September 2011 - Wine-Devel

Re: speeding up builds with ext4 and ccache tuning

by Dan Kegel

Two more things: - ccache's cache size matters. If ccache -s shows the cache is full after two builds, try raising it. The default is 1 GB, and I think with gcc-2.95, I needed a slightly larger cache. - ccache might slow you down if you have less than 3GB of RAM.

Sept. 10, 2011

Re: Marking test cases as flaky?

by Dan Kegel

On Sat, Sep 10, 2011 at 3:01 AM, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > A problem we have is that we have lots and lots of tests in dlls/d3d*/visual.c > and device.c that are unrelated to each other. d3d9/visual has about 100 > separate tests that could be in different files and work just the same way. > > Now I don't suggest moving them into different files, that would make the > directory look pretty ugly. But maybe we could introduce a function that gives > all following ok() calls a different name and makes it easier for external > apps to separate tests. E.g. > > wine_begin_tests(depth_clamp); > ok(...); > ok(...); > ok(...); > wine_begin_tests(stretchrect) > ok(...); > ok(...); > ok(...); Maybe instead of #define ok_(file, line) (winetest_set_location(file, line), 0) ? (void)0 : winetest_ok #define ok ok_(__FILE__, __LINE__) we could have #define ok_(file, func, line) (winetest_set_location(file, func, line), 0) ? (void)0 : winetest_ok #define ok ok_(__FILE__, __FUNCTION__, __LINE__) That would get you most of what you propose without any changes to the tests, I think.

Sept. 10, 2011

Re: [1/3] msxml3: Add xmlparser interfaces

by Francois Gouget

On Wed, 7 Sep 2011, Alistair Leslie-Hughes wrote: > > Changelog: > msxml3: Add xmlparser interfaces This patch defines an XML_ERROR_CODE enumeration type in xmlparser.idl which is a bit strange because I cannot find this type in the Windows SDK. But in the Windows 7 SDK there is an XmlError enumeration type in xmllite.h that looks similar but subtly different. These two enumaration types share some error codes, in particular XML_E_INVALID_HEXIDECIMAL (Microsoft can't spell apparently) but with different values. In this case 0xc00ce51e in Wine vs. 0xc00ce01e in the Windows 7 SDK. So what's going on? -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Before you criticize someone, walk a mile in his shoes. That way, if he gets angry, he'll be a mile away - and barefoot.

Sept. 10, 2011

Re: gdiplus/tests: Don't fail if Microsoft Sans Serif and Times New Roman are not installed.

by Michael Mc Donnell

On Thu, Sep 8, 2011 at 8:11 PM, Francois Gouget <fgouget(a)free.fr> wrote: > On Thu, 8 Sep 2011, Alexandre Julliard wrote: > [...] >> I don't think the name of the family really matters here. By using >> Tahoma the Wine implementation is trying to make sure we always return >> something, and that's what we should check: that we get a family, and >> that it's valid for creating a font or something along those lines. > > We don't have a fallback for Monospace fonts so that if Courier New is > not installed we return FontFamilyNotFound. What you're saying suggests > that this is a bug. Does anyone have a font to recommend as the > fallback? I think that the fallback for Courier New should be Liberation Mono [1] as they both are monospace. [1] http://en.wikipedia.org/wiki/Liberation_fonts

Sept. 10, 2011

Re: The msctf:inputprocessor test is not reliable

by Alexandre Julliard

Francois Gouget <fgouget(a)free.fr> writes: > Or would it be better to change processPendingMessages() to look more > like user32/tests/msg.c's flush_events() function although it looks like > it's overkill for now? That's preferable, yes. -- Alexandre Julliard julliard(a)winehq.org

Sept. 10, 2011

The msctf:inputprocessor test is not reliable

by Francois Gouget

This test is not reliable on my Debian desktop. http://test.winehq.org/data/3b40c03a8f0e3dcbb236d59125b115b2501086f8/linux_… inputprocessor.c:2117: Test failed: OnSetFocus expected current focus 0x12b070 got 0x12b0d0 inputprocessor.c:538: Test failed: Unexpected ThreadMgrEventSink_OnSetFocus sink inputprocessor.c:549: Test failed: Sink reports wrong previous focus If I run it in a loop it fails about once every twelve times: i=0;while ../../../tools/runtest inputprocessor.c && [ $i -lt 100 ];do i=`expr $i + 1`; echo $i; done Introducing a single Sleep(50) as shown below makes it totally reliable here. ShowWindow(wnd3,SW_SHOWNORMAL); SetFocus(wnd3); + Sleep(50); processPendingMessages(); test_CurrentFocus = FOCUS_SAVE; test_PrevFocus = FOCUS_SAVE; test_OnSetFocus = SINK_SAVE; SetFocus(wnd1); processPendingMessages(); sink_check_saved(&test_OnSetFocus,dm2,dm1,"OnSetFocus"); /* ^^^ Line that fails! */ That seems pretty hacky though. I suspect this is all because processPendingMessages() will exit if there's no pending message right after we call SetFocus() without giving enough time for the asynchronous X event to arrive. But I'm surprised that a single Sleep() fixes the issue, especially where it's placed. Also I don't know anything about msctf so I'm unsure how ThreadMgrEventSink_OnSetFocus() figures in all this. So should I just submit a patch to add that Sleep()? Or would it be better to change processPendingMessages() to look more like user32/tests/msg.c's flush_events() function although it looks like it's overkill for now? -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Linux: It is now safe to turn on your computer.

Sept. 10, 2011

urlmon:sec_mgr failure on XP

by Francois Gouget

This test now fails on my Widnows XP VMs that have IE >= 7 with 'All zones reset to default level': http://test.winehq.org/data/3b40c03a8f0e3dcbb236d59125b115b2501086f8/xp_fg-… sec_mgr.c:1695: Test failed: Expected zone 3, but got 1 for 'file/testing/test.test' The patch that introduces the failures is 98d5dfb8 which was committed on the 5th. I did some experiments and I can get this test to pass if I do: * Go to the Internet Properties control panel * Security tab * Select 'Local intranet' * Click on the 'Sites' button * Deselect 'Automatically detect intranet network' * Deselect 'Include all local (intranet) sites not listed in other zones' * Click on 'OK' I also checked on Windows 7 and by default it has the first checkbox checked but the second one unchecked. Disabling 'Automatically detect intranet network' and enabling 'Include all local sites...' to match Windows XP causes the test to fail on Windows 7 too. Finally, on XP even if I deselect all the 'Include XXX' checkboxes but select the 'Automatically detect intranet network' one the test fails while with the same configuration it passes on Windows 7. So I think this reflects a difference in the way the intranet detection works. So I'm hoping that you can fix the test. The easiest fix would be to remove the 'file/testing/test.test' check. An alternative would be to check what the settings are and adjust the expected result or skip the test. But it seems like this would be much more complex. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ 1 + e ^ ( i * pi ) = 0

Sept. 10, 2011

Re: Marking test cases as flaky?

by Stefan Dösinger

A problem we have is that we have lots and lots of tests in dlls/d3d*/visual.c and device.c that are unrelated to each other. d3d9/visual has about 100 separate tests that could be in different files and work just the same way. Now I don't suggest moving them into different files, that would make the directory look pretty ugly. But maybe we could introduce a function that gives all following ok() calls a different name and makes it easier for external apps to separate tests. E.g. wine_begin_tests(depth_clamp); ok(...); ok(...); ok(...); wine_begin_tests(stretchrect) ok(...); ok(...); ok(...); and so on.

Sept. 10, 2011

Re: Auto-selecting local variable/parameters when 'p'rinting values using winedbg?

by Eric Pouech

Le 09/09/2011 13:02, Frédéric Delanoy a écrit : > 2011/9/9 Eric Pouech<eric.pouech(a)orange.fr>: >> WineDbg> set ! symbol_picker scoped >> will do what you want >> unfortunately, there's no way to store it as a default command (you can >> always use --command or --file on command line to help) > Thanks for that quick response! > > Is there some doc on the internal commands, besides the code? > > http://www.winehq.org/docs/winedev-guide/dbg-config may be a bit out of date > > Frédéric > not for that part... feel free to bring to the doc up to date set ! symbol_picker interactive being the other setting (which is the default : if several symbols found, ask the user for which one to pick) scoped meaning give precedence to local symbols over global symbols A+ -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

Sept. 10, 2011

speeding up builds with ext4 and ccache tuning

by Dan Kegel

Here's a little brain dump about speeding up builds. I ran into two problems today: - disk slowness On a machine where builds were slow, the iozone benchmark showed I/O was half or two thirds as fast as a better machine. The slow system turned out to be ext3, whereas the fast ones were ext4. (It's also using gcc-2.95.3 for wine, whereas the others are using the system compiler.) Adding a 7200rpm disk with an ext4 filesystem, and putting both wine-git and ~/.ccache on it, sped up ./configure so that it matched the faster system... but builds were only sped up by 20%. Nice, but not sufficient. - ccache too strict? ccache helps hugely on most of my systems without any tuning, but wasn't working well on my slow system. After looking at ccache's statistics and log file for a while, I came across the CCACHE_SLOPPINESS environment variable; setting that to maximum sloppiness got the cache hit rate sky-high, and build times dropped by a factor of 2, making the slower box suddenly competitive with the faster system. The sloppy setting is probably inappropriate for general use, but ok for regression testing. Those two tweaks get the buildbot cycle time for the headless subset of tests with gcc-2.95 on my e7300 down under nine minutes, yay.

Sept. 10, 2011

Re: [PATCH] cmd: Add help info for xcopy

by Octavian Voicu

2011/9/9 Frédéric Delanoy <frederic.delanoy(a)gmail.com> > Patch is OK, but for consistency, shouldn't the externals be put at > the start of wcmdmain.c, just after the inbuilt[][]10]? > This way, you could see directly which command is built-in, and which > is external. > + make it static const, since it's not used outside of builtins.c. Octavian

Sept. 9, 2011

Re: resource_check_usage Unhandled usage flags 0x8

by Michael Stefaniuc

On 09/09/2011 08:24 PM, Luis C. Busquets Pérez wrote: > I do not from which exact version of wine, whenever I run Civilization > IV bryond the sword, losts of fixmes appear with the message > > fixme:d3d:resource_check_usage Unhandled usage flags 0x8. > > Since those fixmes did not appear before I wonder if this is a bug. Not at all, those are very common. And harmless. > I do not understand how it is that the function resource_check_usage > uses a flag that before it wasn't, was it? Can anybody clarify the usage > of the flag 0x8 and if this is a bug? Not a bug. I figure something changed and you hit now a different code path that sets the D3DUSAGE_WRITEONLY flag. That's just a hint to the driver that the application doesn't intend to read that VBO or index buffer. The driver is free to take that hint or ignore it. bye michael

Sept. 9, 2011

resource_check_usage Unhandled usage flags 0x8

by Luis C. Busquets Pérez

I do not from which exact version of wine, whenever I run Civilization IV bryond the sword, losts of fixmes appear with the message fixme:d3d:resource_check_usage Unhandled usage flags 0x8. Since those fixmes did not appear before I wonder if this is a bug. I do not understand how it is that the function resource_check_usage uses a flag that before it wasn't, was it? Can anybody clarify the usage of the flag 0x8 and if this is a bug?

Sept. 9, 2011

Re: 78632: Subject: po: Update Japanese translation.

by Dan Kegel

Hi Akihiro, your patch does not apply here, perhaps it's slightly out of date? patching file po/ja.po Hunk #22 FAILED at 8231. Hunk #23 succeeded at 8271 (offset 1 line). Hunk #24 succeeded at 8347 with fuzz 2 (offset 1 line). Hunk #25 succeeded at 8540 (offset 1 line). Hunk #26 succeeded at 8549 (offset 1 line). Hunk #27 succeeded at 8566 (offset 1 line). Hunk #28 succeeded at 8581 (offset 1 line). Hunk #29 succeeded at 8647 (offset 1 line). Hunk #30 succeeded at 8791 (offset 1 line). Hunk #31 succeeded at 8888 (offset 1 line). Hunk #32 succeeded at 8912 (offset 1 line). Hunk #33 succeeded at 9004 (offset 1 line). Hunk #34 succeeded at 9045 (offset 1 line). Hunk #35 succeeded at 9872 (offset 1 line). Hunk #36 succeeded at 10329 (offset 1 line). Hunk #37 succeeded at 10490 (offset 1 line). Hunk #38 succeeded at 10733 (offset 1 line). Hunk #39 succeeded at 10758 (offset 1 line). Hunk #40 succeeded at 11246 (offset 1 line). Hunk #41 succeeded at 11260 (offset 1 line). Hunk #42 succeeded at 11381 (offset 1 line). Hunk #43 succeeded at 11415 (offset 1 line). 1 out of 43 hunks FAILED -- saving rejects to file po/ja.po.rej dank(a)pantry:~/wine-git$ git reset --hard origin HEAD is now at cb55d2a crypt32/tests: Remove dead assignment (clang). dank(a)pantry:~/wine-git$ git apply 78632 error: patch failed: po/ja.po:8198 error: po/ja.po: patch does not apply 2011/9/9 <buildbot(a)kegel.com>: > This is an experimental automated build and test service. > Please feel free to ignore this email while we work the kinks out. > > The Buildbot has detected a failed build on builder runtests-default-x86_64 while building Wine. > Full details are available at: http://buildbot.kegel.com/builders/runtests-default-x86_64/builds/0 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting) > BUILD FAILED: failed git > > > For more info about this message, see http://wiki.winehq.org/BuildBot > > >

Sept. 9, 2011

Re: [2/2] dpnet: return a TCP/IP provider in IDirectPlay8PeerImpl_EnumServiceProviders

by Alexandre Julliard

Louis Lenders <xerox_xerox2000(a)yahoo.co.uk> writes: > + info = HeapAlloc(GetProcessHeap(), 0, req_size); > + > + info->pwszName = (LPWSTR)(info + 1); > + > + if(!pguidServiceProvider) > + { > + lstrcpyW(info->pwszName, dp_providerW); > + info->guid = CLSID_DP8SP_TCPIP; > + } > + else > + { > + lstrcpyW(info->pwszName, dp_adapterW); > + info->guid = adapter_guid; > + } > + > + *pSPInfoBuffer = *info; > + *pcReturned = 1; > + > + HeapFree(GetProcessHeap(), 0, info); This doesn't make sense, you are returning freed data. -- Alexandre Julliard julliard(a)winehq.org

Sept. 9, 2011

Re: [2/2] dpnet: return a TCP/IP provider in IDirectPlay8PeerImpl_EnumServiceProviders

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=14103 Your paranoid android. === WNT4WSSP6 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BC === W2KPROSP4 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BC

Sept. 9, 2011

Re: [1/2] dpnet: return success in IDirectPlay8PeerImpl_Initialize

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=14102 Your paranoid android. === WNT4WSSP6 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BC === W2KPROSP4 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BC

Sept. 9, 2011

Re: [PATCH 14/14] vbscript: Added null literal support

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=14101 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: [PATCH 13/14] vbsctipy: Added empty literal support

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=14100 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: [PATCH 09/14] vbscript: Added interp_equal implementation

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=14098 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: [PATCH 06/14] vbscript: Added interp_not implementation

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=14097 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: [PATCH 01/14] vbscript: Added support for |f a, b, c| call statement

by Jacek Caban

On 09/09/11 14:54, Marvin wrote: > 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=14094 > > Your paranoid android. > > > === WNT4WSSP6 (32 bit) === > No test summary line found > That's because the patch contains only non-C test changes. There will be more such false reports with my patches. Jacek

Sept. 9, 2011

Re: [PATCH 03/14] vbscript: Added support for parentheses in expressions

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=14096 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: [PATCH 02/14] vbscript: Added parser/compiler support for |option explicit|

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=14095 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 === W7PROX64 (64 bit) === No test summary line found

Sept. 9, 2011

Re: winebuild: Create stubs for ARM

by Alexandre Julliard

André Hentschel <nerv(a)dawncrow.de> writes: > this improves the testresults, as some tests expect to get stubs through GetProcAddress but never use them. These tests should be fixed (or the functions implemented), it shouldn't be necessary to generate stubs for Winelib. -- Alexandre Julliard julliard(a)winehq.org

Sept. 9, 2011