November 2009 - Wine-devel - list.winehq.org

Re: advapi32/tests: Skip tests for ACL file info if ACL info not stored (try 2)
by Paul Vriens 18 Nov '09

18 Nov '09

On 11/18/2009 09:54 AM, Austin Lund wrote: > + TCHAR volumepathname[50]; Please don't use TCHAR's. > + BOOL testacl = FALSE; > + DWORD filesystemflags; > > if (!pGetSecurityInfo) > { > @@ -3143,6 +3146,13 @@ static void test_GetSecurityInfo(void) > return; > } > > + if (GetVolumePathName(myARGV[0], volumepathname, 50)&& > + GetVolumeInformation(volumepathname, NULL, 0, NULL, NULL,&filesystemflags, NULL, 0)&& > + (filesystemflags& FILE_PERSISTENT_ACLS)) > + { > + testacl = TRUE; > + } GetVolumePathName is not available before W2K so you should use GetProcAddress here and bail out or skip() if the function is not available. It's also probably better to use the explicit A-versions for both calls. -- Cheers, Paul.

1 0

Re: Test for Bug 4543: Testers needed
by Charles Davis 18 Nov '09

18 Nov '09

Vincent Povirk wrote: > You still have the loop comparing to 0xff, and the test won't pass on > windows with the if(0). Whoops, that must have been an old version of the patch. I fixed the first one, and the second one should be fixed, too. Now the BitBlt(...,WHITENESS) call executes on Windows, but not on Wine. (There should be a function in wine/test.h for doing this.) Patch attached. > > On Tue, Nov 17, 2009 at 6:09 PM, Charles Davis <cdavis(a)mymail.mines.edu> wrote: >> All right, this should be the last one. The tests pass on Wine, and >> should pass on Windows if you un-if(0) the call to BitBlt(..., >> WHITENESS). I still don't have any results for the many other Windows >> versions (95, 98, ME, NT4, 2000, Vista, Win7), so I'm going to wait for >> results from these other platforms before I send this to wine-patches. >> >> Chip >> >> >> From 2d5aa7c607496c1b05a7a08adfee09db446754cb Mon Sep 17 00:00:00 2001 >> From: Charles Davis <cdavis(a)mymail.mines.edu> >> Date: Tue, 17 Nov 2009 11:10:50 -0700 >> Subject: [PATCH] gdi32/tests: Test BitBlt() to a metafile. >> To: wine-patches <wine-patches(a)winehq.org> >> Reply-To: wine-devel <wine-devel(a)winehq.org> >> >> --- >> dlls/gdi32/tests/metafile.c | 83 +++++++++++++++++++++++++++++++++++++++++++ >> 1 files changed, 83 insertions(+), 0 deletions(-) >> >> diff --git a/dlls/gdi32/tests/metafile.c b/dlls/gdi32/tests/metafile.c >> index b559901..19401ed 100644 >> --- a/dlls/gdi32/tests/metafile.c >> +++ b/dlls/gdi32/tests/metafile.c >> @@ -665,6 +665,88 @@ static void test_SaveDC(void) >> DestroyWindow(hwnd); >> } >> >> +static int CALLBACK test_BitBlt_record(HDC hdc, HANDLETABLE *table, const ENHMETARECORD *emfr, int nobj, LPARAM data) >> +{ >> + UINT *pi = (UINT *)data; >> + EMRBITBLT *emrbb; >> + >> + if(emfr->iType != EMR_BITBLT) return 1; >> + emrbb = (EMRBITBLT *)emfr; >> + >> + (*pi)++; >> + if(*pi > 2) return 1; >> + ok( emrbb->xDest == 0, "wrong xDest for BitBlt record: got %d, expected 0\n", emrbb->xDest ); >> + ok( emrbb->yDest == 0, "wrong yDest for BitBlt record: got %d, expected 0\n", emrbb->yDest ); >> + ok( emrbb->xSrc == 0, "wrong xSrc for BitBlt record: got %d, expected 0\n", emrbb->xSrc ); >> + ok( emrbb->ySrc == 0, "wrong ySrc for BitBlt record: got %d, expected 0\n", emrbb->ySrc ); >> + ok( emrbb->cxDest == GetSystemMetrics(SM_CXSCREEN), "wrong cxDest for BitBlt record: got %d\n", emrbb->cxDest ); >> + ok( emrbb->cyDest == GetSystemMetrics(SM_CYSCREEN), "wrong cyDest for BitBlt record: got %d\n", emrbb->cyDest ); >> + if(*pi == 1) /* First BitBlt */ >> + { >> + ok( emrbb->dwRop == SRCCOPY, "wrong ROP for BitBlt record: got %x, expected SRCCOPY\n", emrbb->dwRop ); >> + } >> + else >> + { >> + todo_wine ok( emrbb->dwRop == WHITENESS, "wrong ROP for BitBlt record: got %x, expected WHITENESS\n", emrbb->dwRop ); >> + } >> + return 1; >> +} >> + >> +static void test_emf_BitBlt(void) >> +{ >> + HDC hdcDisplay, hdcMetafile, hdcBitmap; >> + HBITMAP hBitmap, hOldBitmap; >> + HENHMETAFILE hMetafile; >> + ENHMETAHEADER emh; >> + BYTE *bits; >> + BOOL ret; >> + UINT i; >> + >> + hdcDisplay = CreateDCA("DISPLAY", NULL, NULL, NULL); >> + ok( hdcDisplay != 0, "CreateDCA error %d\n", GetLastError() ); >> + hdcMetafile = CreateEnhMetaFileA(hdcDisplay, NULL, NULL, NULL); >> + ok( hdcMetafile != 0, "CreateEnhMetaFileA failed\n" ); >> + >> + hdcBitmap = CreateCompatibleDC(hdcDisplay); >> + ok( hdcBitmap != 0, "CreateCompatibleDC failed\n" ); >> + hBitmap = CreateCompatibleBitmap(hdcDisplay, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN)); >> + ok( hBitmap != 0, "CreateCompatibleBitmap failed\n" ); >> + hOldBitmap = SelectObject(hdcBitmap, hBitmap); >> + >> + ret = BitBlt(hdcMetafile, 0, 0, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN), hdcBitmap, 0, 0, SRCCOPY); >> + ok( ret, "BitBlt failed\n" ); >> + if(0) /* Crashes on wine */ >> + { >> + ret = BitBlt(hdcMetafile, 0, 0, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN), 0, 0, 0, WHITENESS); >> + ok( ret, "BitBlt failed\n" ); >> + } >> + >> + hMetafile = CloseEnhMetaFile(hdcMetafile); >> + ok( hMetafile != 0, "CloseEnhMetaFile failed\n" ); >> + >> + ret = GetEnhMetaFileHeader( hMetafile, sizeof(emh), &emh ); >> + ok( ret != 0, "GetEnhMetaFileHeader failed\n" ); >> + ok( emh.dSignature == ENHMETA_SIGNATURE, "bad metafile signature %x\n", emh.dSignature ); >> + >> + bits = HeapAlloc( GetProcessHeap(), 0, emh.nBytes ); >> + if(!bits) >> + { >> + skip( "not enough memory for EMF bits\n" ); >> + return; >> + } >> + ret = GetEnhMetaFileBits( hMetafile, emh.nBytes, bits ); >> + ok( ret != 0, "GetEnhMetaFileBits error %d\n", GetLastError() ); >> + for(i = 0; i < emh.nBytes; i++) >> + todo_wine ok( bits[i] == 0xff, "unexpected EMF bits: got %02x, expected 0xff\n", bits[i] ); >> + >> + EnumEnhMetaFile(0, hMetafile, test_BitBlt_record, &i, NULL); >> + todo_wine ok( i == 2, "too many/not enough BitBlt records: got %d, expected 2\n", i ); >> + >> + SelectObject(hdcBitmap, hOldBitmap); >> + DeleteObject(hBitmap); >> + DeleteDC(hdcBitmap); >> +} >> + >> static void test_mf_SaveDC(void) >> { >> HDC hdcMetafile; >> @@ -2628,6 +2710,7 @@ START_TEST(metafile) >> /* For enhanced metafiles (enhmfdrv) */ >> test_ExtTextOut(); >> test_SaveDC(); >> + test_emf_BitBlt(); >> >> /* For win-format metafiles (mfdrv) */ >> test_mf_SaveDC(); >> -- >> 1.6.5.2 >> >> >> > > > Chip >From 990018e6e74279c3243d2c3b48472e94d8839c8c Mon Sep 17 00:00:00 2001 From: Charles Davis <cdavis(a)mymail.mines.edu> Date: Tue, 17 Nov 2009 11:10:50 -0700 Subject: [PATCH] gdi32/tests: Test BitBlt() to a metafile. To: wine-patches <wine-patches(a)winehq.org> Reply-To: wine-devel <wine-devel(a)winehq.org> --- dlls/gdi32/tests/metafile.c | 67 +++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 67 insertions(+), 0 deletions(-) diff --git a/dlls/gdi32/tests/metafile.c b/dlls/gdi32/tests/metafile.c index b559901..dbeff43 100644 --- a/dlls/gdi32/tests/metafile.c +++ b/dlls/gdi32/tests/metafile.c @@ -665,6 +665,72 @@ static void test_SaveDC(void) DestroyWindow(hwnd); } +static int CALLBACK test_BitBlt_record(HDC hdc, HANDLETABLE *table, const ENHMETARECORD *emfr, int nobj, LPARAM data) +{ + UINT *pi = (UINT *)data; + EMRBITBLT *emrbb; + + if(emfr->iType != EMR_BITBLT) return 1; + emrbb = (EMRBITBLT *)emfr; + + (*pi)++; + if(*pi > 2) return 1; + ok( emrbb->xDest == 0, "wrong xDest for BitBlt record: got %d, expected 0\n", emrbb->xDest ); + ok( emrbb->yDest == 0, "wrong yDest for BitBlt record: got %d, expected 0\n", emrbb->yDest ); + ok( emrbb->xSrc == 0, "wrong xSrc for BitBlt record: got %d, expected 0\n", emrbb->xSrc ); + ok( emrbb->ySrc == 0, "wrong ySrc for BitBlt record: got %d, expected 0\n", emrbb->ySrc ); + ok( emrbb->cxDest == GetSystemMetrics(SM_CXSCREEN), "wrong cxDest for BitBlt record: got %d\n", emrbb->cxDest ); + ok( emrbb->cyDest == GetSystemMetrics(SM_CYSCREEN), "wrong cyDest for BitBlt record: got %d\n", emrbb->cyDest ); + if(*pi == 1) /* First BitBlt */ + { + ok( emrbb->dwRop == SRCCOPY, "wrong ROP for BitBlt record: got %x, expected SRCCOPY\n", emrbb->dwRop ); + } + else + { + todo_wine ok( emrbb->dwRop == WHITENESS, "wrong ROP for BitBlt record: got %x, expected WHITENESS\n", emrbb->dwRop ); + } + return 1; +} + +static void test_emf_BitBlt(void) +{ + HDC hdcDisplay, hdcMetafile, hdcBitmap; + HBITMAP hBitmap, hOldBitmap; + HENHMETAFILE hMetafile; + BOOL ret; + UINT i; + + hdcDisplay = CreateDCA("DISPLAY", NULL, NULL, NULL); + ok( hdcDisplay != 0, "CreateDCA error %d\n", GetLastError() ); + hdcMetafile = CreateEnhMetaFileA(hdcDisplay, NULL, NULL, NULL); + ok( hdcMetafile != 0, "CreateEnhMetaFileA failed\n" ); + + hdcBitmap = CreateCompatibleDC(hdcDisplay); + ok( hdcBitmap != 0, "CreateCompatibleDC failed\n" ); + hBitmap = CreateCompatibleBitmap(hdcDisplay, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN)); + ok( hBitmap != 0, "CreateCompatibleBitmap failed\n" ); + hOldBitmap = SelectObject(hdcBitmap, hBitmap); + + ret = BitBlt(hdcMetafile, 0, 0, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN), hdcBitmap, 0, 0, SRCCOPY); + ok( ret, "BitBlt failed\n" ); + if(0) /* Crashes on wine */ + { + ret = BitBlt(hdcMetafile, 0, 0, GetSystemMetrics(SM_CXSCREEN), GetSystemMetrics(SM_CYSCREEN), 0, 0, 0, WHITENESS); + ok( ret, "BitBlt failed\n" ); + } + + hMetafile = CloseEnhMetaFile(hdcMetafile); + ok( hMetafile != 0, "CloseEnhMetaFile failed\n" ); + + i = 0; + EnumEnhMetaFile(0, hMetafile, test_BitBlt_record, &i, NULL); + todo_wine ok( i == 2, "too many/not enough BitBlt records: got %d, expected 2\n", i ); + + SelectObject(hdcBitmap, hOldBitmap); + DeleteObject(hBitmap); + DeleteDC(hdcBitmap); +} + static void test_mf_SaveDC(void) { HDC hdcMetafile; @@ -2628,6 +2694,7 @@ START_TEST(metafile) /* For enhanced metafiles (enhmfdrv) */ test_ExtTextOut(); test_SaveDC(); + test_emf_BitBlt(); /* For win-format metafiles (mfdrv) */ test_mf_SaveDC(); -- 1.6.5.2

1 0

Re: today's git broke winetricks gecko :-(
by James Mckenzie 18 Nov '09

18 Nov '09

Jacek Caban wrote: >Sent: Nov 17, 2009 7:40 AM >To: Tom Wickline <twickline(a)gmail.com> >Cc: James Mckenzie <jjmckenzie51(a)earthlink.net>, Vincent Povirk <madewokherd+8cd9(a)gmail.com>, wine-devel(a)winehq.org >Subject: Re: today's git broke winetricks gecko :-( > >Tom Wickline wrote: >> >> >> On Tue, Nov 17, 2009 at 10:27 PM, James Mckenzie >> <jjmckenzie51(a)earthlink.net <mailto:jjmckenzie51(a)earthlink.net>> wrote: >> >> >> > >> +1 as well. Not all UNIXes have Gecko support packages built for them >> >> >> +1 as well, I haven't tested the changes on OpenSolaris yet, but I >> plan to. :) > >Which makes -0 from both of you as you clearly don't understand the way >Gecko package works. These are .dll files so they are only arch >dependent from Wine point of view. Are there not installer packages for the different UNIXes for Gecko? If there are not, then the package should install everywhere, including MacOSX and OpenSolaris. If they are system/OS dependent, then my comment stands. James McKenzie

5 8

Re: flaky uninit variable in jscript
by Dan Kegel 17 Nov '09

17 Nov '09

There was a second new error in that file that is probably low hanging fruit: Conditional jump or move depends on uninitialised value(s) at invoke_constructor (function.c:243) by Function_value (function.c:521) by invoke_prop_func (dispex.c:259) by jsdisp_call (dispex.c:870) by disp_call (dispex.c:899) by new_expression_eval (engine.c:1568) by expr_eval (engine.c:42) by assign_expression_eval (engine.c:3110) by expr_eval (engine.c:42) by expression_statement_eval (engine.c:679) by stat_eval (engine.c:37) by exec_source (engine.c:473) by exec_global_code (jscript.c:107) by JScriptParse_ParseScriptText (jscript.c:717) by parse_script (run.c:765) by run_from_res (run.c:908) by run_tests (run.c:1057) by func_run (run.c:1095) by run_test (test.h:535) by main (test.h:585) Uninitialised value was created by a stack allocation at invoke_constructor (function.c:227) Both are still happening as of the last run I have, http://kegel.com/wine/valgrind/logs/2009-11-17-07.31/vg-jscript_run.txt - Dan On Thu, Nov 5, 2009 at 7:25 PM, Dan Kegel <dank(a)kegel.com> wrote: > Hi Jacek, > could you have a look at this? I think there are some variants that > aren't being > cleanly initialized. > > http://kegel.com/wine/valgrind/logs/2009-11-05-09.09/diff-jscript_run.txt > http://kegel.com/wine/valgrind/logs/2009-11-05-09.09/vg-jscript_run.txt > > Conditional jump or move depends on uninitialised value(s) > at VARIANT_ValidateType (variant.c:546) > by VariantClear (variant.c:599) > by exec_global_code (jscript.c:111) > by JScriptParse_ParseScriptText (jscript.c:717) > by parse_text (script.c:624) > by parse_inline_script (script.c:679) > by parse_script_elem (script.c:703) > by doc_insert_script (script.c:817) > by nsRunnable_Run (mutation.c:464) > by ??? > by ??? > by add_script_runner (mutation.c:196) > by push_mutation_queue (mutation.c:253) > by nsDocumentObserver_DoneAddingChildren (mutation.c:684) > by ??? > by ??? > by ??? > by ??? > by ??? > by ??? > Uninitialised value was created by a stack allocation > at exec_global_code (jscript.c:94) > > Thanks, > Dan >

1 0

re: Working on avifil32, sudden "non enough memory" error
by Dan Kegel 17 Nov '09

17 Nov '09

Julius wrote: >With one file however (GOODTIME.AVI from the win95 cd) I get this > error after about a second: > err:heap:HEAP_ValidateInUseArena Heap 0x110000: in-use arena 0x184420 > next block has PREV_FREE flag > non enough memory > > I have no idea where this comes from and why it happens. (It appears to > be unrelated to the memcpy call in avifile.c around line 1114.) With > +heap I also get this before the error: > HIGHPERF/GOODTIME.AVI: heap.c:403: HEAP_GetPtr: Controletest '0' faalt. > (the assert fails) Sounds like heap corruption. You might want to cut down the file to the shortest one that causes the crash, then go after it (possibly using valgrind or the like) to look for the error. Does your implementation of avifil32 and friends work properly on windows?

3 3

Re: [PATCH 2/2] [Msvcrt]: in undname functions, no longer use a fixed-size array for storing internal information
by Paul Vriens 17 Nov '09

17 Nov '09

On 11/15/2009 10:07 PM, Eric Pouech wrote: > > (fix for #20681) Hi Eric, This new test fails on all 32bit systems: http://test.winehq.org/data/tests/msvcrt:cpp.html The only box where there are no failures is a 64bit one, but no tests are actually run there ;) Could you have a look? -- Cheers, Paul.

2 1

Re: ddraw/tests: Fix test failure for D3DFMT_A2R10G10B10 pixel format in GetDC tests
by Stefan Dösinger 17 Nov '09

17 Nov '09

Am 17.11.2009 um 12:16 schrieb Austin Lund: >>> <0001-ddraw-tests-Fix-test-failure-for-D3DFMT_A2R10G10B10-.txt> >> Where did this test fail? On Wine or Windows? If it was on windows, which windows version and GPU did you use? >> > > WinXP SP2 running in Virtual box 3.0.8 with guest additions installed: > > http://test.winehq.org/data/d51b4e19fb99e30eadd6207d8dba7b0d36363c3d/xp_apl… VirtualBox uses Wine's ddraw/d3d infrastructure. It is likely that you tested the new test against an old Wine prior to when this test and fixes were committed. Your patch contradicts the behavior I saw on a Geforce 7400 Go on Windows Vista, and on a Radeon 9000 Mobility on Windows XP. As such I don't think we should enable GetDC on this format unless we find evidence on a real system that A2R10G10B10 allows GetDC calls. If we find such a system the test should be changed to accept both outcomes(ie, comment out the A2R10G10B10 test)

1 0

new crash in urlmon/misc test?
by Dan Kegel 17 Nov '09

17 Nov '09

http://kegel.com/wine/valgrind/logs/2009-11-16-19.12/diff-urlmon_misc.txt http://kegel.com/wine/valgrind/logs/2009-11-16-19.12/vg-urlmon_misc.txt show the new crash Unhandled exception: page fault on read access to 0xf0f0f0f0 in 32-bit code (0x0c530e3b). Register dump: CS:0023 SS:002b DS:002b ES:0000 FS:000b GS:0013 EIP:0c530e3b ESP:7f21fc48 EBP:7f21fc58 EFLAGS:00000084( - -- S - -P- ) EAX:f0f0f0f0 EBX:0ce3cff4 ECX:0cf416a0 EDX:00000409 ESI:7f21fe70 EDI:7bf00de0 Stack dump: 0x7f21fc48: 0cda560d 0cdfd62d 00000000 f0f0f0f0 0x7f21fc58: 7f21fc98 0cde15cd f0f0f0f0 7bf00de0 0x7f21fc68: 0cf0cebd 0cf34f5a 7f03f380 0cf41ff4 0x7f21fc78: 0cda560d 0cddff10 00000000 0cf36569 0x7f21fc88: 0cf3616c 00000409 0ce3cff4 0ce3cff4 0x7f21fc98: 7f21fcd8 0cde1b39 00000409 00000000 Backtrace: =>0 0x0c530e3b strlenW+0x15(str=0xf0f0f0f0) [include/wine/unicode.h:216] in libwine.so.1 1 0x0cde15cd CompareStringW+0xc6(lcid=1033, style=0, str1=0xf0f0f0f0, len1=-1, str2=0xcf365bc, len2=-1) [dlls/kernel32/locale.c:2737] in kernel32 2 0x0cde1b39 lstrcmpW+0x7a(str1=0xf0f0f0f0, str2=0xcf365bc) [dlls/kernel32/locale.c:2892] in kernel32 3 0x0cf0ef56 test_FindMimeFromData+0x119() [dlls/urlmon/tests/misc.c:623] in urlmon_test 4 0x0cf134b6 func_misc+0x44() [dlls/urlmon/tests/misc.c:1349] in urlmon_test - Dan

2 1

new valgrind results...
by Dan Kegel 17 Nov '09

17 Nov '09

Yesterday's valgrind results are at http://kegel.com/wine/valgrind/logs/2009-11-16-19.12/ These were done with updated suppressions; see http://code.google.com/p/winezeug/source/list?path=/trunk/valgrind/valgrind… so don't get too excited about errors going away. (I really should include a histogram of used suppressions so we can know which ones still need fixing.) I now terminate hung tests (or those that just take too long) with SIGALARM, which shows up as "Command terminated by signal 14". For instance, see the bottom of http://kegel.com/wine/valgrind/logs/2009-11-16-19.12/vg-ole32_marshal.txt With luck, my automated script will now need no babying, and results will show up 'promptly' five hours after Alexandre commits. (I used to run tests in parallel to reduce that lag; perhaps I'll try doing that again sometime, carefully.)

1 0

Re: wineserver: Preserve existing UNIX permission bits unless a matching ACCESS_ALLOWED is specified in an ACL, always preserve directory bit
by Paul Vriens 17 Nov '09

17 Nov '09

On 11/17/2009 04:48 PM, Paul Chitescu wrote: > > --- ./file.c.orig 2009-08-19 14:25:58.000000000 +0300 > +++ ./file.c 2009-11-17 17:26:31.000000000 +0200 Patches should be relative to the root. -- Cheers, Paul.

1 0