of generating an error if we attempt to > + * attach s3tc textures to an FBO */ This was an issue with some versions of fglrx, but should be fixed in current versions.

Oct. 17, 2011

Re: [3/7] d3d9/tests: Add a test for locking partial blocks with different pools

by Henri Verbeet

On 16 October 2011 23:01, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + hr = IDirect3DDevice9_CreateTexture(device, 4, 4, 1, > + pools[i].pool == D3DPOOL_DEFAULT ? D3DUSAGE_DYNAMIC : 0, > + D3DFMT_DXT5, pools[i].pool, &texture, NULL); Same comment about dynamic textures as for patch 2/7. > + rect.left = 0; rect.top = 0; rect.right = 2; rect.bottom = 2; I don't care strongly about this, but note that SetRect() exists.

Oct. 17, 2011

Re: [2/7] wined3d: Locking partial blocks is not allowed in DEFAULT pool resources

by Henri Verbeet

On 16 October 2011 22:56, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + hr = IDirect3DDevice9_CreateTexture(device, 128, 128, 1, D3DUSAGE_DYNAMIC, dxt_formats[i].fmt, > + D3DPOOL_DEFAULT, &texture, NULL); You should check D3DCAPS2_DYNAMICTEXTURES before using dynamic textures. Also, wouldn't it be more appropriate to make this part of the test in patch 3/7?

Oct. 17, 2011

Re: vbscript: Added lexer support for '_'

by Jacek Caban

Hi Bruno, On 10/14/11 20:07, Bruno Jesus wrote: > On Fri, Oct 14, 2011 at 13:44, Jacek Caban<jacek(a)codeweavers.com> wrote: >> --- >> dlls/vbscript/lex.c | 10 ++++++++++ >> dlls/vbscript/tests/lang.vbs | 7 +++++++ >> dlls/vbscript/tests/run.c | 2 ++ >> 3 files changed, 19 insertions(+), 0 deletions(-) >> > Hi, Jacek, I'm far away from my linux box and I'm just passing by to > remember that there is a 24 continuation limit. > > Here is an invalid operation because it has 26 continuations: > x = _ > 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 - _ > 3 + _ > 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 _ > + 3 I've tested it and that doesn't seem to be true. > And this is another test that must pass too (there is no new operand > after the last _ ) : > x = _ > 3 _ > + 3 _ > + 3 _ > + 3 _ This will work only if the next line is empty (exactly like my implementation does). Jacek

Oct. 17, 2011

Re: d3dx9_36: Buffer size does not depend on the size of the created buffer

by Henri Verbeet

This is because the buffer is supposed to be created by D3DXCreateBox().

Oct. 17, 2011

Re: riched20: Prevent setting an invalid default format rect

by Thomas Faber

Hey, On 2011-10-15 12:10, Thomas Faber wrote: > The patch fixes this by adding checks to ME_SetDefaultFormatRect to > use the minimal rectangle that is still valid in this case. I've made some tests (will post them later) and it turns out this rectangle is also invalid in Windows in this case, so this is not the correct solution. Please ignore the patch, I will try to come up with a better one soon. Regards, Tom

Oct. 17, 2011

Re: [patch] Support NULL puLen in VerQueryValue

by Rafael Espíndola

>> You should write some tests to examine the behavior of VerQueryValue >> with NULL puLen. >> > > I just found out that wine has a test for it, but it has been commented out: The attached patch also uncomment the tests. Cheers, Rafael

Oct. 15, 2011

Fwd: [Announce] Google Code-in 2011

by Austin English

---------- Forwarded message ---------- From: Carol Smith <carols(a)google.com> Date: Fri, Oct 14, 2011 at 15:28 Subject: [Announce] Google Code-in 2011 To: Google Summer of Code Mentors List <google-summer-of-code-mentors-list(a)googlegroups.com> Hi everyone, We’re pleased to announce that we are planning to run the Google Code-in contest again this November. We had a successful contest last year with 20 mentoring organizations and 361 participating students. We will again be giving 13-17 year old students around the world an opportunity to get involved with open source projects doing tasks ranging from documentation to outreach to coding programs. We’re looking forward to a fun contest this year and hope you’ll consider participating as a mentor and/or org admin yourself, and spreading the word to friends, family, and colleagues about the program. Please check out the timeline [0] for more details on the program. We will soon be accepting applications for open source organizations to become mentoring organizations for this year’s contest. There will be a separate email thread to this list inviting you to apply to be a mentoring organization on Monday, 17 October. Please take some time to discuss with your respective organizations about whether you feel you have the time and mentor capacity to participate this year. We won’t kid you: working with teengers can be demanding. It’s important that your mentors/org be prepared to put the time in over the holiday, but if you participate you will all have the satisfaction of sharing your project and your passion for open source with a truly amazing collection of young people, and getting them started in our community. We've put together an Information Page [1] for this year's program to give you some lessons learned from last year's program. *Please* look this over before deciding to apply to participate in the program. We encourage you to join the Google Code-in contest discussion list [2] and announcement list [3]. Please see our blog post [4] for further details and contact us directly if you have specific questions that aren’t answered in the FAQs [5]. [0] - http://www.google-melange.com/gci/events/google/gci2011 [1] - http://code.google.com/p/google-code-in/wiki/GCIAdminMentorInformation [2] - http://groups.google.com/group/gci-discuss [3] - http://groups.google.com/group/gci-announce [4] - http://google-opensource.blogspot.com/2011/10/google-code-in-are-you-in.html [5] - http://www.google-melange.com/gci/document/show/gci_program/google/gci2011/… Cheers, The Google OSPO Team -- You received this message because you are subscribed to the Google Groups "Google Summer of Code Mentors List" group. To post to this group, send email to google-summer-of-code-mentors-list(a)googlegroups.com. To unsubscribe from this group, send email to google-summer-of-code-mentors-list+unsubscribe(a)googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-summer-of-code-mentors-list?hl=en. -- -Austin

Oct. 14, 2011

Re: [6/6] wined3d: Catch nop palette updates

by Stefan Dösinger

On Thursday 13 October 2011 22:27:46 Henri Verbeet wrote: > What makes it slow? Is it purely the upload of the palette itself, or > is this because e.g. realize_palette() causes a download + upload of > the surface? It's the upload in realize_palette. On my system it isn't really an issue with the opengl renderer, even though the color conversion is done only in software. The gdi renderer profits more from eliminating the redundant changes(The game calls SetPaletteEntries with the same entries about 4 times per second)

Oct. 14, 2011

Re: vbscript: Added lexer support for '_'

by Bruno Jesus

On Fri, Oct 14, 2011 at 13:44, Jacek Caban <jacek(a)codeweavers.com> wrote: > --- > dlls/vbscript/lex.c | 10 ++++++++++ > dlls/vbscript/tests/lang.vbs | 7 +++++++ > dlls/vbscript/tests/run.c | 2 ++ > 3 files changed, 19 insertions(+), 0 deletions(-) > Hi, Jacek, I'm far away from my linux box and I'm just passing by to remember that there is a 24 continuation limit. Here is an invalid operation because it has 26 continuations: x = _ 3 _ + 3 _ + 3 _ + 3 _ + 3 - _ 3 + _ 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 _ + 3 And this is another test that must pass too (there is no new operand after the last _ ) : x = _ 3 _ + 3 _ + 3 _ + 3 _ As I said I can't test in wine but maybe you will find this information useful. Best wishes, Bruno -- universe* god::bigbang (void); //and then it all began...

Oct. 14, 2011

Re: 79927: [3/3] cmd: Make RENAME work for read-only files

by Frédéric Delanoy

2011/10/14 Dan Kegel <dank(a)kegel.com>: > gcc complains here: > builtins.c: In function 'WCMD_rename': > builtins.c:1782:9: error: unused variable 'attribs' [-Werror=unused-variable] > OK resent Thx

Oct. 14, 2011

Re: 79927: [3/3] cmd: Make RENAME work for read-only files

by Dan Kegel

gcc complains here: builtins.c: In function 'WCMD_rename': builtins.c:1782:9: error: unused variable 'attribs' [-Werror=unused-variable] On Fri, Oct 14, 2011 at 2:27 AM, <buildbot(a)kegel.com> wrote: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/383 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed shell_2 > > Errors: > builtins.c: In function 'WCMD_rename': > builtins.c:1782:9: error: unused variable 'attribs' [-Werror=unused-variable] > make[1]: *** [builtins.o] Error 1 > make: *** [programs/cmd] Error 2 > make: *** Waiting for unfinished jobs.... > >

Oct. 14, 2011

Re: ddraw: Keep track of interface attached by AddAttachedSurface and detach correct interface when parent is released. (try 3)

by Octavian Voicu

On Thu, Oct 13, 2011 at 11:12 AM, Saulius Krasuckas <saulius2(a)ar.fi.lt> wrote: > I am troubled by seeing (via test.winehq.org) your ddraw:refcount tests to > account only results for "VMware SVGA 3D" adapter/driver (and only those > with 3D acceleration disabled), I guess. > > But if this driver is 3D-enabled or just present real physical adapter > (with no 3d acceleration) then ddraw:refcount contains 6 static failuDDSCAPS_3DDEVICEres > such as: Hi Saulius, Thanks for pointing this out. I only tested with testbot and I guess all of the machines use that adapter (at least the default ones do), so that's why I didn't catch the error. The application that I was debugging used stencil buffers, so I sticked with those -- although I did have a feeling it's not a good idea to test using DDSCAPS_3DDEVICE. I'm going to write a patch to fix the tests by switching to more general surface types and maybe also move the tests to dsurface.c. @Henri: I was in doubt between refcount.c and dsurface.c when writing the tests, but refcount.c allowed me to write the tests for both IDirectDrawSurface and IDirectDrawSurface7 in the same function and so reuse the caps for them. Now that I QI the v1 interface, maybe I can still combine them somehow. Octavian

Oct. 14, 2011

Re: [2/6] wined3d: Pass the destination surface to blitters

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 13.10.2011 um 22:27 schrieb Henri Verbeet: > It's mostly a matter of getting rid of set_shader() / unset_shader() > and introducing calls for doing color and depth blits. This shouldn't > be much of a surprise, I'm pretty sure we looked at this when Roderick > was still working on the blit code. No surprise there. The main question I have is, how do you handle blitting offscreen back/frontbuffers to the screen, and how do you select the correct color fixup shader for this? > On 13 October 2011 20:58, Stefan Dösinger <stefan(a)codeweavers.com> wrote: >> initiates the data transfer(That could include blitters that call >> glTexSubImage2D when blitting from sysmem->vidmem) >> > That's not necessarily something that belongs in individual blitters > either. The idea is to make the blitters do GPU -> GPU transfers only, > and do CPU <-> GPU mostly as uploads / downloads. I have some patches > that go in that direction already. Do you mean catch it in wined3d_surface_blt with checks that test for those operations? Or do you mean keep the current software blt then upload the entire destination surface? In the end CPU <-> GPU blits should call surface_upload_data/surface_download_data, no question about that. I'd just use wined3d_select_blitter to select a blitter that calls those functions instead of explicit checks in wined3d_surface_blit. -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) iQIcBAEBAgAGBQJOmARJAAoJEN0/YqbEcdMw+HsP/21icwJaEUXEFdDdVGGffmb0 dqnyH29cnV7do24OH44ABtv+4cq35oepg0NyRLxcPjTqSLsQgiQnCTB80oRZVw+J 4TuaXPTn+FH+eIzJRp1lpjO8LhY78iTMwqZE+ZlRGz/fST3z0kHSuXQTFVZKnME/ RfVrmRn/0xULM6Ey0rf63iPhcmXdmgshE/vV3mX2/j5yTakwjoV0FmpRz4eibbpq 2OdaNcYHXXdtFqBArLZp6y5DslQUWTjqc0U41t8XndCL2g0xxr5uEXcSxqd+hbgA 4KVubj/13ni86BGavCxoQI+fjosYLdZ1kQ2J/u4jXK9lfOvQw5ESl895d+yrblyR kyfRhZBKlmfT94jAfGaClqsAVKWcPf52hc8uBljhjwfTGB6TpdaTZSSPmny4eGVi b8uN9146HbkboYMiyKZvxSLwBzhVcngFCRJgpbvKm4has8G4m1NjL15OCLL3pfu9 Xleq3czqvrNLgx2kq+gXDPlOeeyyeElsYebfA5wigh7Ioxo7ZyX3cqexacPuFOMN oMt2q4QgkMdmVzvIyLfRh5kOdjkEfYzOAniqPDDGKqLZXl3d8ndCiIAI3aITLEW2 AFEvxHrughwxac9PBQv3UTazlHJwg5/R7B5LuGbD/N09m5axs02Pz0k+N7fUBztk zwJ9JqrChMlbW6BDrlZF =kBZI -----END PGP SIGNATURE-----

Oct. 14, 2011

Update: One more review: wineps.drv: Implement PS_USERSTYLE on printers

by Daniel Santos

Sorry, I left an fprintf in that last patch. On 10/14/2011 02:57 AM, Daniel Santos wrote: > Is anybody intimately familiar with the life cycle of device context and > pens? I don't necessarily mean that you went to bed with either of > them, but that would be OK too. I mean, this is the 2010s, we're > supposed to be modern here. Specifically, I want to make sure I'm not > leaking my dynamically allocated memory block. This block is allocated > when the pen is selected (not when it's created). > > Thanks in advance > Daniel

Oct. 14, 2011

One more review: wineps.drv: Implement PS_USERSTYLE on printers

by Daniel Santos

Is anybody intimately familiar with the life cycle of device context and pens? I don't necessarily mean that you went to bed with either of them, but that would be OK too. I mean, this is the 2010s, we're supposed to be modern here. Specifically, I want to make sure I'm not leaking my dynamically allocated memory block. This block is allocated when the pen is selected (not when it's created). Thanks in advance Daniel

Oct. 14, 2011

Re: 79921: ntdll: Move adding VPROT_READ for Solaris to the same place where it's done for FreeBSD. Take 2.

by Dmitry Timoshkov

Dan Kegel <dank(a)kegel.com> wrote: > Fails to apply here: > $ patch -p1 < vprot.patch > patching file dlls/ntdll/virtual.c > Hunk #1 FAILED at 187. > 1 out of 2 hunks FAILED -- saving rejects to file dlls/ntdll/virtual.c.rej Becuase it depends on the 1/2 patch sent before. -- Dmitry.

Oct. 14, 2011

Re: 79921: ntdll: Move adding VPROT_READ for Solaris to the same place where it's done for FreeBSD. Take 2.

by Dan Kegel

Fails to apply here: $ patch -p1 < vprot.patch patching file dlls/ntdll/virtual.c Hunk #1 FAILED at 187. 1 out of 2 hunks FAILED -- saving rejects to file dlls/ntdll/virtual.c.rej On Thu, Oct 13, 2011 at 10:02 PM, <buildbot(a)kegel.com> wrote: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/373 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed git > > Errors: > error: patch failed: dlls/ntdll/virtual.c:187 > error: dlls/ntdll/virtual.c: patch does not apply > >

Oct. 14, 2011

Re: [6/6] wined3d: Catch nop palette updates

by Henri Verbeet

On 13 October 2011 20:40, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > On Thursday 13 October 2011 14:40:23 Henri Verbeet wrote: >> This looks questionable. Which applications does this help? > Age of Empires 2. It uploads the same palette regularly, which makes the main > menu quite unresponsive. > What makes it slow? Is it purely the upload of the palette itself, or is this because e.g. realize_palette() causes a download + upload of the surface?

Oct. 13, 2011

Re: [3/6] wined3d: Use the destination palette when blitting between P8 surfaces

by Henri Verbeet

On 13 October 2011 21:15, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > On Thursday 13 October 2011 14:40:09 Henri Verbeet wrote: >> Which makes much more sense. > And is the plan for offscreen P8 surfaces. It won't work for blits to the > frontbuffer(at the moment) or for the non-shader P8 surface conversion. > The point is mostly that don't see much point in extending the existing hacks (with all the potential for regressions associated with that) when it's clear that it's not the way it's supposed to work.

Oct. 13, 2011

Re: [2/6] wined3d: Pass the destination surface to blitters

by Henri Verbeet

On 13 October 2011 20:58, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > What does the interface you plan look like? > It's mostly a matter of getting rid of set_shader() / unset_shader() and introducing calls for doing color and depth blits. This shouldn't be much of a surprise, I'm pretty sure we looked at this when Roderick was still working on the blit code. > initiates the data transfer(That could include blitters that call > glTexSubImage2D when blitting from sysmem->vidmem) > That's not necessarily something that belongs in individual blitters either. The idea is to make the blitters do GPU -> GPU transfers only, and do CPU <-> GPU mostly as uploads / downloads. I have some patches that go in that direction already.

Oct. 13, 2011

Re: [1/6] wined3d: YUV formats are block-based

by Stefan Dösinger

On Thursday 13 October 2011 14:39:38 Henri Verbeet wrote: > On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > diff --git a/dlls/d3d9/tests/surface.c b/dlls/d3d9/tests/surface.c > > diff --git a/dlls/ddraw/tests/overlay.c b/dlls/ddraw/tests/overlay.c > > These tests don't prove anything, they pass on current Wine as well. Because height=1 blocks are quite uninteresting. You notice the 2x1 blocks when you try to lock a rectangle with an odd left or right coordinate. But only with DEFAULT pool surfaces and not SCRATCH ones. There are no DXTN default pool surfaces, and textures break the double unlock test. So the test needs splitting into a texture and surface part. Detours ftw. > > + {WINED3DFMT_YUY2, 2, 1, 4, FALSE }, > > + {WINED3DFMT_UYVY, 2, 1, 4, FALSE }, > > + /* YV12 actually has 2x2 blocks with 6 bytes per block, but the > > blocks + * are spread across the surface. */ > > + {WINED3DFMT_YV12, 1, 1, 1, FALSE}, > > Blocks with 1 height aren't really blocks. YV12 is a strange format in many ways. What I care about is that we don't write to it. I'll write some more tests for YV12. If Windows rejects 1:1 blits, or blits only the luminance data that'd make things much simpler.

Oct. 13, 2011

Re: [3/6] wined3d: Use the destination palette when blitting between P8 surfaces

by Stefan Dösinger

On Thursday 13 October 2011 14:40:09 Henri Verbeet wrote: > On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > Native transfers the 8 bit data without modifications to the destination > > surface and does the palette lookup when showing the final front buffer > > contents on the screen. > > Which makes much more sense. And is the plan for offscreen P8 surfaces. It won't work for blits to the frontbuffer(at the moment) or for the non-shader P8 surface conversion.

Oct. 13, 2011

Re: [2/6] wined3d: Pass the destination surface to blitters

by Stefan Dösinger

On Thursday 13 October 2011 14:39:58 Henri Verbeet wrote: > On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > -static HRESULT arbfp_blit_set(void *blit_priv, struct wined3d_context > > *context, struct wined3d_surface *surface) + static HRESULT > > arbfp_blit_set(void *blit_priv, struct wined3d_context *context, struct > > wined3d_surface *src, + const struct wined3d_surface *dst) > > Whitespace error. The parameter you're adding isn't used anywhere, not > even in the same patchset. Patch 3: - upload_palette(src, context); + upload_palette(src, dst, context); > Perhaps more important though is the point > that "set_shader" is pretty much the wrong interface for blitters and > should go away. Note that the blitter interface is something I'm > actually working on as part of the ddraw work, it's just that other > things keep coming up. What does the interface you plan look like? I had a short look at this and my proposal is to separate the blitters into two interfaces: One that sets the shader for the color conversion(blit_shader without colorfill and depthfill), and one that does the drawable setup and initiates the data transfer(That could include blitters that call glTexSubImage2D when blitting from sysmem->vidmem) The shader setup blitter could be hidden behind a quad draw blitter if you find a way to integrate swapchain_blit and surface_blt_to_drawable(especially frontbuffer texture -> frontbuffer drawable blit)

Oct. 13, 2011

Re: [6/6] wined3d: Catch nop palette updates

by Stefan Dösinger

On Thursday 13 October 2011 14:40:23 Henri Verbeet wrote: > This looks questionable. Which applications does this help? Age of Empires 2. It uploads the same palette regularly, which makes the main menu quite unresponsive.

Oct. 13, 2011

Re: wined3d: Refuse to create a shader unsupported by the backend.

by Alexandre Julliard

Matteo Bruni <mbruni(a)codeweavers.com> writes: > --- > dlls/d3d9/tests/shader.c | 22 ++++++++++++++++++++-- > dlls/wined3d/device.c | 2 ++ > dlls/wined3d/shader.c | 19 +++++++++++++++++++ > dlls/wined3d/wined3d_private.h | 1 + > 4 files changed, 42 insertions(+), 2 deletions(-) It breaks the tests here: ../../../tools/runtest -q -P wine -M d3d10.dll -T ../../.. -p d3d10_test.exe.so effect.c && touch effect.ok effect.c:2661: Test failed: D3D10CreateEffectFromMemory failed! wine: Unhandled page fault on read access to 0x00000000 at address 0x6869a2a3 (thread 0024), starting debugger... -- Alexandre Julliard julliard(a)winehq.org

Oct. 13, 2011

Re: [6/6] wined3d: Catch nop palette updates

by Henri Verbeet

This looks questionable. Which applications does this help?

Oct. 13, 2011

Re: [3/6] wined3d: Use the destination palette when blitting between P8 surfaces

by Henri Verbeet

On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > Native transfers the 8 bit data without modifications to the destination > surface and does the palette lookup when showing the final front buffer > contents on the screen. Which makes much more sense.

Oct. 13, 2011

Re: [2/6] wined3d: Pass the destination surface to blitters

by Henri Verbeet

On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > -static HRESULT arbfp_blit_set(void *blit_priv, struct wined3d_context *context, struct wined3d_surface *surface) > + static HRESULT arbfp_blit_set(void *blit_priv, struct wined3d_context *context, struct wined3d_surface *src, > + const struct wined3d_surface *dst) Whitespace error. The parameter you're adding isn't used anywhere, not even in the same patchset. Perhaps more important though is the point that "set_shader" is pretty much the wrong interface for blitters and should go away. Note that the blitter interface is something I'm actually working on as part of the ddraw work, it's just that other things keep coming up.

Oct. 13, 2011

Re: [1/6] wined3d: YUV formats are block-based

by Henri Verbeet

On 13 October 2011 01:07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > diff --git a/dlls/d3d9/tests/surface.c b/dlls/d3d9/tests/surface.c > diff --git a/dlls/ddraw/tests/overlay.c b/dlls/ddraw/tests/overlay.c These tests don't prove anything, they pass on current Wine as well. > + {WINED3DFMT_YUY2, 2, 1, 4, FALSE }, > + {WINED3DFMT_UYVY, 2, 1, 4, FALSE }, > + /* YV12 actually has 2x2 blocks with 6 bytes per block, but the blocks > + * are spread across the surface. */ > + {WINED3DFMT_YV12, 1, 1, 1, FALSE}, Blocks with 1 height aren't really blocks.

Oct. 13, 2011

Re: [PATCH 5/5] server/named_pipe.c: remove the ps_disconnected_server state, if DisconnectNamedPipe is called, the server i

by Marvin

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=14877 Your paranoid android. === WXPPROSP3 (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W2K3R2SESP2 (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === WVISTAADM (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W2K8SE (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PRO (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PROX64 (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === TEST64_W7SP1 (32 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PROX64 (64 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === TEST64_W7SP1 (64 bit pipe) === pipe.c:1605: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368

Oct. 13, 2011

Re: [PATCH 4/5] kernel32, ntdll: implement most of GetNamedPipeHandleState (try 3)

by Marvin

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=14876 Your paranoid android. === WXPPROSP3 (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W2K3R2SESP2 (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === WVISTAADM (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W2K8SE (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PRO (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PROX64 (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === TEST64_W7SP1 (32 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === W7PROX64 (64 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368 === TEST64_W7SP1 (64 bit pipe) === pipe.c:1561: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 1368

Oct. 13, 2011

Re: For review please: wineps.drv: Implement PS_USERSTYLE on printers

by Bruno Jesus

Hi, Daniel. On Thu, Oct 13, 2011 at 06:14, Daniel <danielfsantos(a)att.net> wrote: > Hello guys. I would like to get a little review on this patch, a follow > up to the patches that fix this problem on the screen. Specifically, > I'm most concerned about making sure my dynamically allocated memory is > freed under all conditions and there's no potential snafus that I'm not > aware of. Then, of course, if somebody sees something off, please do > comment. + int i; + EXTLOGPEN *elp = NULL; + INT size = GetObjectW( hpen, 0, NULL ); + + if (!size) return 0; + else if (size == sizeof(logpen)) { + GetObjectW( hpen, sizeof(logpen), &logpen ); + } else { Alexandre replied in your patch email regarding this part of the patch, take a look at it. + const char *pattern = "%d"; /* don't put a space in front of the 1st number */ Are you sure const is doing what you think it is? + pattern = " %d"; You're setting the pattern var inside the loop, isn't it better to start with "%d " and trim the string after the for loop? + if (elp) HeapFree( GetProcessHeap(), 0, elp ); There is no need to check elp before calling HeapFree, it takes care of null pointers. + BOOL own_dash; /* TRUE if we need to free dash */ You are using spaces, the rest of the variable declarations use tabs. I usually don't feel safe about reviewing patches so I would appreciate if anyone care to review my review. Best wishes, Bruno -- universe* god::bigbang (void); //and then it all began...

Oct. 13, 2011

Re: wined3d: recognize "Brian Paul" as a mesa card vendor

by Henri Verbeet

On 13 October 2011 02:23, Austin English <austinenglish(a)gmail.com> wrote: > See http://bugs.winehq.org/show_bug.cgi?id=28684 / > http://cgit.freedesktop.org/mesa/mesa/tree/src/mesa/main/getstring.c#n88 > Yeah, this makes sense. You'd probably mostly run into this if you're not using the regular dri based drivers, but that's valid, if somewhat obscure.

Oct. 13, 2011

Re: ddraw: Keep track of interface attached by AddAttachedSurface and detach correct interface when parent is released. (try 3)

by Henri Verbeet

On 13 October 2011 10:12, Saulius Krasuckas <saulius2(a)ar.fi.lt> wrote: > refcount.c:214: Test failed: AddAttachedSurface returned 8876000a > refcount.c:235: Test failed: AddAttachedSurface returned 8876000a > refcount.c:237: Test failed: Got refcount 1, expected 2 > refcount.c:262: Test failed: AddAttachedSurface returned 8876000a > refcount.c:275: Test failed: AddAttachedSurface returned 8876000a > refcount.c:277: Test failed: Got refcount 1, expected 2 > refcount: 112 tests executed (0 marked as todo, 6 failures), 0 skipped > It probably doesn't make sense to attach a depth buffer to a P8 offscreenplain surface. I think it would make more sense to move these tests to the existing tests for AddAttachedSurface() in dsurface.c.

Oct. 13, 2011

Re: [1/4] winex11.drv: Add on-screen PS_USERSTYLE support to pens (take 4)

by Frédéric Delanoy

On Thu, Oct 13, 2011 at 06:08, Daniel <danielfsantos(a)att.net> wrote: > Fixes on-screen portion of http://bugs.winehq.org/show_bug.cgi?id=26091 You should configure your mail/git config so your real name appears in patches, and not just your first name "Daniel"

Oct. 13, 2011

Re: mscoree: Register interfaces

by Alexandre Julliard

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > @@ -10,4 +10,6 @@ C_SRCS = \ > metahost.c \ > mscoree_main.c > > +IDL_R_SRCS = mscoree_v1.idl There's no reason to name the file v1, that's a leftover from stdole.tlb but it doesn't make sense here. -- Alexandre Julliard julliard(a)winehq.org

Oct. 13, 2011

Re: [1/3] winex11.drv: Add on-screen PS_USERSTYLE support to pens (bugbot feeding #5)

by Alexandre Julliard

Daniel <danielfsantos(a)att.net> writes: > @@ -42,29 +42,23 @@ HPEN X11DRV_SelectPen( PHYSDEV dev, HPEN hpen ) > X11DRV_PDEVICE *physDev = get_x11drv_dev( dev ); > LOGPEN logpen; > int i; > + EXTLOGPEN *elp = NULL; > + INT size = GetObjectW( hpen, 0, NULL ); > > - if (!GetObjectW( hpen, sizeof(logpen), &logpen )) > - { > - /* must be an extended pen */ > - EXTLOGPEN *elp; > - INT size = GetObjectW( hpen, 0, NULL ); > - > - if (!size) return 0; > - > + if (!size) return 0; > + else if (size == sizeof(logpen)) { > + GetObjectW( hpen, sizeof(logpen), &logpen ); > + physDev->pen.ext = 0; > + } else { You are changing too much here, there's no reason to duplicate the GetObject call for the standard case. Also please make sure to include your full name in your mail headers. -- Alexandre Julliard julliard(a)winehq.org

Oct. 13, 2011

Re: [msi] initialize variable to placate static analyzers

by Michael Stefaniuc

Hello Dan, Dan Kegel wrote: > Avoids the spurious compiler warning > > msi.c: In function 'MSI_ApplyPatchW': > msi.c:353:10: error: 'r' may be used uninitialized in this function > > Hans says "It's compiler confusion but I'm fine with a patch that initializes > r with ERROR_FUNCTION_FAILED." why don't you just initialize r in the declaration of r? It's the standard way of doing it and doesn't require a suboptimal comment (that warning is gcc version dependent). bye michael

Oct. 13, 2011

For review please: wineps.drv: Implement PS_USERSTYLE on printers

by Daniel

Hello guys. I would like to get a little review on this patch, a follow up to the patches that fix this problem on the screen. Specifically, I'm most concerned about making sure my dynamically allocated memory is freed under all conditions and there's no potential snafus that I'm not aware of. Then, of course, if somebody sees something off, please do comment. It does indeed completely fix http://bugs.winehq.org/show_bug.cgi?id=26091 (printed a cute piece of music with dashed lines). Thanks all! Daniel

Oct. 13, 2011

Re: [Wine] WineHQ database compromise

by Maarten Lankhorst

Hey, On 10/12/2011 12:46 AM, Josh Juran wrote: > On Oct 11, 2011, at 3:37 PM, Conan Kudo (ニール・ゴンパ) wrote: > >> On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: >> >>> Since bugzilla passwords were sent in cleartext anyway, I sincerely hope none of them were otherwise valuable. (Remember FireSheep?) >> Wait, what? Bugzilla sends passwords in cleartext? That isn't very smart... Is there no way to replace this with some sort of client based hashing or something? > To clarify, your browser sends your password to bugzilla in cleartext, since HTTPS isn't an option. > > Firesheep was a lesson that even once passwords are secure, session credentials are still vulnerable to sniffing. Some sites went to HTTPS-only sessions after that. > If I go to any https://*.winehq.org website I get the certificate for test.winehq.org , otherwise you could use the firefox https anywhere to force https on. Or better yet, force automatic redirect to https, with Strict-Transport-Security: https://hacks.mozilla.org/2010/08/firefox-4-http-strict-transport-security-… If winehq can't get more ips for every subdomain (ssl sucks), would the solution be moving it to https://winehq.org/{bugs,appdb,test,source} ? Cheers, Maarten

Oct. 13, 2011

Re: ddraw: Keep track of interface attached by AddAttachedSurface and detach correct interface when parent is released. (try 3)

by Saulius Krasuckas

Hello, Octavian. * On Tue, 11 Oct 2011, Octavian Voicu wrote: > > try 3: > - fix test failure by moving error check from *_DeleteAttachedSurface to > ddraw_surface_delete_attached_surface, after another check (thanks Henri). > > try 2: > - use IUnknown_Release(attached_iface) instead of manual dispatch, > as suggested by Henri; > - change type of attached_iface from void* to IUnknown* to reflect > fact that it's a pointer to an interface; > - add test to show that attaching a surface only attaches a specific > interface and return error if DeleteAttachedSurface is called with > another interface than the attached one (instead of detaching the > surface anyway). > > Fixes a bug that causes a stencil buffer leak in the game The Longest Journey. > Problem can be seen in the console output log attached to bug #11819. > This patch doesn't fix the original bug, but now game exits cleanly. > > The root of the problem is that when releasing a surface, its attached > surfaces are detached using IDirectDraw7_DeleteAttachedSurface, regardless > of what interface was originally attached. I am troubled by seeing (via test.winehq.org) your ddraw:refcount tests to account only results for "VMware SVGA 3D" adapter/driver (and only those with 3D acceleration disabled), I guess. But if this driver is 3D-enabled or just present real physical adapter (with no 3d acceleration) then ddraw:refcount contains 6 static failures such as: refcount.c:214: Test failed: AddAttachedSurface returned 8876000a refcount.c:235: Test failed: AddAttachedSurface returned 8876000a refcount.c:237: Test failed: Got refcount 1, expected 2 refcount.c:262: Test failed: AddAttachedSurface returned 8876000a refcount.c:275: Test failed: AddAttachedSurface returned 8876000a refcount.c:277: Test failed: Got refcount 1, expected 2 refcount: 112 tests executed (0 marked as todo, 6 failures), 0 skipped This relation can be easily tracked by looking at d3d9:visual or (if they aren't present) at opengl32:opgenl tests. Eg. failures occuring here: * Intel(R) 82865G Graphics Controller http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/xp_af-… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/xp_af-… * NVIDIA GeForce FX 5200 http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_s… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_s… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/xp_s2-… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/xp_s2-… * ATI Radeon HD 4800 Series (telling it has no IDirect3D[7] support for DDraw) http://test.winehq.org/data/a843acef5567bc30901612ce5013da543259b21b/win7_w… http://test.winehq.org/data/a843acef5567bc30901612ce5013da543259b21b/win7_w… * VMware SVGA II http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/2000_f… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/2000_f… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/2000_f… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/2000_f… * VMware SVGA 3D (telling it has not IDirect3D[7] support for DDraw) http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_f… http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_f… All other cases succeed. AFAICS all of them represent the only driver (with 3D support turned off): * VMware SVGA 3D http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_w… visual.c:194: Driver string: "vm3dum.dll" visual.c:195: Description string: "VMware SVGA 3D (Microsoft Corporation - WDDM)" visual.c:197: Device name string: "\\.\DISPLAY1" visual.c:199: Driver version 7.14.1.40 visual.c:12627: Tests skipped: Creating the device failed visual: 5 tests executed (0 marked as todo, 0 failures), 1 skipped. with following results: http://test.winehq.org/data/d0c38ece9db657431559024abb34285cd8b717b6/win7_w… refcount: 121 tests executed (0 marked as todo, 0 failures), 0 skipped. IMHO, it would be nice to conform to real world configuration. Wish you best luck with fixing your tests:) S.

Oct. 13, 2011

Re: 79869: [2/4] gdi32: remove PS_USERSTYLE FIXME and work-arounds (take 4)

by Dan Kegel

Still fails tests here. Sorry if I wasn't clear before. The wine tests are supposed to pass after *every* patch in your series. That means you have to merge patches 2 and 3 together, I think. On Wed, Oct 12, 2011 at 9:27 PM, <buildbot(a)kegel.com> wrote: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/297 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed shell_3 > > Errors: > alarum: failed command was ../../../wine gdi32_test.exe.so pen.c > pen.c:317: Test succeeded inside todo block: expected 7, got 7 > make: *** [pen.ok] Error 1 > * Call to xpconnect wrapped JSObject produced this error: * > * Call to xpconnect wrapped JSObject produced this error: * > GnuTLS error: GnuTLS internal error. > >

Oct. 13, 2011

Re: ntdll: Define PATH_MAX when needed

by Dmitry Timoshkov

André Hentschel <nerv(a)dawncrow.de> wrote: > Am 12.10.2011 23:53, schrieb Dan Kegel: > > When is PATH_MAX not defined? > > > > It's part of posix (see > > http://pubs.opengroup.org/onlinepubs/009695399/basedefs/limits.h.html > > ) > > > > Are you trying to build on Gnu Hurd? > > Yes that's for Hurd. Austin already pointed out dynamic allocation might be a better solution, so... Notice that PATH_MAX and MAX_PATH are completely different things going from different sources which can't be replaced by each other. -- Dmitry.

Oct. 13, 2011

Re: [Wine] WineHQ database compromise

by Marcus Meissner

On Thu, Oct 13, 2011 at 10:23:58AM +0200, Maarten Lankhorst wrote: > Hey, > > On 10/12/2011 12:46 AM, Josh Juran wrote: > > On Oct 11, 2011, at 3:37 PM, Conan Kudo (ニール・ゴンパ) wrote: > > > >> On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: > >> > >>> Since bugzilla passwords were sent in cleartext anyway, I sincerely hope none of them were otherwise valuable. (Remember FireSheep?) > >> Wait, what? Bugzilla sends passwords in cleartext? That isn't very smart... Is there no way to replace this with some sort of client based hashing or something? > > To clarify, your browser sends your password to bugzilla in cleartext, since HTTPS isn't an option. > > > > Firesheep was a lesson that even once passwords are secure, session credentials are still vulnerable to sniffing. Some sites went to HTTPS-only sessions after that. > > > If I go to any https://*.winehq.org website I get the certificate for test.winehq.org , otherwise you could use the firefox https anywhere to force https on. > > Or better yet, force automatic redirect to https, with Strict-Transport-Security: > https://hacks.mozilla.org/2010/08/firefox-4-http-strict-transport-security-… > > If winehq can't get more ips for every subdomain (ssl sucks), would the solution be moving it to https://winehq.org/{bugs,appdb,test,source} ? Or a wildcard SSL cert for *.winehq.org. Ciao, Marcus

Oct. 13, 2011

Re: [4/3] gdi32/tests: update test for PS_USERSTYLE

by Daniel

Yeah, I'll recombine them in a bit, but I didn't really remove the test, per-se, I removed the if/else that performs the test with todo_wine when the style is PS_USERSTYLE (see the line below, that, of course, is improperly indented -- why I don't allow single statements in if, else, while, etc. without a {} block in my own projects). Daniel On 10/12/2011 07:12 PM, Austin English wrote: > On Wed, Oct 12, 2011 at 19:06, Daniel <danielfsantos(a)att.net> wrote: >> OK, I hope [4/3] is acceptable. These are the test updates I forgot >> (removes a todo_wine). > Two problems: > A) It needs to be combined with the original patch, so that it passes make test. > B) Why remove the test? Just remove 'todo_wine' itself, since the test > now passes on wine.. >

Oct. 13, 2011

re: [4/3] gdi32/tests: update test for PS_USERSTYLE

by Dan Kegel

> OK, I hope [4/3] is acceptable. It's fine for humans, but for buildbot, you really need to send the whole patch series again, with the test fixes right in the same patch as the code that fixes them (so that the wine tree passes all tests after each patch in the series). Otherwise your code will not be tested.

Oct. 13, 2011

Re: [4/3] gdi32/tests: update test for PS_USERSTYLE

by Austin English

On Wed, Oct 12, 2011 at 19:06, Daniel <danielfsantos(a)att.net> wrote: > OK, I hope [4/3] is acceptable. These are the test updates I forgot > (removes a todo_wine). Two problems: A) It needs to be combined with the original patch, so that it passes make test. B) Why remove the test? Just remove 'todo_wine' itself, since the test now passes on wine.. -- -Austin

Oct. 13, 2011

Re: 79857: [2/3] gdi32: remove PS_USERSTYLE FIXME and work-arounds (take 3)

by Daniel

> Fails the gdi32/pen.ok test here, with error > pen.c:317: Test succeeded inside todo block: expected 7, got 7 > > Looks like you need to update a test. whoops, I forgot about my tests. Will do.

Oct. 12, 2011

Re: 79857: [2/3] gdi32: remove PS_USERSTYLE FIXME and work-arounds (take 3)

by Dan Kegel

Fails the gdi32/pen.ok test here, with error pen.c:317: Test succeeded inside todo block: expected 7, got 7 Looks like you need to update a test. On Wed, Oct 12, 2011 at 4:00 PM, <buildbot(a)kegel.com> wrote: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/280 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed shell_3 > > Errors: > alarum: failed command was ../../../wine gdi32_test.exe.so pen.c > pen.c:317: Test succeeded inside todo block: expected 7, got 7 > make: *** [pen.ok] Error 1 > * Call to xpconnect wrapped JSObject produced this error: * > * Call to xpconnect wrapped JSObject produced this error: * > GnuTLS error: GnuTLS internal error. > >

Oct. 12, 2011

[RFC] Use ICU in wine ?

by Rafał Mużyło

On Mon Oct 10 12:48:27 CDT 2011, André Hentschel wrote a few things: Well, in the bug you've mentioned (#5163) there was a link to that file in mono, but while it seems nice, just looking at mono code makes my eyes bleed and the perspective of rewriting it in perl would fill me with terror. There's also the question of how close did mono get to Windows in this case. Still, that would at most fixed the default table, the tailoring needs to be implemented anyway.

Oct. 12, 2011

Re: ntdll: Define PATH_MAX when needed

by André Hentschel

Am 12.10.2011 23:53, schrieb Dan Kegel: > When is PATH_MAX not defined? > > It's part of posix (see > http://pubs.opengroup.org/onlinepubs/009695399/basedefs/limits.h.html > ) > > Are you trying to build on Gnu Hurd? Yes that's for Hurd. Austin already pointed out dynamic allocation might be a better solution, so... -- Best Regards, André Hentschel

Oct. 12, 2011

Re: wined3d: Report every texture format as filterable and blendable.

by Matteo Bruni

2011/10/12 Dan Kegel <dank(a)kegel.com>: > Fails d3d9's visual tests here. > > os: Linux 3.0.0-12-generic, Ubuntu 11.10 > gpu: GeForce GT 240/PCI/SSE2 3.3.0 NVIDIA 280.13 > > On Wed, Oct 12, 2011 at 2:52 PM, <buildbot(a)kegel.com> wrote: >> This is an experimental automated build and test service. >> Please feel free to ignore this email while we work the kinks out. >> >> For more info about this message, see http://wiki.winehq.org/BuildBot >> >> The Buildbot has detected a failed build on builder runtests-default while building Wine. >> Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/273 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) >> BUILD FAILED: failed shell_3 >> >> Errors: >> alarum: failed command was ../../../wine d3d9_test.exe.so visual.c >> fixme:win:EnumDisplayDevicesW ((null),0,0x33f0cc,0x00000000), stub! >> fixme:d3d_draw:drawPrimitive Using software emulation because manual fog coordinates are provided >> fixme:d3d:state_shademode WINED3DSHADE_PHONG isn't supported >> fixme:d3d:state_shademode WINED3DSHADE_PHONG isn't supported >> visual.c:9147: Tests skipped: D3DFMT_G16R16 textures not supported as render targets. >> visual.c:9239: Test failed: Offscreen failed for D3DFMT_R16F: Got color 0x20ffff, expected 0x18ffff. >> visual.c:9239: Test failed: Offscreen failed for D3DFMT_G16R16F: Got color 0x2010ff, expected 0x1818ff. >> visual.c:9239: Test failed: Offscreen failed for D3DFMT_R32F: Got color 0x20ffff, expected 0x18ffff. >> visual.c:9239: Test failed: Offscreen failed for D3DFMT_G32R32F: Got color 0x2010ff, expected 0x1818ff. >> visual.c:9239: Test failed: Offscreen failed for D3DFMT_A32B32G32R32F: Got color 0x201000, expected 0x181800. >> visual.c:7741: Tests skipped: Card has unconditional pow2 support, skipping conditional NP2 tests >> make: *** [visual.ok] Error 5 >> * Call to xpconnect wrapped JSObject produced this error: * >> * Call to xpconnect wrapped JSObject produced this error: * >> GnuTLS error: GnuTLS internal error. >> >> > Interesting. I didn't test on your kind of GPUs but I understood that it should support the feature I'm enabling. I'm not sure right now if it really doesn't support it or it is a driver bug of some kind. Thank you for catching it, if you have Windows installed on that machine I'll probably ask you to run a small testcase soon...

Oct. 12, 2011

Re: wined3d: Report every texture format as filterable and blendable.

by Dan Kegel

Fails d3d9's visual tests here. os: Linux 3.0.0-12-generic, Ubuntu 11.10 gpu: GeForce GT 240/PCI/SSE2 3.3.0 NVIDIA 280.13 On Wed, Oct 12, 2011 at 2:52 PM, <buildbot(a)kegel.com> wrote: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default/builds/273 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed shell_3 > > Errors: > alarum: failed command was ../../../wine d3d9_test.exe.so visual.c > fixme:win:EnumDisplayDevicesW ((null),0,0x33f0cc,0x00000000), stub! > fixme:d3d_draw:drawPrimitive Using software emulation because manual fog coordinates are provided > fixme:d3d:state_shademode WINED3DSHADE_PHONG isn't supported > fixme:d3d:state_shademode WINED3DSHADE_PHONG isn't supported > visual.c:9147: Tests skipped: D3DFMT_G16R16 textures not supported as render targets. > visual.c:9239: Test failed: Offscreen failed for D3DFMT_R16F: Got color 0x20ffff, expected 0x18ffff. > visual.c:9239: Test failed: Offscreen failed for D3DFMT_G16R16F: Got color 0x2010ff, expected 0x1818ff. > visual.c:9239: Test failed: Offscreen failed for D3DFMT_R32F: Got color 0x20ffff, expected 0x18ffff. > visual.c:9239: Test failed: Offscreen failed for D3DFMT_G32R32F: Got color 0x2010ff, expected 0x1818ff. > visual.c:9239: Test failed: Offscreen failed for D3DFMT_A32B32G32R32F: Got color 0x201000, expected 0x181800. > visual.c:7741: Tests skipped: Card has unconditional pow2 support, skipping conditional NP2 tests > make: *** [visual.ok] Error 5 > * Call to xpconnect wrapped JSObject produced this error: * > * Call to xpconnect wrapped JSObject produced this error: * > GnuTLS error: GnuTLS internal error. > >

Oct. 12, 2011

re: ntdll: Define PATH_MAX when needed

by Dan Kegel

When is PATH_MAX not defined? It's part of posix (see http://pubs.opengroup.org/onlinepubs/009695399/basedefs/limits.h.html ) Are you trying to build on Gnu Hurd?

Oct. 12, 2011

Please Test again, user32: use uniscribe in the single line edit control

by Aric Stewart

Apologies I sent this to the wrong list just now. Ok, Thanks to Dan I have found and corrected the issues that this created with the tests. I believe all the tests should continue to pass with this version of the patch. Has anyone tried it live in any applications? -aric --- dlls/user32/edit.c | 252 +++++++++++++++++++++++++++++----------------------- 1 files changed, 142 insertions(+), 110 deletions(-)

Oct. 12, 2011

Re: shell32/tests: Better check the result of SHGetDesktopFolder

by Austin English

2011/10/12 André Hentschel <nerv(a)dawncrow.de>: > Might fix the crash in some NT4 systems > --- > dlls/shell32/tests/brsfolder.c | 3 ++- > 1 files changed, 2 insertions(+), 1 deletions(-) > > diff --git a/dlls/shell32/tests/brsfolder.c b/dlls/shell32/tests/brsfolder.c > index 7b87967..6e5e73b 100644 > --- a/dlls/shell32/tests/brsfolder.c > +++ b/dlls/shell32/tests/brsfolder.c > @@ -331,7 +331,8 @@ static void test_selection(void) > > hr = SHGetDesktopFolder(&desktop_object); > ok (SUCCEEDED(hr), "SHGetDesktopFolder failed with hr 0x%08x\n", hr); > - if (FAILED(hr)) { > + ok (desktop_object, "Expected desktop_object to be a valid interface\n"); > + if (FAILED(hr) || !desktop_object) { > skip("SHGetDesktopFolder failed - skipping\n"); > return; > } > -- > > Best Regards, André Hentschel Causes a compiler warning on Buildbot: Errors: brsfolder.c: In function 'test_selection': brsfolder.c:334:5: error: passing argument 1 of 'winetest_ok' makes integer from pointer without a cast [-Werror] make[1]: *** [brsfolder.o] Error 1 -- -Austin

Oct. 12, 2011

Re: Warning report for wine-1.3.30-55-g583e887

by Josh Juran

On Oct 12, 2011, at 1:31 PM, Octavian Voicu wrote: > On Wed, Oct 12, 2011 at 11:22 PM, Josh Juran <josh(a)iswifter.net> wrote: >> My understanding is that accessing element n or greater in an array[n] >> is undefined behavior, but declaring a huge array and allocating only >> part of it is valid. > > It's a commonly used pattern in C. Agreed. > You declare a size-one array at the > end of a structure, then you allocate enough memory to hold the > structure plus (n-1) extra elements of the array. In wine size is > usually calculated with the macro FIELD_OFFSET(struct, member[size]). I understand. I'm just passing on what I read on another mailing list, which is that the construction is non-portable (even if it happens to work everywhere), and that might be the justification for the warning. A purist approach would call for avoiding this idiom, but pragmatically, disabling the warning is sufficient. Josh

Oct. 12, 2011

Re: Warning report for wine-1.3.30-55-g583e887

by Octavian Voicu

On Wed, Oct 12, 2011 at 11:22 PM, Josh Juran <josh(a)iswifter.net> wrote: > My understanding is that accessing element n or greater in an array[n] > is undefined behavior, but declaring a huge array and allocating only > part of it is valid. It's a commonly used pattern in C. You declare a size-one array at the end of a structure, then you allocate enough memory to hold the structure plus (n-1) extra elements of the array. In wine size is usually calculated with the macro FIELD_OFFSET(struct, member[size]). Octavian

Oct. 12, 2011

Re: Warning report for wine-1.3.30-55-g583e887

by Josh Juran

On Oct 12, 2011, at 10:11 AM, Dmitry Timoshkov wrote: > Jerome Leclanche <adys.wh(a)gmail.com> wrote: > >> It would be nice to fix the "array subscript is above array bounds" >> warning spam for 1.4.0. They make up 90% of the warnings. > > The compiler is misguided, you can safely ignore those warnings. My understanding is that accessing element n or greater in an array[n] is undefined behavior, but declaring a huge array and allocating only part of it is valid. Josh

Oct. 12, 2011

Re: Warning report for wine-1.3.30-55-g583e887

by Michael Stefaniuc

On 10/12/2011 02:29 PM, Jerome Leclanche wrote: > It would be nice to fix the "array subscript is above array bounds" > warning spam for 1.4.0. They make up 90% of the warnings. The fix would be to add a configure check for that bogus warning on access to the array[n]; n > 1 struct _tag { ... array[1]; } and to add -Wno-array-bounds in that case. bye michael

Oct. 12, 2011

Re: [Corrected^2] dlls/ntdll/serial.c: Generate a single EV_TXEMPTY when the\ TX buffer turns empty

by Alexandre Julliard

Uwe Bonnes <bon(a)elektron.ikp.physik.tu-darmstadt.de> writes: > @@ -807,8 +808,22 @@ typedef struct async_commio > static NTSTATUS get_irq_info(int fd, serial_irq_info *irq_info) > { > NTSTATUS status = STATUS_NOT_IMPLEMENTED; > -#ifdef TIOCGICOUNT > struct serial_icounter_struct einfo; You can't move this out of the #idef. -- Alexandre Julliard julliard(a)winehq.org

Oct. 12, 2011

Re: WineHQ database compromise

by GOUJON Alexandre

On 10/11/2011 09:13 PM, Jeremy White wrote: > I am sad to say that there was a compromise of the WineHQ database system. "Nothing Is Invulnerable" So, now or later, your system will be compromised. The only thing you have to do is to be prepared to face an incident and of course secure your systems to slow the attacker(s) down. The bugzilla case does not really worry me because it's only bugs. But as CEO, you have to protect your company and your customers. I'm of course a simple "user" of wine and I have absolutely not the right to tell you what to do. But something was open, broken or whatever .. and now you have to spend time and energy to try to repair the breach. Just don't let it happen again. There are lots of methods to analyse risk. Depending on what level of security you want, it will cost more or less. Just think about it. Anyway, thanks for the quick reply, communication is really important in this situation.

Oct. 12, 2011

Re: Warning report for wine-1.3.30-55-g583e887

by Dmitry Timoshkov

Jerome Leclanche <adys.wh(a)gmail.com> wrote: > It would be nice to fix the "array subscript is above array bounds" > warning spam for 1.4.0. They make up 90% of the warnings. The compiler is misguided, you can safely ignore those warnings. -- Dmitry.

Oct. 12, 2011

Re: d3d8/tests: skip the visual test if d3d cannot be initialized

by Austin English

On Wed, Oct 12, 2011 at 04:39, Henri Verbeet <hverbeet(a)gmail.com> wrote: > On 12 October 2011 01:51, Austin English <austinenglish(a)gmail.com> wrote: >> Current d3d8 does a win_skip(), while d3d9 does a skip(). This unifies >> the behavior to match the d3d9 behavior. >> > I seem to recall this being intentional, so that you get a failure if > your setup has e.g. no OpenGL. If the goal is to ensure some test fails with that setup, opengl32/tests/opengl.c already ensures that: ../../../tools/runtest -q -P wine -M opengl32.dll -T ../../.. -p opengl32_test.exe.so opengl.c && touch opengl.ok err:wgl:X11DRV_WineGL_InitOpenglInfo couldn't initialize OpenGL, expect problems opengl.c:1295: Test failed: Unable to find pixel format. *** Error code 1 I was looking to make d3d8/d3d9 consistent. Instead we could make d3d9 fail if OpenGL isn't available, as d3d8 does now. -- -Austin

Oct. 12, 2011

Re: [4/4] wined3d: Don't allow fbo blits to frontbuffers that need fixups.

by Stefan Dösinger

On Wednesday 12 October 2011 11:35:48 Henri Verbeet wrote: > On 11 October 2011 22:31, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > static BOOL fbo_blit_supported(const struct wined3d_gl_info *gl_info, > > enum wined3d_blit_op blit_op, > > > > const RECT *src_rect, DWORD src_usage, WINED3DPOOL src_pool, > > const struct wined3d_format *src_format, > > > > - const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, > > const struct wined3d_format *dst_format) + const RECT *dst_rect, > > DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format > > *dst_format, + const struct wined3d_surface *dst_surface) > > I don't think we want this. Does this actually ever happen? I had to double-check this. It used to be a problem before you introduced the shadow ddraw frontbuffer. Now it works by luck because the shadow frontbuffer has a palette set and is converted on upload time, so it is a RGB surface that doesn't need read-time conversion. There are two things that make the fbo blit work correctly although it should be broken: *) The offscreen P8 surfaces should be converted with the destination's palette. The palette happens to be the same for the shadow frontbuffer, but for no other surface. *) The shadow frontbuffer has SFLAG_CONVERTED set. Thus blits to it happen in software, otherwise we'd write P8 data to it that had no palette when it was converted to RGB(the ERR in d3dfmt_p8_init_palette). So in the end we never happen to have a P8 surface that only has GL_ALPHA8 data right now. No surface uses COMPLEX_FIXUP_P8 right now. It will be a problem again once we load offscreen surfaces(render target or just offscreenplain) with GL_ALPHA8 textures that require COMPLEX_FIXUP_P8. I'll send the patch again with the patchset that re-enables shader P8 conversion. And yeah, we don't actually want to pass the destination surface to fbo_blit_supported. It can go away again once we have the shadow frontbuffers in wined3d instead of ddraw. The ARB and FFP blitter have a similar problem when blitting to onscreen P8 surfaces, but unlike the FBO blitter they can do the palette lookup. They can accept the blit no matter what the destination is, but they have to make sure to select the right color fixup and palette.

Oct. 12, 2011

Re: 79794: Help Test Please: use uniscribe in the single line edit control

by Aric Stewart

Thank Dan, I had not gotten to a full test run. I will look into those. -aric On 10/11/11 8:01 PM, Dan Kegel wrote: > Hi Aric, > failed again on second run, > http://buildbot.kegel.com:8010/builders/runtests-default/builds/209 > > Failing tests are > comctl32_test.exe.so listview.c > comctl32_test.exe.so updown.c > user32_test.exe.so edit.c > > On Tue, Oct 11, 2011 at 1:26 PM, Dan Kegel<dank(a)kegel.com> wrote: >> Hey Aric, >> don't know if I trust the bot at the moment, but it says the tests >> failed here... >> I'll rerun them to see if the failure is repeatable. >> >> ---------- Forwarded message ---------- >> From:<buildbot(a)kegel.com> >> Date: Tue, Oct 11, 2011 at 1:21 PM >> Subject: Re: 79794: Help Test Please: use uniscribe in the single line >> edit control >> To: dank(a)kegel.com, wine-tests-results(a)winehq.org >> >> >> This is an experimental automated build and test service. >> Please feel free to ignore this email while we work the kinks out. >> >> For more info about this message, see http://wiki.winehq.org/BuildBot >> >> The Buildbot has detected a failed build on builder runtests-default >> while building Wine. >> Full details are available at: >> http://buildbot.kegel.com/builders/runtests-default/builds/186 (though >> maybe not for long, as I'm still reinstalling the buildbot >> periodically while experimenting) >> BUILD FAILED: failed shell_3 >> >> Errors: >> alarum: Timeout! Killing child ../../../wine. >> alarum: failed command was ../../../wine comctl32_test.exe.so listview.c >> alarum: ../../../wine terminated abnormally >> make: *** [listview.ok] Error 99 >> alarum: Timeout! Killing child ../../../wine. >> alarum: failed command was ../../../wine comctl32_test.exe.so updown.c >> alarum: ../../../wine terminated abnormally >> make: *** [updown.ok] Error 99 >> * Call to xpconnect wrapped JSObject produced this error: * >> * Call to xpconnect wrapped JSObject produced this error: * >> alarum: failed command was ../../../wine user32_test.exe.so edit.c >> edit.c:1138: Test failed: expected 1 got 0 >> edit.c:1156: Test failed: expected 1 got 0 >> edit.c:1174: Test failed: expected 1 got 0 >> make: *** [edit.ok] Error 3 >> GnuTLS error: GnuTLS internal error. >>

Oct. 12, 2011

Re: [PATCH 5/5] server/named_pipe.c: remove the ps_disconnected_server state, if DisconnectNamedPipe is called, the server i

by Marvin

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=14844 Your paranoid android. === WNT4WSSP6 (32 bit pipe) === pipe.c:1603: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 203 === W2KPROSP4 (32 bit pipe) === pipe.c:1603: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 997

Oct. 12, 2011

Re: [PATCH 4/5] kernel32, ntdll: implement most of GetNamedPipeHandleState (try 2)

by Marvin

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=14843 Your paranoid android. === WNT4WSSP6 (32 bit pipe) === pipe.c:1559: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 203 === W2KPROSP4 (32 bit pipe) === pipe.c:1559: Test failed: expected ERROR_CANNOT_IMPERSONATE, got 997

Oct. 12, 2011

Re: [2/4] wined3d: reject ARB shader blits if the destination is not fbo attachable

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 12.10.2011 um 11:26 schrieb Henri Verbeet: > On 11 October 2011 22:30, Stefan Dösinger <stefan(a)codeweavers.com> wrote: >> + if (wined3d_settings.offscreen_rendering_mode == ORM_FBO) >> + { >> + if (!((dst_format->flags & WINED3DFMT_FLAG_FBO_ATTACHABLE) || (dst_usage & WINED3DUSAGE_RENDERTARGET))) >> + return FALSE; >> + } >> + else if (!(dst_usage & WINED3DUSAGE_RENDERTARGET)) >> + { >> + return FALSE; >> + } > Formats are never FBO-attachable if we don't have > FBOs. It's not clear this is what you want either way though. This > pretty much looks like a copy/paste from fbo_blit_supported(), and the > WINED3DUSAGE_RENDERTARGET check there is to allow formats that don't > have a FBO-attachable internal format, but were created with the > rtInternal format (which should always be attachable). There's really > no equivalent check for backbuffer ORM, it just renders something and > hopes we can read it back. I'm not sure I see what you mean. What I want is to make sure the blitter doesn't draw to a surface that GL can't draw to. In the FBO ORM case the fbo_attachable || rendertarget case seems just right, for the reasons you mention. With backbuffer ORM we don't really know where we can draw(e.g. stuff like surface size vs framebuffer size comes into play), but I think RENDERTARGET usage is a decent heuristic since we'll probably have to be able to draw to it anyways, or things fails. I could add some surface_is_drawable() function or equivalent surface flag. That might be a good idea on its own, but I don't see a better way to define if a surface can be rendered to than the above statement. -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) iQIcBAEBAgAGBQJOlYhaAAoJEN0/YqbEcdMwpxgQAJTF3i75QLb+gosDaGkFODbk nhNUt/PHRw7k8yTVFBHp5aBWNCROdcr9yFANMj4AD74ii1HGNj9jzGISyfogvfqv RNHHHvwdff7b8Ev8MtKBEY3w9fTPvl7w0FzdzoSTFBF2SxkeEMc6QnecaHIfcepr SCf69/FFeVkPniOiVa3RCbLXyvrWG6ifMpJpXV8DFw23NTqb/JGyQIjodW2hc+Qy Gpgz4hwBw/jywOeFPCyOXtpR0bTBdD03mix3XWeYYaZg1Ukm+jyARQOvBSFU0dgB xhEcVTR0aNGA5WAgpKxUT1f3jRQc9f36U7HS5jTZemFR71wAXFPO0pMm7RNNrMXQ imhzE9+56OS7o96Vk4Rz8UdFYDBgDvTjO8nlp9j5aDOi8ylsX9TstDdRi2kh3ILI Satb83g0brU6JZaDqsLd5xZI8ZTPL55MniKRGvXJ1EEepViC0isgvMmrM1b/I2Wo N0FSoLkukXAfJHcal9YD6nAQeoI1HbM9HNNmAMPMCtDb9Y8Q0VSSWoeh71iP8cGF DziZZ3uSnX90kDrt7q6maXDFSWUS8YKZz2lviRxupb2LzwaSMAP2Aw1XEQz/YImH HxZkPuY4kZdQNShcYOyv9ZqNKUBrD5geFhmErYqxSL+Ede1nM7Lcp52NhbVRnTLI 9ksujocgWYGZYbzbhwym =04xu -----END PGP SIGNATURE-----

Oct. 12, 2011

Warning report for wine-1.3.30-55-g583e887

by Jerome Leclanche

It would be nice to fix the "array subscript is above array bounds" warning spam for 1.4.0. They make up 90% of the warnings. Apart from those, theres 25 more warning blocks to fix, most of them unused-but-set-variables. JL

Oct. 12, 2011

Re: [1/4] wined3d: Use an rbtree for the blit programs

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 12.10.2011 um 11:57 schrieb Henri Verbeet: > You already have the P8 shader. My bad. What will actually be added is the NONE conversion with color keying. But adding a color keying flag is what makes the current approach unsustainable because it's orthogonal to the other properties. > In the case of color keying it should > probably at least be in the same series as the code that uses it, I can resend the patch later, but IMO getting rid of the current spaghetti shader selection code makes sense on its own. > but > I'm not sure an rbtree would really have a lot of advantages over a > simple array there. I chose the rbtree for consistency with the ffp and pshader/vshader selection, but an array is probably more efficient because I don't expect to have more than 4 different blit shaders at a time. -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) iQIcBAEBAgAGBQJOlXiYAAoJEN0/YqbEcdMwXLMP/0OUs1gZK6lJDmTOo7/HJRqt xh0+4pTZqWfQsPOePPJ9Erz5wtPSToKayShimwpbdvsmfdUgsGuWGqR22MipgADJ Q7t8muOEe8K9tFb5QgLwNhK39VQMASh+WohuUNFazMmBXbRIETT2UaGkh/VipihK gkI/7JnA8ju/Y41golqiEdnYShnmSTcnTXgaQp4NZDYZDgkT9IFHOmdd+uE4NvE2 ngYciZywyA8Rt37On4DjEMw9Jv32NKpnfoocXx7sPWFFpcqeDejCqIj32yVbWZBo Wj8gi4IbfAE+xdMWFBS2PIlCR1rbv6wwfaC9jB9rAYVpKin4hvk7DDPr4fqCTJ5q eTi5w1ehoH4MOeH8YPnu9BtxrKkrUwXYGKsivKB2jDBTzjm0G+IiZxKT00rvBrmL WpId1Y3Zau7TkTKeoy/W6LThjZzLeyw48Mmf5XTUabp8U/TBIwbPHzn5bHuYZJle AFt/74iWwbIfMVgeRGgSbXXCk56aV1+ykK9HT7DrH4Ne/74tZjZM4+kf54s/H9T+ hjj9lmR3mdm6wCnJqPdX+9+DyiYLSCgkdVa3Cd9TDaElsoMaggJIraxU8PFidU7c 3tzyg5AE+7QjqjKgv1ZZQhyYUWK85Z29iij5BWpzlqvaqfsKQD0fC6inaRhbkLG6 t5j6zcErEey2yeNttNas =3kLs -----END PGP SIGNATURE-----

Oct. 12, 2011

Re: [1/4] wined3d: Use an rbtree for the blit programs

by Henri Verbeet

On 12 October 2011 11:47, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > Because when we add more conversions(P8) and settings(color keying) the current approach will result ugly unmaintainable if blocks. > You already have the P8 shader. In the case of color keying it should probably at least be in the same series as the code that uses it, but I'm not sure an rbtree would really have a lot of advantages over a simple array there.

Oct. 12, 2011

Re: [1/4] wined3d: Use an rbtree for the blit programs

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 12.10.2011 um 11:26 schrieb Henri Verbeet: > Why do you need this? Because when we add more conversions(P8) and settings(color keying) the current approach will result ugly unmaintainable if blocks. -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) iQIcBAEBAgAGBQJOlWIoAAoJEN0/YqbEcdMwdsoP/1FWotm97eiLNanErb/fXL6C t9dlTzVVXbcUboFSw9uBhSY+GQNj+sjZ3sQ5wLJSsQbsypW4MMOYjoBIUD4n+N91 6QY2xK5Bys7DnE7l8T9k3jCss5Fi93TpOkrQjDi0I7C0OreL7/yueM3NJQwn4PiS PJE9C5upvARhPII158NqYtVrd5TmImpLes6yshI9k4gQc4XfDZt9n0x0ERWmL8ag zS8nVR2Zaa3HaRuk/BMnf5X8FWiImwT/JapK9AyroWV4Lcg4MlKaOLGl43KaRTjQ 8IbMDPxFdWOFyU4yOfu8MEwLPDj1g7wG/CxCDNrKXMgRxgTmam9mtogM+e/dGWFX H8x+Td7h56syIU9l7g+L/4ODVugIKBxBlDdeJVUYbjOOpxuaUjsiQXG5aUcfYLf2 +1hVoKXuk7XfobCF3XKRW0hQIZGW9iJ2NAfLQ6OWK/tCZIsQ+GQafjJRPSAMgC7E y/WQ6YuuTA9eGo9si8wiuzJ5xCJoqjG7Y4oB7aih/l0DjG8MwKwWuxk85kcMg6xF 9Hnkz7738HyXhb6Y2cyUtPVsJTao5DZkBFXKs+SiCQhOnLFKPrXyIB8Zk2R4uLl3 9v+OU9khdzjTvu+f8uRw+V3HW4X+inDdS2UueGNlRHh/rYFMHX8GIYKeVK+8PS/C 1ZEjQ+1hE7S664QuNv3g =Ptfk -----END PGP SIGNATURE-----

Oct. 12, 2011

Re: d3d8/tests: skip the visual test if d3d cannot be initialized

by Henri Verbeet

On 12 October 2011 01:51, Austin English <austinenglish(a)gmail.com> wrote: > Current d3d8 does a win_skip(), while d3d9 does a skip(). This unifies > the behavior to match the d3d9 behavior. > I seem to recall this being intentional, so that you get a failure if your setup has e.g. no OpenGL.

Oct. 12, 2011

Re: [4/4] wined3d: Don't allow fbo blits to frontbuffers that need fixups.

by Henri Verbeet

On 11 October 2011 22:31, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > static BOOL fbo_blit_supported(const struct wined3d_gl_info *gl_info, enum wined3d_blit_op blit_op, > const RECT *src_rect, DWORD src_usage, WINED3DPOOL src_pool, const struct wined3d_format *src_format, > - const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format *dst_format) > + const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format *dst_format, > + const struct wined3d_surface *dst_surface) I don't think we want this. Does this actually ever happen?

Oct. 12, 2011

Re: [3/4] wined3d: reject ffp blits if the destination is not fbo attachable

by Henri Verbeet

On 11 October 2011 22:30, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + if (wined3d_settings.offscreen_rendering_mode == ORM_FBO) > + { > + if (!((dst_format->flags & WINED3DFMT_FLAG_FBO_ATTACHABLE) || (dst_usage & WINED3DUSAGE_RENDERTARGET))) > + return FALSE; > + } > + else if (!(dst_usage & WINED3DUSAGE_RENDERTARGET)) > + { > + return FALSE; > + } Same consideration as for 2/4. Guess I should have caught this when bccfd7cc introduced it.

Oct. 12, 2011

Re: [1/4] wined3d: Use an rbtree for the blit programs

by Henri Verbeet

Why do you need this?

Oct. 12, 2011

Re: [2/4] wined3d: reject ARB shader blits if the destination is not fbo attachable

by Henri Verbeet

On 11 October 2011 22:30, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + if (wined3d_settings.offscreen_rendering_mode == ORM_FBO) > + { > + if (!((dst_format->flags & WINED3DFMT_FLAG_FBO_ATTACHABLE) || (dst_usage & WINED3DUSAGE_RENDERTARGET))) > + return FALSE; > + } > + else if (!(dst_usage & WINED3DUSAGE_RENDERTARGET)) > + { > + return FALSE; > + } This is redundant. Formats are never FBO-attachable if we don't have FBOs. It's not clear this is what you want either way though. This pretty much looks like a copy/paste from fbo_blit_supported(), and the WINED3DUSAGE_RENDERTARGET check there is to allow formats that don't have a FBO-attachable internal format, but were created with the rtInternal format (which should always be attachable). There's really no equivalent check for backbuffer ORM, it just renders something and hopes we can read it back.

Oct. 12, 2011

Re: 79794: Help Test Please: use uniscribe in the single line edit control

by Dan Kegel

Hi Aric, failed again on second run, http://buildbot.kegel.com:8010/builders/runtests-default/builds/209 Failing tests are comctl32_test.exe.so listview.c comctl32_test.exe.so updown.c user32_test.exe.so edit.c On Tue, Oct 11, 2011 at 1:26 PM, Dan Kegel <dank(a)kegel.com> wrote: > Hey Aric, > don't know if I trust the bot at the moment, but it says the tests > failed here... > I'll rerun them to see if the failure is repeatable. > > ---------- Forwarded message ---------- > From: <buildbot(a)kegel.com> > Date: Tue, Oct 11, 2011 at 1:21 PM > Subject: Re: 79794: Help Test Please: use uniscribe in the single line > edit control > To: dank(a)kegel.com, wine-tests-results(a)winehq.org > > > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > For more info about this message, see http://wiki.winehq.org/BuildBot > > The Buildbot has detected a failed build on builder runtests-default > while building Wine. > Full details are available at: > http://buildbot.kegel.com/builders/runtests-default/builds/186 (though > maybe not for long, as I'm still reinstalling the buildbot > periodically while experimenting) > BUILD FAILED: failed shell_3 > > Errors: > alarum: Timeout! Killing child ../../../wine. > alarum: failed command was ../../../wine comctl32_test.exe.so listview.c > alarum: ../../../wine terminated abnormally > make: *** [listview.ok] Error 99 > alarum: Timeout! Killing child ../../../wine. > alarum: failed command was ../../../wine comctl32_test.exe.so updown.c > alarum: ../../../wine terminated abnormally > make: *** [updown.ok] Error 99 > * Call to xpconnect wrapped JSObject produced this error: * > * Call to xpconnect wrapped JSObject produced this error: * > alarum: failed command was ../../../wine user32_test.exe.so edit.c > edit.c:1138: Test failed: expected 1 got 0 > edit.c:1156: Test failed: expected 1 got 0 > edit.c:1174: Test failed: expected 1 got 0 > make: *** [edit.ok] Error 3 > GnuTLS error: GnuTLS internal error. >

Oct. 12, 2011

Re: [Wine] WineHQ database compromise

by Josh Juran

On Oct 11, 2011, at 3:54 PM, Conan Kudo (ニール・ゴンパ) wrote: > 2011/10/11 Josh Juran <josh(a)iswifter.net> > >> To clarify, your browser sends your password to bugzilla in cleartext, since HTTPS isn't an option. > > Shouldn't it be possible to modify the login environment so that a salted hash of the password is produced before sending it to the server, to strengthen the security a little bit? That protects the password itself, but not the privilege it guards. It also essentially makes Javascript a requirement, which currently it isn't. Josh

Oct. 11, 2011

Re: [Wine] WineHQ database compromise

by Conan Kudo (ニール・ゴンパ)

2011/10/11 Josh Juran <josh(a)iswifter.net> > On Oct 11, 2011, at 3:37 PM, Conan Kudo (ニール・ゴンパ) wrote: > > > On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: > > > >> Since bugzilla passwords were sent in cleartext anyway, I sincerely hope > none of them were otherwise valuable. (Remember FireSheep?) > > > > Wait, what? Bugzilla sends passwords in cleartext? That isn't very > smart... Is there no way to replace this with some sort of client based > hashing or something? > > To clarify, your browser sends your password to bugzilla in cleartext, > since HTTPS isn't an option. > > Firesheep was a lesson that even once passwords are secure, session > credentials are still vulnerable to sniffing. Some sites went to HTTPS-only > sessions after that. > > Josh > > > Shouldn't it be possible to modify the login environment so that a salted hash of the password is produced before sending it to the server, to strengthen the security a little bit?

Oct. 11, 2011

Re: [Wine] WineHQ database compromise

by Austin English

2011/10/11 Josh Juran <josh(a)iswifter.net>: > On Oct 11, 2011, at 3:37 PM, Conan Kudo (ニール・ゴンパ) wrote: > >> On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: >> >>> Since bugzilla passwords were sent in cleartext anyway, I sincerely hope none of them were otherwise valuable. (Remember FireSheep?) >> >> Wait, what? Bugzilla sends passwords in cleartext? That isn't very smart... Is there no way to replace this with some sort of client based hashing or something? > > To clarify, your browser sends your password to bugzilla in cleartext, since HTTPS isn't an option. > > Firesheep was a lesson that even once passwords are secure, session credentials are still vulnerable to sniffing. Some sites went to HTTPS-only sessions after that. http://bugs.winehq.org/show_bug.cgi?id=23791 -- -Austin

Oct. 11, 2011

Re: [Wine] WineHQ database compromise

by Josh Juran

On Oct 11, 2011, at 3:37 PM, Conan Kudo (ニール・ゴンパ) wrote: > On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: > >> Since bugzilla passwords were sent in cleartext anyway, I sincerely hope none of them were otherwise valuable. (Remember FireSheep?) > > Wait, what? Bugzilla sends passwords in cleartext? That isn't very smart... Is there no way to replace this with some sort of client based hashing or something? To clarify, your browser sends your password to bugzilla in cleartext, since HTTPS isn't an option. Firesheep was a lesson that even once passwords are secure, session credentials are still vulnerable to sniffing. Some sites went to HTTPS-only sessions after that. Josh

Oct. 11, 2011

Re: [Wine] WineHQ database compromise

by Conan Kudo (ニール・ゴンパ)

On Tue, Oct 11, 2011 at 3:39 PM, Josh Juran <josh(a)iswifter.net> wrote: > On Oct 11, 2011, at 12:13 PM, Jeremy White wrote: > > > Unfortunately, the attackers were able to download the full login > > database for both the appdb and bugzilla. This means that they have all > > of those emails, as well as the passwords. The passwords are stored > > encrypted, but with enough effort and depending on the quality of the > > password, they can be cracked. > > > > This, I'm afraid, is a serious threat; it means that anyone who uses the > > same email / password on other systems is now vulnerable to a malicious > > attacker using that information to access their account. > > Since bugzilla passwords were sent in cleartext anyway, I sincerely hope > none of them were otherwise valuable. (Remember FireSheep?) > > Josh > > Wait, what? Bugzilla sends passwords in cleartext? That isn't very smart... Is there no way to replace this with some sort of client based hashing or something?

Oct. 11, 2011

Re: WineHQ database compromise

by Jeremy White

> Almost 2 years ago I have sent you an email privately about a security > hole with the database. To be exactly, the date of the email is Wed, > Jul 29, 2009, 12:00 AM (GMT +02:00). I guess that's probably the same > trick the bad guys have used... Hmm. I can't find any such email in my archives - can you resend it to me? Are you sure it was me, and not the other Jeremy? I'd be curious to see if it matches up to the forensics we have. Cheers, Jeremy

Oct. 11, 2011

Re: [PATCH] dsound: Don't claim to support hardware buffers

by Marvin

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=14829 Your paranoid android. === W2K3R2SESP2 (32 bit dsound) === dsound.c:329: Test failed: IDirectSound_CreateSoundBuffer() failed to create a secondary buffer 88780032

Oct. 11, 2011

Re: WineHQ database compromise

by Matijn Woudt

On Tue, Oct 11, 2011 at 9:13 PM, Jeremy White <jwhite(a)codeweavers.com> wrote: > Hi, > > I am sad to say that there was a compromise of the WineHQ database system. > > What we know at this point that someone was able to obtain unauthorized > access to the phpmyadmin utility. We do not exactly how they obtained > access; it was either by compromising an admins credentials, or by > exploiting an unpatched vulnerability in phpmyadmin. > Jeremy, Almost 2 years ago I have sent you an email privately about a security hole with the database. To be exactly, the date of the email is Wed, Jul 29, 2009, 12:00 AM (GMT +02:00). I guess that's probably the same trick the bad guys have used... Kind regards, Matijn Woudt

Oct. 11, 2011

Re: [PATCH] dsound: Don't claim to support hardware buffers

by Chris Robinson

Hi. On Tuesday, October 11, 2011 3:31:59 PM Andrew Eikum wrote: > The specific drvcaps values and error code were tested against Windows > 7. How does this affect the primary buffer? Is it not allowed to be a hardware buffer either?

Oct. 11, 2011