September 2011 - Wine-devel

Re: loop fails
by Dan Kegel 24 Sep '11

24 Sep '11

2011/9/24 André Hentschel <nerv(a)dawncrow.de>: > Hey Dan, > sadly i'm not going to be at wineconf this year, but i thought the results might help anyway. maybe i'm participating per IRC and it's my desktop computer anyway, so we'll see. > the script didn't end cleanly, there was a heap related crash that made it hang so i'm not sure about it. > > ended in the urlmon test: > > Backtrace: > =>0 0x7bc48485 RtlAllocateHeap+0xb5(heap=0x110000, flags=0x2, size=0x6000) [/home/dawncrow/dev/wine/132/dlls/ntdll/../../../include/wine/list.h:99] in ntdll (0x0032e328) > 1 0x696866b7 MSVCRT_malloc+0x36(size=0x6000) [/home/dawncrow/dev/wine/132/dlls/msvcrt/../../../dlls/msvcrt/heap.c:312] in msvcrt (0x0032e348) > 2 0x702d1962 in mozjs (+0x151961) (0x02326318) > 3 0x00000000 (0x010c6740) oooh. So it only took six loops to crash in urlmon, eh? (For the record, the script ran both dlls/urlmon/tests/protocol.ok and dlls/urlmon/tests/url.ok, don't know which one crashed.) Can you install debug gecko, e.g. by running http://kegel.com/wine/dgecko.sh.txt ( which just does the steps on http://wiki.winehq.org/Gecko ) and re-run? Also, out of six loops, you got the following test failures: 6 dlls/dsound/tests/ds3d.ok 6 dlls/dsound/tests/ds3d8.ok 6 dlls/wininet/tests/urlcache.ok 6 dlls/dsound/tests/dsound.ok 6 dlls/dsound/tests/dsound8.ok 5 dlls/winhttp/tests/winhttp.ok 5 dlls/vbscript/tests/run.ok 5 dlls/mshtml/tests/style.ok 5 dlls/mshtml/tests/htmldoc.ok 5 dlls/gdi32/tests/bitmap.ok 2 dlls/ws2_32/tests/sock.ok 1 dlls/user32/tests/msg.ok The urlcache one may go away if you clear out ~/.wine. The msg one probably would go away if you had the blessed window manager (does anyone know what the procedure there is these days?) The others seem interesting. If you were attending wineconf, I would have asked you to hit the appropriate developers up for help resolving them. Maybe you still can via email or irc... I'm looking forward to seeing loop-fail results from other folks. - Dan

2 1

Re: hhctrl.ocx: Don't show unimplemented buttons.
by Erich Hoover 24 Sep '11

24 Sep '11

> + case IDTB_NOTES: > + case IDTB_CONTENTS: > + case IDTB_INDEX: > + case IDTB_SEARCH: > + case IDTB_HISTORY: > + case IDTB_FAVORITES: > + /* These are officially unimplemented as of the Windows 7 SDK */ > + break; Could you point me to a link on this behavior? I'm pretty sure that many of these buttons send a signal that applications can hook into and override the default (do-nothing) behavior. Erich Hoover ehoover(a)mines.edu

3 2

Re: [1/4] ddraw/tests: Use sizeof instead of a magic number
by Stefan Dösinger 24 Sep '11

24 Sep '11

Am 24.09.2011 um 12:48 schrieb Stefan Dösinger: > <PGP.txt> WTF? What did my Mail client do here???

1 0

Congratulations! DIB engine 421
by Keith 23 Sep '11

23 Sep '11

Hi, Amazing I'm a long time follower of the project and to see a DIB engine implemented is a huge milestone given all the trials and tribulations leading up to it. Honestly never thought it would happen everyone needs to give themselves a pat on the back today :-) Regards, Keith

1 0

Re: dpnet: fill the DPN_SP_CAPS structure in IDirectPlay8Peer_GetSPCaps with some sane values + tests
by Marvin 23 Sep '11

23 Sep '11

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=14486 Your paranoid android. === WNT4WSSP6 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BF === W2KPROSP4 (32 bit peer) === peer.c:52: Test failed: CoCreateInstance failed with 0x80040154 peer: unhandled exception c0000005 at 004013BF

1 0

Re: [PATCH] comdlg32: Initialize 2 RECT to 0 (Coverity)
by Andrew Talbot 23 Sep '11

23 Sep '11

Marcus Meissner wrote: > Hi, > > If the filename and toolbar field are not present, we will be using > uninitialized RECTs, so initialize them. > > CID 5033, 5034 > > Ciao, Marcus > --- > dlls/comdlg32/itemdlg.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/dlls/comdlg32/itemdlg.c b/dlls/comdlg32/itemdlg.c > index bc309c9..d9756c0 100644 > --- a/dlls/comdlg32/itemdlg.c > +++ b/dlls/comdlg32/itemdlg.c > @@ -1106,6 +1106,8 @@ static SIZE update_layout(FileDialogImpl *This) > SIZE ret; > > GetClientRect(This->dlg_hwnd, &dialog_rc); > + memset (&toolbar_rc, 0, sizeof(toolbar_rc)); > + memset (&filename_rc, 0, sizeof(filename_rc)); > > missing_width = max(0, 320 - dialog_rc.right); > missing_height = max(0, 200 - dialog_rc.bottom); Hi Marcus, Instead of using memset(), would it not be better to initialise the RECTs to zero when declaring them, as follows? RECT toolbar_rc = { 0 }; RECT filename_rc = { 0 }; -- Andy.

1 0

Re: [PATCH] comdlg32: Initialize 2 RECT to 0 (Coverity)
by Alexandre Julliard 23 Sep '11

23 Sep '11

Marcus Meissner <meissner(a)suse.de> writes: > Hi, > > If the filename and toolbar field are not present, we will be using > uninitialized RECTs, so initialize them. It's not clear that 0 is much better. If there's really a possibility for these things to be missing the layout should take that into account. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: dpnet: return a TCP/IP provider in IDirectPlay8PeerImpl_EnumServiceProviders (try 3, resend)
by Bruno Jesus 23 Sep '11

23 Sep '11

On Fri, Sep 23, 2011 at 07:50, Louis Lenders <xerox_xerox2000(a)yahoo.co.uk> wrote: > Note: the 2 failing tests on test-bot that this patch will produce, are not real failures as winnt and win2000 will not run the test because of missing dpnet > > try 3: rewrote patch in a more straightforward way > > this makes previously sent tests pass, and fixes crash in bug http://bugs.winehq.org/show_bug.cgi?id=28008 Use the broken macro to skip or expect a different return from the tests that crash in NT and add a comment there. Example: http://source.winehq.org/source/dlls/ole32/tests/usrmarshal.c#L103 Best wishes, Bruno -- universe* god::bigbang (void); //and then it all began...

1 0

Re: dpnet: return a TCP/IP provider in IDirectPlay8PeerImpl_EnumServiceProviders (try 3, resend)
by Marvin 23 Sep '11

23 Sep '11

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=14466 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

1 0

Re: mshtml: Add IHTMLCurrentStyle2/3/4 support
by Alexandre Julliard 23 Sep '11

23 Sep '11

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > diff --git a/dlls/mshtml/Makefile.in b/dlls/mshtml/Makefile.in > index 3933645..62a2d6f 100644 > --- a/dlls/mshtml/Makefile.in > +++ b/dlls/mshtml/Makefile.in > @@ -14,6 +14,7 @@ C_SRCS = \ > htmlbody.c \ > htmlcomment.c \ > htmlcurstyle.c \ > + htmlcurstyle2.c \ Please add this to the existing file. There are already way too many source files in this dll. -- Alexandre Julliard julliard(a)winehq.org

1 0