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

Wineconf?
by Damjan Jovanovic 19 Nov '09

19 Nov '09

Hi Can someone please write a bit about Wineconf 2009, for those of us that didn't go? Also is there videos/presentations to see? Thank you Damjan Jovanovic

4 7

switching to redmine?
by Dan Kegel 19 Nov '09

19 Nov '09

http://yokozar.org/blog/archives/171 is a nice writeup (thanks Scott) of wineconf. It all sounds good except for that switching to redmine part... The current wiki and bugzilla are working reasonably well, doesn't seem worth the disruption to switch. I'm highly skeptical of these newfangled wiki+bugzilla+scm+mailinglist thingies. It's really hard to do all those jobs well. Sourceforge was the first, and it was really rough. Google Code is kind of ok, but it doesn't support git yet, and its svn has performance issues (for the moment). Launchpad is also kind of ok, and kind of supports git... Yet Another WBSM Thingy that I've never heard of seems unlikely to be as robust as the above systems. - Dan

3 2

Buffer overrun in LsaLookupNames2() found with valgrind+wine+new heap patch
by Dan Kegel 19 Nov '09

19 Nov '09

Here's the first error it helped Valgrind find in tonight's automated run: ../../../tools/runtest -q -P wine -M advapi32.dll -T ../../.. -p advapi32_test.exe.so lsa.c && touch lsa.ok ... Invalid write of size 1 at memmove (mc_replace_strmem.c:613) by RtlCopySid (sec.c:376) by CopySid (security.c:905) by lookup_local_wellknown_name (security.c:2800) by lookup_name (lsa.c:308) by LsaLookupNames2 (lsa.c:411) by test_LsaLookupNames2 (lsa.c:336) by func_lsa (lsa.c:362) Address 0x7f03c550 is 6 bytes after a block of size 26 alloc'd at notify_alloc (heap.c:279) by RtlAllocateHeap (heap.c:1521) by LsaLookupNames2 (lsa.c:402) by test_LsaLookupNames2 (lsa.c:336) by func_lsa (lsa.c:362) Offhand that looks like a real problem. Aric, you may have touched that code last, could you have a look? Oh, and here's a bit that was missing from the patch. - Dan

1 0

re: Problem to download git
by Dan Kegel 18 Nov '09

18 Nov '09

I'm having network errors with git tonight, too. e.g. $ git pull fatal: The remote end hung up unexpectedly

2 1

ntdll: implement FLG_HEAP_ENABLE_TAIL_CHECK
by Dan Kegel 18 Nov '09

18 Nov '09

(Without a change like this, Valgrind can't properly detect buffer overruns, but even without Valgrind, this change will improve Wine's ability to detect buffer overruns.) With this change, users can enable buffer overrun detection in Wine by setting WINE_GLOBAL_FLAGS=0x10. This is like the overrun detection in Windows enabled by setting NtGlobalFlags to 0x10, but checks 16 bytes rather than 8, and works for large allocations, not just small ones. With this enabled, a remarkable number of tests break. I'm posting this to wine-devel for review while I add valgrind annotations (which should make it easier to tell whether the errors it detects are real or spurious) and fix the spurious breakages. Note that this change is a no-op under normal circumstances; without setting WINE_GLOBAL_FLAGS, you shouldn't see any breakage.

2 4

Resetting screen resolution on crash
by Erich Hoover 18 Nov '09

18 Nov '09

I was reading the brief summary on WineConf ( http://yokozar.org/blog/archives/171) and saw an interesting item: > We need some help from Freedesktop.org for solving bug 10841. Essentially > we need a standard way of saying “hey reset the resolution when I’m done, > even if I crash.” I was “volunteered” for the task of approaching the > respective projects. I believe this is a case that we can detect on our own, attached is a small standalone example of using fork() in such a way that you could reset the screen resolution even if an application crashed. Erich Hoover ehoover(a)mines.edu

2 1

Re: [PATCH 6/6] msxml3: Implemented xmlnode_get_nodeTypedValue date related data types handling
by Paul Vriens 18 Nov '09

18 Nov '09

On 11/18/2009 12:51 AM, Piotr Caban wrote: > --- > dlls/msxml3/node.c | 67 ++++++++++++++++++++++++++++++++++++++++ > dlls/msxml3/tests/domdoc.c | 72 > ++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 139 insertions(+), 0 deletions(-) > > Hi Piotr, These new tests introduce failures on XP and W2K3: http://test.winehq.org/data/tests/msxml3:domdoc.html (These two are the only platforms in as of this writing). Could you have a look? -- Cheers, Paul.

1 0

Re: ole32: Add a test for freeing the parent of an open stream.
by Alexandre Julliard 18 Nov '09

18 Nov '09

"Vincent Povirk" <vincent(a)codeweavers.com> writes: > From 1d583fc07db661bd0f0d0d00ea82c6208c3b744f Mon Sep 17 00:00:00 2001 > From: Vincent Povirk <vincent(a)codeweavers.com> > Date: Tue, 17 Nov 2009 11:20:34 -0600 > Subject: [PATCH 4/7] ole32: Add a test for freeing the parent of an open stream. It doesn't work here: ../../../tools/runtest -q -P wine -M ole32.dll -T ../../.. -p ole32_test.exe.so storage32.c && touch storage32.ok wine: Unhandled page fault on read access to 0x454556d6 at address 0x687dc2dd (thread 0045), starting debugger... Unhandled exception: page fault on read access to 0x454556d6 in 32-bit code (0x687dc2dd). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:687dc2dd ESP:0032f980 EBP:0032f9b8 EFLAGS:00010216( R- -- I -A-P- ) EAX:45455246 EBX:688587e4 ECX:00110058 EDX:00000100 ESI:45455246 EDI:00000011 Stack dump: 0x0032f980: 00000000 00000100 00000000 00000080 0x0032f990: 0032f9dc 0032f9b0 0032f9b8 7bc45af1 0x0032f9a0: 00000040 00040000 00141c80 7bc346cf 0x0032f9b0: 00141c58 688587e4 0032fa68 687dc339 0x0032f9c0: 45455246 00000002 0032f9dc 00000000 0x0032f9d0: 00000000 0032f9dc 00000000 00000000 Backtrace: =>0 0x687dc2dd StorageImpl_ReadRawDirEntry+0x3d(This=0x45455246, index=256, buffer="") [/home/julliard/wine/wine/dlls/ole32/storage32.c:3002] in ole32 (0x0032f9b8) 1 0x687dc339 StorageImpl_ReadDirEntry+0x39(This=0x45455246, index=2, buffer=0x32fa98) [/home/julliard/wine/wine/dlls/ole32/storage32.c:3127] in ole32 (0x0032fa68) 2 0x687d8268 StgStreamImpl_SetSize+0x98(iface=<register ESI not in topmost frame>, libNewSize=<register EDI not in topmost frame>) [/home/julliard/wine/wine/dlls/ole32/stg_stream.c:614] in ole32 (0x0032fb28) 3 0x687d86ee StgStreamImpl_Write+0x1ce(iface=0x141cd8, pv=0x6867cadc, cb=17, pcbWritten=<register EDI not in topmost frame>) [/home/julliard/wine/wine/dlls/ole32/stg_stream.c:420] in ole32 (0x0032fb78) 4 0x68662ea2 test_parent_free+0x222() [/home/julliard/wine/wine/dlls/ole32/tests/storage32.c:1322] in ole32_test (0x0032fbc8) 5 0x6866deea func_storage32+0x14ca() [/home/julliard/wine/wine/dlls/ole32/tests/storage32.c:2470] in ole32_test (0x0032fd98) 6 0x68671b40 run_test+0x130(name="storage32.c") [/home/julliard/wine/wine/dlls/ole32/tests/../../../include/wine/test.h:537] in ole32_test (0x0032fdd8) 7 0x68671d2e main+0x11e(argc=<register ECX not in topmost frame>, argv=<register ECX not in topmost frame>) [/home/julliard/wine/wine/dlls/ole32/tests/../../../include/wine/test.h:586] in ole32_test (0x0032fe88) 8 0x686727b4 __wine_spec_exe_entry+0x84(peb=0x7ffdf000) [/home/julliard/wine/wine/dlls/winecrt0/exe_entry.c:36] in ole32_test (0x0032feb8) 9 0x7b8768a5 start_process+0x55(peb=<register ESI not in topmost frame>) [/home/julliard/wine/wine/dlls/kernel32/process.c:1025] in kernel32 (0x0032fee8) 10 0x7bc6de64 call_thread_func+0xc() in ntdll (0x0032fef8) 11 0x7bc6e080 call_thread_entry_point+0x70(entry=0x7b876850, arg=0x7ffdf000) [/home/julliard/wine/wine/dlls/ntdll/signal_i386.c:2342] in ntdll (0x0032ffc8) 12 0x7bc4b20a in ntdll (+0x3b20a) (0x0032ffe8) 13 0x68027e4d wine_call_on_stack+0x1d() in libwine.so.1 (0x00000000) 0x687dc2dd StorageImpl_ReadRawDirEntry+0x3d [/home/julliard/wine/wine/dlls/ole32/storage32.c:3002] in ole32: movl 0x490(%eax),%eax 3002 hr = BlockChainStream_ReadAt( -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: To error out or to skip tests?
by Greg Geldorp 18 Nov '09

18 Nov '09

Hi Jörg > > The picture on test.winehq.org is a disaster: None of the w95, w98, nt4, > > 2k, 2k3, Vista, 2k8, w7 machines have performed winmm:wave tests! > > They all have no sound configured. Mostly useless. > > A lot of those VMware boxes are mine. They come in two flavors. Some of > them (starting with esx-) are ESX VMs, ESX is VMware's data center product, > which simply doesn't support sound. The others (starting with gvg-) are > Server VMs, which do support sound. The reason I didn't configure sound > on them is because I'm simply not interested in sound. I'll consider > enabling sound, but to be honest I'm a bit put off by your aggressive tone. > > Anyway, the end result is that there will always be ESX VMs that can't > run sound tests. After a bit of mucking around I got sound to work on the Linux host where those VMware Server VMs run, and enabled sound in the VMs where possible. I could not find correct sound drivers for some of the old stuff (NT4 and below), so those still don't have sound, neither do the ESX VMs. Best regards, Ge.

3 2

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

18 Nov '09

Austin Lund <austin.lund(a)gmail.com> writes: > @@ -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; > + } Testing file system flags is not a good idea, that won't work on Wine. It's better to test the results of the actual call. -- Alexandre Julliard julliard(a)winehq.org

2 2