December 2005 - Wine-Devel - list.winehq.org

Re: msvcrt: Printf Integer Conversion try 2

by Alexandre Julliard

Jesse Allen <the3dfxdude(a)gmail.com> writes: >> Changes: >> Add pf_is_integer_format to detect integer formats >> Add pf_integer_conv for I64 types >> Foward I32 and I to libc's printf >> Correctly handle long long >> Update pf_fill to handle the sign for signed integers >> Fix pf_fill to ignore flags->PadZero for printing zeroes when >> flags->LeftAlign is set >> Fix existing pf_is_*_format to require explicit upper and lower case formats >> Still tolower( flags.Format ) before we use libc's printf (to avoid >> regressions?) >> Allow flags.Sign to be ' ' (space) for the sign place holder, also >> make sure '+' overrides it. > > Any comments now? You should split the patch, you are trying to fix too many things at once. Also it still fails some tests: printf.c:45: Test succeeded inside todo block: Problem with long long printf.c:46: Test succeeded inside todo block: return count wrong printf.c:280: Test failed: failed printf.c:281: Test failed: return count wrong printf.c:465: Test succeeded inside todo block: Problem with long long -- Alexandre Julliard julliard(a)winehq.org

Dec. 19, 2005

Re: Page fault: completely lost while debugging

by Vitaliy Margolen

Sunday, December 18, 2005, 5:54:17 PM, Sergio wrote: > Hi all, > I am having one problem with one project I have just made compile. It's one > of the plugins of the windows' multiprotocol IM program, Miranda. > I will try to explain what my problem is. > After solving some issues with compilation (mainly some undefined names - > functions - in std namespace - it's a C++ project), I tried to run the > application and when asking the plugin to connect to one IRC server (it is > the application's IRC protocol plugin) it crashes with this error: > Unhandled exception: page fault on read access to 0x00000004 in 32-bit code > (0x41d551de). > In 32 bit mode. > fixme:dbghelp:elf_load_debug_info_from_map Alpha-support for Dwarf2 > information for irc<elf> > Register dump: > CS:0073 SS:007b DS:007b ES:007b FS:174f GS:0033 > EIP:41d551de ESP:43510adc EBP:43510af8 EFLAGS:00210206( - 00 - RIP1) > EAX:00000004 EBX:41db6434 ECX:00000000 EDX:7c0cbe30 > ESI:00000000 EDI:41c14f10 > Stack dump: > 0x43510adc: 41d5514e 01000000 4f4d2e6c 41db6434 > 0x43510aec: 7c126a38 4017c0d4 41db6434 43510b28 > 0x43510afc: 41d9b4b5 7c126a38 00000000 00000000 > 0x43510b0c: 41d9b45c 401f6b31 407bdbd0 4078f8e0 > 0x43510b1c: 41d55142 7c126a38 407d4528 43510b40 > 0x43510b2c: 407c9e31 4078f8e0 41d9b450 4078f8e0 > 02e9: sel=174f base=41c14000 limit=00001fff 32-bit rw- > Backtrace: > =>1 0x41d551de in irc (+0x51de) (0x41d551de) > 2 0x41d9b4b5 in irc (+0x4b4b5) (0x41d9b4b5) > 3 0x407c9e31 in msvcrt (+0x29e31) (0x407c9e31) > 4 0x404ebbfb in kernel32 (+0x6bbfb) (0x404ebbfb) > 5 0x401d6cd3 in ntdll (+0x36cd3) (0x401d6cd3) > 6 0x40051297 start_thread+0x87 in libpthread.so.0 (0x40051297) > 7 0x4011c37e __clone+0x5e in libc.so.6 (0x4011c37e) > 0x41d551de: movl 0x0(%eax),%eax > > Which if I understand the dump well is due to a problem in __clone function > trying to access 0x00000004 (eax). Am I right? Nope you looking at the wrong end. The address at fault is 0x41d551de which is frame 1. Also it looks like you have stripped wine libs. That's why you don't see function names nor their parameters. > 005d:Call ntdll.RtlLeaveCriticalSection(41164de8) ret=4111a044 thread ID function name called arguments return address > 005d:Ret ntdll.RtlLeaveCriticalSection() retval=00000000 ret=4111a044 thread ID function name returning from return value return address > All this is absolutely cryptic to me. Does anyone ge something from this > information? Is there any way to get to know where the aplication is > crashing? What you might need is a WINEDEBUG=+relay,+seh debug channels. That should point you at the place where exception occurred. PS You might want to look at developer's guide on winehq.org. Vitaliy.

Dec. 19, 2005

Re: msvcrt: Printf Integer Conversion try 2

by Jesse Allen

On 12/16/05, Jesse Allen <the3dfxdude(a)gmail.com> wrote: > Hi Alexandre, > > This patch is much improved. It passes all tests this time including > additional ones you will see in the new tests patch. > > Changes: > Add pf_is_integer_format to detect integer formats > Add pf_integer_conv for I64 types > Foward I32 and I to libc's printf > Correctly handle long long > Update pf_fill to handle the sign for signed integers > Fix pf_fill to ignore flags->PadZero for printing zeroes when > flags->LeftAlign is set > Fix existing pf_is_*_format to require explicit upper and lower case formats > Still tolower( flags.Format ) before we use libc's printf (to avoid > regressions?) > Allow flags.Sign to be ' ' (space) for the sign place holder, also > make sure '+' overrides it. > > > Any comments now?

Dec. 19, 2005

Re: msvcrt: Printf Integer Conversion try 2

by Jesse Allen

On 12/16/05, Jesse Allen <the3dfxdude(a)gmail.com> wrote: > Hi Alexandre, > > This patch is much improved. It passes all tests this time including > additional ones you will see in the new tests patch. > > Changes: > Add pf_is_integer_format to detect integer formats > Add pf_integer_conv for I64 types > Foward I32 and I to libc's printf > Correctly handle long long > Update pf_fill to handle the sign for signed integers > Fix pf_fill to ignore flags->PadZero for printing zeroes when > flags->LeftAlign is set > Fix existing pf_is_*_format to require explicit upper and lower case formats > Still tolower( flags.Format ) before we use libc's printf (to avoid > regressions?) > Allow flags.Sign to be ' ' (space) for the sign place holder, also > make sure '+' overrides it. > > >

Dec. 19, 2005

crashing tests in d3d9

by Louis. Lenders

Hi, the tests in d3d9 all crash for me : wine d3d9_test.exe.so stateblock fixme:d3d:IWineD3DDeviceImpl_GetAvailableTextureMem (0x7fec77c0) : stub, emulating 64Mib for now, returning 64Mib wine: Unhandled page fault on read access to 0x00000000 at address 0xb7eb77a0 (thread 0009), starting debugger... WineDbg starting on pid 0x8 Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0xb7eb77a0). In 32 bit mode. Register dump: CS:0073 SS:007b DS:007b ES:007b FS:003b GS:0033 EIP:b7eb77a0 ESP:7fc5fb58 EBP:7fc5fb5c EFLAGS:00010206( - 00 - RIP1) EAX:7fec7418 EBX:7ee0e18c ECX:7fec7417 EDX:00000000 ESI:7fec7418 EDI:7fec73f0 Stack dump: 0x7fc5fb58: 7fec77c0 7fc5fda0 7edc3794 7fec7418 0x7fc5fb68: 00000000 00001f01 7ee0e8dc 7edf8740 0x7fc5fb78: 00000000 00000000 0000ffff 000001b4 0x7fc5fb88: 00000008 00000001 00000000 7c04dc58 0x7fc5fb98: 00000000 00000000 7fec7418 00000300 0x7fc5fba8: 7ee0ee60 7fc5fbc4 7fc5fbe4 0000a838 Backtrace: =>1 0xb7eb77a0 (0xb7eb77a0) 2 0x7edc3794 IWineD3DImpl_FillGLCaps+0x68 [/LFS/winecvs/wine/dlls/wined3d/directx.c:234] in wined3d (0x7edc3794) 3 0x7edc799e IWineD3DImpl_CreateDevice(iface=0x7fec73f0, Adapter=0x0, DeviceType=0x1, hFocusWindow=0x0, BehaviourFlags=0x20, pPresentationParameters=0x7fc5fe14, ppReturnedDeviceInterface=0x7fec71c8, parent=0x7fec71c0, D3DCB_CreateAdditionalSwapChain=0x7ef8a014) [/LFS/winecvs/wine/dlls/wined3d/directx.c:1808] in wined3d (0x7edc799e) 4 0x7ef8a2f2 IDirect3D9Impl_CreateDevice+0xde(iface=0x7fec7320, Adapter=0x0, DeviceType=0x1, hFocusWindow=0x0, BehaviourFlags=0x20, pPresentationParameters=0x7fc5fe98, ppReturnedDeviceInterface=0x7fc5fe94) [/LFS/winecvs/wine/dlls/d3d9/directx.c:312] in d3d9 (0x7ef8a2f2) 5 0x7ffb7648 func_stateblock [/LFS/winecvs/wine/dlls/d3d9/tests/stateblock.c:46] in d3d9_test (0x7ffb7648) 6 0x7ffb7eb4 run_test [/LFS/winecvs/wine/dlls/d3d9/tests/../../../include/wine/test.h:355] in d3d9_test (0x7ffb7eb4) 7 0x7ffb8149 __wine_spec_exe_entry(peb=0x7bc6f680) [/LFS/winecvs/wine/dlls/winecrt0/exe_entry.c:36] in d3d9_test (0x7ffb8149) 8 0x7b862855 start_process+0xc1(arg=0x0) [/LFS/winecvs/wine/dlls/kernel/process.c:1027] in kernel32 (0x7b862855) 9 0xb7f8f57b wine_switch_to_stack+0x17 in libwine.so.1 (0xb7f8f57b) 0xb7eb77a0: movzbl 0x0(%edx),%eax etc. Anone an idea why they crash? ___________________________________________________________ To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com

Dec. 19, 2005

Re: Create new mailing list wine-isv?

by Robert Shearman

Peter Beutner wrote: > Dmitry Timoshkov schrieb: > >> "Peter Beutner" <p.beutner(a)gmx.net> wrote: > >>>> Wine is a very good way of testing the waters with a Linux market. >>>> If a significant part of the market share starts coming from Linux >>>> or other Unix operating systems then the company can start offering >>>> winelib extensions that integrate better with the environment in >>>> which they are running. >>> >>> >>> I doubt that this will happen. If the windows version works with >>> wine the company will more likely continue to work on that. See your >>> money argument. >> >> >> Another myth about Wine. > > Glad to hear that. So there are already companies shipping winelib > extensions for their products? I know of at least one company that was considering this for interfacing their software with a USB hardware device that the software shipped with. -- Rob Shearman

Dec. 19, 2005

binary only dll , resolving links through winebuild

by Ananth Rao

This can be resolved by - create the .def file using the spec file - Link the application to this dll whicle building the application executable ######################################################### THIS EMAIL MESSAGE IS FOR THE SOLE USE OF THE INTENDED RECIPIENT(S) AND MAY CONTAIN CONFIDENTIAL AND PRIVILEGED INFORMATION. ANY UNAUTHORIZED REVIEW, USE, DISCLOSURE OR DISTRIBUTION IS PROHIBITED.BEFORE OPENING ANY ATTACHMENTS PLEASE CHECK FOR VIRUSES AND DEFECTS.IF YOU ARE NOT THE INTENDED RECIPIENT, PLEASE NOTIFY US IMMEDIATELY BY REPLY E-MAIL AND DELETE THE ORIGINAL MESSAGE. ##########################################################

Dec. 19, 2005

Re: Alexandre Julliard : Take advantage of the __EXCEPT_PAGE_FAULT macro.

by Robert Shearman

Alexandre Julliard wrote: >Robert Shearman <rob(a)codeweavers.com> writes: > > > >>The idea is that the ole_filter will catch all exceptions other than >>page faults so that we don't hide obvious programmer errors. However, >>there obviously should be cases for some other exceptions that we don't >>want to handle. >> >> > >That doesn't sound right, we should have cases for exceptions that we >want to handle, not catching all of them. What sort of exceptions do >you expect to get here? > E_NOINTERFACE, RPC_E_DISCONNECTED, RPC_X_NULL_REF_POINTER, etc. I believe native also catches all but a few exceptions here, but I'll test this hypothesis. In the mean time, can you revert the patch? I think that it will have broken anything that uses MIDL generated proxies. -- Rob Shearman

Dec. 19, 2005

Re: Steam Winsock regression

by Andreas Mohr

Hi, On Sun, Dec 18, 2005 at 11:51:01PM -0800, James Liggett wrote: > On Sun, 2005-12-18 at 23:50 -0800, Scott Ritchie wrote: > > On Sun, 2005-12-18 at 22:35 -0800, James Liggett wrote: > > > Hi, > > > For a while now I've been noticing a very bad regression in winsock with > > > steam. If I try to view information about a server using the "View game > > > info command," my Internet connection fails completely. Not only that, > > > but sometimes *all* of the machines on my network (both WinXP and Linux) > > > lose their connections as well. I did the usual drudgery and found the > > > offending patch: > > > http://www.winehq.com/pipermail/wine-cvs/2005-November/019293.html > > > > > > Can someone take a look at this one ASAP? Thanks very much! > > > > > > James > > > > Does that make this a security issue? > Maybe for people running Wines older than current CVS (this would > include 0.9.3, it has been fixed in current) Nonetheless, it's a pretty > bad problem, but the next release will include the fix. I don't think Scott meant it that way ;-) Who cares about Wine, here's a change that makes half the machines in the network lose connection randomly, that's much bigger things to worry about potentially than simply non-working Wine socket functionality... Thus one should probably attempt to investigate a bit more. Andreas Mohr

Dec. 19, 2005

Re: ntdll: Implement few Dbg* functions.

by Eric Pouech

Eric Pouech wrote: > IMO, the Ui ones seem to be rather old (NT4 or something) whereas the > ones I'm talking about are more recent (XP), but seem better defined > (API wise). For example, only XP version seems to provide the wait for > debug object and continue APIs. That's why it seems to me more > interesting to me that we move to the XP flavor. actually, you could implement it on top of Ui flavor as well we'll have to dig deeper to tell the differences A+ -- Eric Pouech

Dec. 19, 2005